EPA/600/6-90/003
January 1990
METHODOLOGY FOR ASSESSING HEALTH
RISKS ASSOCIATED WITH INDIRECT
EXPOSURE TO COMBUSTOR EMISSIONS
INTERIM FINAL
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
S
-------
DISCLAIMER
This document has been reviewed in accordance with U.S.
Environmental Protection Agency policy and approved for
publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
11
-------
PREFACE
Incineration is a widely-used method for the treatment" "or r
disposal of waste materials such as municipal solid waste/ «
hazardous waste, and sewage sludge. The ability to make careful/:'
scientific estimates of potential risks to human health or the
environment from atmospheric emissions is critical to the planning,
design and siting of waste combustion facilities.
This methodology document seeks to provide risk assessors with
the guidance necessary to estimate the health risks that result
from exposure to toxic pollutants in combustor emissions by
pathways other than inhalation. Whereas procedures for assessing
human health risks from inhalation of pollutant emissions are well
established, this methodology enables estimation of the indirect
human exposures and health risks that can result from the transfer
of emitted pollutants to soil, vegetation and water bodies.
This methodology is not intended to be prescriptive; that is,
it does not comprise a set of guidelines or recommended approaches
that the U.S. EPA believes should be applied in all circumstances.
Rather, it provides a set of procedures that the risk assessor can
draw upon where applicable for a specific assessment. The
appropriate use of these procedures and the discussion of
uncertainties surrounding the results remain important
responsibilities of the risk assessor. This document is a
revision of an earlier report that underwent review by U.S. EPA's
Science Advisory Board and also served'as a basis for much of the
information presented in the document rAssessment of Health Risks
Associated with Municipal Waste Combustion Emissions (U.S. EPA,
1987c). Revisions were carried out by Syracuse Research
Corporation, under U.S. EPA Contract No. 68-C8-0004, and Oak Ridge
National Laboratories' Office of Risk Analysis, under Interagency
Agreement DW89932633-01.
Measures planned in the near future to further improve the
utility of this methodology include (1) the availability of
computer software for the execution of calculations in this
methodology and (2) demonstrations of the use of the methodology
to test its applicability to specific sites and problems. For more
information, please contact Randall Bruins, Environmental Criteria
and Assessment Office, U.S. EPA, Cincinnati, OH 45268.
iii
-------
TABLE OF CONTENTS
Page
1. INTRODUCTION 1-1
1.1. PURPOSE 1-1
1.2. BACKGROUND 1-2
1.3. SCOPE 1-5
2. HUMAN EXPOSURE SCENARIOS 2-1
2.1. INTRODUCTION 2-1
2.2. CREATING EXPOSURE SCENARIOS 2-2
2.3. DEFINING THE LENGTH OF TIME OF
EMISSIONS 2-6
2.4. DEFINING THE EXPOSED INDIVIDUAL 2-7
2.4.1. Location of the Individual
Relative to the Combustion
Source 2-7
2.4.2. Age of Exposed Individual 2-11
2.4.3. Human Body Weight (BW) 2-12
2.4.4. Length of Time Individual Spends
at Location 2-14
2.5. PATHWAYS OF HUMAN EXPOSURE 2-17
2.5.1. Food Ingestion Pathways 2-17
2.5.2. Soil Exposure Pathways 2-20
2.5.3. Water Exposure Pathways 2-23
2.6. DEFINING EXAMPLE EXPOSURE SCENARIOS 2-27
2.6.1. Exposure Scenario A 2-28
2.6.2. Exposure Scenario B 2-31
2.6.3. Exposure Scenario C 2-33
3. AIR DISPERSION AND DEPOSITION OF EMITTED
POLLUTANTS 3-1
3.1. BACKGROUND AND PURPOSE 3-1
3.2. MODEL SELECTION 3-2
3.3. MODEL DESCRIPTIONS 3-5
IV
-------
TABLE OF CONTENTS (cont.)
3.3.1. General Model Development
3.3.2. COMPDEP Mode]
3.3.3. RTDMDEP
3.3.4. ISCST
3.4. ANALYSIS PROCEDURES
3.4.1. Source Characteristics
3.4.2. Meteorologic Data
3.4.3. Receptor Locations ,
3.4.4. Model Run Descriptions ,
3.4.5. Reflection Coefficients Used
by ISCST
3.5. COMPARISON OF MODEL RESULTS ,
3.5.1. Calculated Air Concentrations
Neglecting Deposition Effects..
3.5.2. Calculated Air Concentrations
Including Deposition Effects...
3.5.3. Calculated Values of Pollutant
Deposition
3.5.4. Summary of Air Dispersion
Modeling
3.6. DETERMINATION OF AREAL-AVERAGE
DEPOSITION
4. CALCULATING SOIL CONCENTRATIONS
4.1. INTRODUCTION. . . ;
4.2. OVERVIEW OF SOIL CONCENTRATION
CALCULATIONS
4.2.1. Calculating Cumulative Soil
Concentration
4.2.2. Calculating the Soil Loss
Constant „
4.3. DESCRIPTION OF INPUT VARIABLES
4.3.1. Dry and Wet Deposition Rates
(Dyd and Dyw)
4.3.2. Time Period of Deposition (Tc).
4.3.3. Soil Depth (Z)
4.3.4. Bulk Density of the Soil (BD)..
4.3.5. Soil Loss Constant (ks)
Page
3-5
3-5
3-11
3-13
3-14
3-15
3-15
3-17
3-18
3-18
3-25
3-25
3-33
3-37
3-41
3-44
4-1
4-1
4-3
4-3
4-5
4-7
4-8
4-8
4-8
4-11
4-11
-------
TABLE OF CONTENTS (cont.)
Page
4.4. EXAMPLE CALCULATIONS 4-18
4.4.1. Cadmium (Cd) 4-19
4.4.2. Benzo(a)Pyrene (B(a)P) 4-23
5. DETERMINING EXPOSURE THROUGH THE TERRESTRIAL
FOOD CHAIN 5-1
5.1. INTRODUCTION 5-1
5.2. CALCULATING CONCENTRATION OF POLLUTANT
IN PLANTS 5-1
5.2.1. Plant Pollutant Concentration
Due to Root Uptake 5-3
5.2.2. Plant Pollutant Concentration
Due to Direct Deposition. 5-8
5.2.3. Plant Pollutant Concentration
Due to Air-to-Plant Transfer.... 5-16
5.2.4. Phytotoxicity 5-23
5.3. CALCULATING CONCENTRATION OF POLLUTANT IN
ANIMAL TISSUES 5-25
5.3.1. Quantity of Plants Consumed by
Animals (Qp,-j) 5-26
5.3.2. Quantity of Soil Consumed by.
Animals (QSj) 5-27
5.3.3. Biotransfer Factors (Baj) 5-29
5.4. CALCULATING HUMAN DAILY INTAKE 5-31
5.4.1. Calculating Daily Intake from
Contaminated Plants 5-31
5.4.2. Calculating Daily Intake from
Contaminated Animal Tissues 5-41
5.5. EXAMPLE CALCULATIONS 5-44
5.5.1. Benzo(a)pyrene 5-46
5.5.2. Cadmium 5-61
5.5.3. Summary 5-76
6. DETERMINING EXPOSURE FROM SOIL INGESTION 6-1
6.1. INTRODUCTION 6-1
6.2 HUMAN DAILY INTAKE (DI) 6-1
6.2.1. Soil Concentration (Sc) 6-2
6.2.2. Soil Ingestion Rate (Cs) 6-4
6.2.3. Body Weight (BW) 6-6
vi
-------
TABLE OF CONTENTS (cont.)
6.3. EXAMPLE CALCULATIONS
6.3.1. Benzo(a)Pyrene
6.3.2. Cadmium
7. DETERMINING EXPOSURE FROM DERMAL ABSORPTION
VIA SOIL
7.1. INTRODUCTION
7.2. DAILY DERMAL INTAKE (DDI)
7.2.1. Contact Time (CT)
7.2.2. Surface Area (SA)
7.2.3. Contact Amount (CA)
7.2.4. Absorption Fraction (AF).......
7.2.5. Soil Concentration (So)
7.2.6. Body Weight (BW)
7.3. EXAMPLE CALCULATIONS
7.3.1. Benzo (a) Pyrene
7.3.2. Cadmium
8 . DUST RESUSPENSION
8.1. INTRODUCTION
8.2. SOIL EROSION PROCESS
8.3. FACTORS AFFECTING WIND EROSION.
8.4. EXAMPLE CALCULATIONS
8.4.1. Estimation of Soil
Concentration
8.4.2. Annual Average Emission Rate...
8.4.3. Worst-Case 24-hr Emission Rate.
8.4.4. Results
8.5. CONCLUSION
9. CALCULATING WATER CONCENTRATIONS
9.1. INTRODUCTION
9.2. SURFACE WATER
Page
6-*6
6-6
6-8
7-1
7-1
7-1
7-2
7-6
7-7
7-8
7-10
7-10
7-10
7-11
7-13
8-1
8-1
8-1
8-2
8-3
8-5
8-7
8-14
8-17
8-19
9-1
9-1
9-1
vn
-------
TABLE OF CONTENTS (cont.)
Page
9.2.1. Overview of the Model 9-2
9.2.2. Calculating Contaminant
Concentrations for Tier 1 9-4
9.2.3. Calculating Contaminant
Concentrations for Tier 2 or 3.. 9-8
9.3. PRECIPITATION 9-19
9.3.1. Water Concentration (We) 9-19
9.4. GROUNDWATER 9-20
9.4.1. Overview of the Model 9-21
9.4.2. Determining Leachate Contaminant
Concentration 9-24
9.4.3. Conclusion 9-27
9.5. EXAMPLE CALCULATIONS 9-28
9.5.1. Benzo(a)Pyrene 9-28
9.5.2. Cadmium 9-51
10. DETERMINING EXPOSURE FROM WATER INGESTION 10-1
10.1. INTRODUCTION 10-1
10.2. DAILY INTAKE FROM WATER (DI) 10-1
10.2.1. Water Concentration (We) 10-2
10.2.2. Water Consumption Rate (Cw)... 10-2
10.3. EXAMPLE CALCULATIONS 10-3
11. DETERMINING EXPOSURE FROM FISH INTAKE 11-1
11.1. INTRODUCTION 11-1
11.2. CALCULATING DAILY INTAKE FROM FISH 11-1
11.2.1. Water Concentration (We) 11-2
11.2.2. Bioconcentration Factor (BCF). 11-2
11.2.3. Fish Ingestion Rate (Cf) 11-6
11.3. EXAMPLE CALCULATIONS 11-9
11.3.1. Benzo(a)Pyrene 11-9
11.3.2. Cadmium 11-11
Vlll
-------
TABLE OF CONTENTS (cont.)
Page
12. DETERMINING EXPOSURE FROM DERMAL ABSORPTION
FROM WATER 12-1
12 .1. INTRODUCTION . 12-1
12 . 2 . DAILY DERMAL INTAKE FROM WATER 12-1
12.2.1. Contact Time and Frequency.... 12-2
12.2.2. Surface Area (SA) 12-3
13 . HAZARD IDENTIFICATION 13-1
13 .1. INTRODUCTION 13-1
13.2. ELEMENTS OF HAZARD IDENTIFICATION... 13-2
13.2.1. Human Studies/Epidemiologic
Studies. 13-3
13.2.2. Animal Studies 13-4
13.2.3. Supportive Evidence 13-4
13 . 3 . WEIGHT OF EVIDENCE. 13-7
13.3.1. EPA Classification of
Benzo(a)Pyrene and Cadmium.... 13-10
13 .4. SUMMARY AND CONCLUSION. 13-12
14 . DOSE-RESPONSE ASSESSMENT. „ 14-1
14.1. INTRODUCTION 14-1
14.2. DOSE-RESPONSE ASSESSMENT FOR
SYSTEMIC TOXICITY. 14-1
14.2.1. Selection of the Critical
Data 14-2
14.2.2. Reference Dose 14-4
14.2.3. Selection of Uncertainty and
Modifying Factors 14-5
14.2.4. Confidence in the RfD 14-6
14.2.5. Calculation of RfD for
Cadmium. . 14-7
14.3. DOSE-RESPONSE ASSESSMENT FOR
CARCINOGENS 14-8
ix
-------
TABLE OF CONTENTS (cont.)
Page
14.3.1. Selection of Data Sets 14-8
14.3.2. Choice of Extrapolation
Model 14-9
14.3.3. Route-to-Route Extrapolation.. 14-10
14.3.4. Confidence in Quantitation.... 14-10
14.3.5. Calculation of q.,* for
Benzo(a)Pyrene 14-11
14.4. SUMMARY 14-11
15. RISK CHARACTERIZATION 15-1
15.1. INTRODUCTION 15-1
15.2. ADJUSTMENTS TO EXPOSURE 15-2
15.2.1. Relative Effectiveness of
Exposure (RE) 15-3
15.2.2. Exposure Duration Adjustment
(EDA) 15-6
15.2.3. Estimating Total Exposure by
Each Route 15-8
15.2.4. Total Background Intake of
Pollutant by Each Route (TBI). 15-12
15.3. RISK ESTIMATION 15-15
15.3.1. Comparison with the RfD 15-16
15.3.2. Determination of the Excess
Risk (ER) for Non-Threshold
Toxicants (Carcinogens) 15-21
15.4. CHARACTERIZATION OF UNCERTAINTY 15-25
15.4.1. Uncertainty Analysis for
Cadmium 15-26
15.4.2. Uncertainty Analysis for
Benzo(a)pyrene 15-27
15.5. CONCLUSION 15-27
16. REFERENCES 16-1
APPENDIX A A_l
X
-------
LIST OF TABLES
Table
2-1
2-2
2-3
3-1
3-2
3-3
3-4
3-5
3-6
3-7
3-8
3-9
3-10
3-11
3-14
3-15
3-16
Ground Level Concentration of Pollutants and
Population Size Surrounding Two Model Combustors,
Geographic Mobility Rates as Related to Age.,
Description of Exposure Scenarios
Source Characteristics for MWC Analysis
Receptor Locations Used by COMPDEP and ISCST.
Receptor Locations Used by RTDMDEP for Terrain
Slope of 30, 45, and 60°
COMPDEP Model Options.
RTDMDEP Model Options.
ISCST Model Options...
Reflection Coefficients and Gravitational
Settling Velocities Used in ISCST Model Runs.
Average Atmospheric Concentrations (/Ltg/m3) ,
Neglecting Depositions Effects, Predicted by
COMPDEP and ISCST „
Average Atmospheric Concentrations
Neglecting Depositions Effects, Predicted by
RTDMDEP, COMPDEP and ISCST
Average Atmospheric Concentrations (jig/m3) ,
Including Dry Deposition Effects Predicted by
COMPDEP
Average Atmospheric Concentrations (/Ltg/m3) ,
Including Dry Depositions Effects, Predicted by
RTDMDEP
3-12 Average Annual Dry Deposition (g/m /yr)
Predicted by COMPDEP and ISCST.
3-13 Average Annual Dry Deposition (g/m /yr)
Predicted by RTDMDEP, COMPDEP and ISCST
Annual Dry Deposition for Benzo(a)Pyrene (g/m2)..
Annual Dry Deposition for Cadmium (g/m2) .........
Annual Wet Deposition for Benzo(a)Pyrene (g/m2)..
xi
Page
2-9
2-16
2-29
3-16
3-19
,3-21
3-22
3-23
3-24
3-26
3-27
3-31
3-34
3-36
3-38
3-40
3-46
3-47
3-48
-------
LIST OF TABLES (cont.)
Table Page
3-17 Annual Wet Deposition for Cadmium (g/m2) .......... 3-49
3-18 Average Hourly Concentrations for Benzo(a)Pyrene
........................................... 3_50
3-19 Emission Rates (kg/yr) for Planned MWCs ........... 3-52
3-20 Summary of Deposition Rates for Cadmium and
Benzo(a)Pyrene .................................... 3-53
4-1 Assumptions for Soil Concentration Calculations ___ 4-4
4-2 Input Variables for Determination of Cadmium
Soil Concentrations ............................... 4-20
4-3 Input Variables for Determination of
Benzo(a)Pyrene Soil Concentrations ............ . . . . 4-24
4-4 Summary of Soil Concentrations (mg/g) . „ ........... 4-29
5-1 Average Values of Productivity (kg DW/m2) ......... 5-15
5-2 Mean Wind Speed for Selected United States
Stations .......................................... 5-22
5-3 Parameters Used to Calculate az ................... 5-24
5-4 Biotransfer Factors for Selected Metals (Ba-. ..... 5-30
5-5 Description of Seven Plant Food Groups ............ 5-34
5-6 Consumption Rate of Seven Plant Groups (Cp. ....... 5-37
5-7 Percentage of Some Plant Foods Produced At
Home (Fpj) ................. 4 ...................... 5_39
5-8 Consumption Rates of Meat and Diary Products (Ca.)
(g DW/kg BW/day) ................................ ( . 5_43
5-9 Percentage of Animal Foods Produced At Home
5-45
5-10 Input Variables for Scenario B Exposure
Calculations: B(a)P ............................... 5-47
5-11 Input Variables for Scenario B Exposure
Calculations: Cadmium ............................. 5-62
5-12 Summary of Terrestrial Food Chain Exposure for
Scenario B ........................................ 5-77
xii
-------
LIST OF TABLES (cont.)
Table
6-1
6-2
7-1
7-2
9-1
9-2
9-3
9-4
9-5
9-6
9-7
9-8
9-9
11-1
15-1
15-2
15-3
15-4
15-5
Assumptions for Soil Ingestion Exposure
Pathway
Input Variables for Soil Ingestion Pathway.
Assumptions and Uncertainties for the Dermal
Exposure Model
Input Variables for Dermal Exposure Model...
Surface Water Methodology Assumptions
Groundwater Infiltration Model for Organics.
Groundwater Infiltration Model for Metals...
Input Variables for Determination of
Benzo(a)Pyrene Water Concentrations..
Annual Fallout and Facility Life for
Benzo (a) Pyrene.
Benzo(a)Pyrene Concentration in Receiving
Water for the Three Different Scenarios...
Input Variables for Determination of Cadmium
Water Concentrations
Annual Fallout and Facility Life for Cadmium.
Cadmium Concentration in Receiving Water for
the Three Different Scenarios.
Values for Input Variables Used in the
Scenario B Example
Results of Example Calculations of Daily Intake of
Benz(a)pyrene in Scenario B.
Results of Example Calculations of Daily Intake-
Adjusted of Benzo(a)pyrene in Scenario B
Results of Example Calculations of Daily Intake of
Cadmium in Scenario B
Breakdown of Cadmium Exposure for the Terrestrial
Food Chain and Human Soil Ingestion Pathways
Contribution to Daily Intake of Cadmium by
Vegetation Food Groups
Page
6-3
6-7
7-3
7-12
9-5
9-25
9-26
9-29
9-31
9-52
9-53
9-54
9-75
11-10
15-9
15-10
15-19
15-28
15-29
Xiii
-------
LIST OF TABLES (cont.)
Table
15-6 Contribution to Daily Intake of Cadmium by Meat
Food Groups 15-30
15-7 Breakdown of Benzo(a)pyrene Exposure for the
Terrestrial Food Chain and Human Soil
Ingestion Pathways 15-
32
15-8 Contribution to Daily Intake of Benzo(a)pyrene
by Vegetation Food Groups 15-33
15-9 Contribution of Daily Intake of Benzo(a)pyrene
by Meat Food Groups 15-34
xiv
-------
LIST OF FIGURES
Figure
2-1
2-2
4-1
8-1
8-2
9-1
15-1
15-2
Decision Tree for Defining Exposure Scenarios
Cumulative Population (Expressed as Percentage
of the Total Population within 50 km) and
Ground-Level Concentration (Expressed as
Percentage of Maximum Ground-Level Concentration)
as a Function of Distance from the Facility where
Ground-Level Concentration is the Areal Average
within Each Distance ,
Determination of Cumulative Soil Concentration..
Graph of Function F(x) Needed to Estimate
Unlimited Erosion
Ratio of Wind Speed at 7m to Friction Velocity
As a Function of Roughness Height.
Logic Flow for Groundwater Pathway Evaluation....
Page
2-4
2-8
4-2
8-9
8-11
9-22
Probability Distribution for Cadmium Daily
Intake Values by the Terrestrial Food Chain
and Human Soil Ingest ion Pathways. 15-31
Probability Distribution for Benzo(a)pyrene Daily
Intake Values by Terrestrial Food Chain
and Human Soil Ingestion Pathways. 15-35
xv
-------
LIST OF ABBREVIATIONS
A Area of source (m2)
Aa Adsorbed contaminant mass per unit of land in
watershed (kg/km )
AF Absorption fraction (%/day)
Ah, Area planted to crop i that is harvested (m2)
a{ and b{ Empirically determined parameters for each Pasquill
atmospheric stability category
Aj Concentration of pollutant in the jth animal tissue
group (/jg pollutant/g animal tissue DW)
AWQC Ambient water quality criteria
BHhk Human cancer potency for the hth pollutant by the
k™ route per(mg/kg)/day
Baj Biotransfer factor for the jth animal tissue
group (d/kg)
B(a)P Benzo(a)pyrene
BCF Bioconcentration factor (L/g)
BCFa Adjusted bioconcentration factor (L/kg)
BCFu Unadjusted bioconcentration factor (L/kg)
BD Soil bulk density (g/cm3)
BIhk Background intake for the hth pollutant by the kth
route (mg/kg BW/day)
Bri Plant-soil bioconcentration factor for the ith plant
group (jig pollutant/g plant tissue DW)/(ug
pollutant/g soil)
Bv Air-to-leaf concentration ([/xg/g DW]/[/zg/g])
Bvi Air-to-plant bioconcentration factor for the ith
plant group [ (p.g pollutant/g plant tissue DW)/(//g
pollutant/g air)
Body weight (kg)
c Assumed constant of l.VxlO"4 (atm-cm)
C "Cover management" factor (unitless)
xvi
-------
CA
Caj
Cd
Cd(
Cf
CHEMa
a
CHEMd
CHEM,
s1
CHEM.
•s2
CHEMW
Ci
L91
Ci
Lg23
Ci
sh1
cm
CN
Cs
LIST OF ABBREVIATIONS (cont.)
Contact amount (mg/cm2)
Daily human consumption of the jth animal tissue
group (g animal tissue DW/kg BW/day)
Cadmium
Pollutant concentration associated with particles
of the ith size category during the portion of the
hour in which no precipitation occurs (g/m )
Fish ingestion rate (g/kg BW/day)
Mass of chemical present in the water body in
adsorbed form (mg/L)
Mass of chemical present in the water body in
dissolved form (mg/L)
Mass of adsorbed chemical per mass of sediment in
the water body (/ig/L)
Mass of adsorbed chemical per unit volume of water
(/zg/L)
Mass of dissolved chemical per unit volume of water
Contaminant concentration in receiving water for
long-term, Tier 1 loading scenario (mg/L)
Contaminant concentration in receiving water for
long-term, Tier 2/3 loading scenario (mg/L)
Contaminant concentration in receiving water for
short-term, Tier 1 loading scenario (mg/L)
Concentration of contaminant in the receiving water
for the short-term or long-term loading scenario
(mg/L)
Centimeter
Curve number
Daily dietary consumption of the ith plant group by
the appropriate age group (g plant tissue DW/kg
BW/day)
Mean sediment concentration in the water body
(kg/m3)
Soil ingestion rate (g soil/day)
xvii
-------
CT
Cv
CW
CW,
Cy
d
D
Da
Da
DDI
DEPdj
DEPwf
DI
DIA
DIhk
DIFF
Dp
d_
LIST OF ABBREVIATIONS (cont.)
Contact time (hours/day)
Concentration of pollutant in air due to
volatilization from soil
DW
Consumption rate of water (L/kg BW/day)
Pollutant concentration in the portion of the hour
in which precipitation does occur associated with
particles in the ith size category (g/m3)
Concentration of pollutant in air due to direct
emissions (jig/m2)
day
Particle diameter (/zm) (Chapter 3 only)
Diffusion coefficient of pollutant in air (cm2/s)
Dissolved contaminant mass per unit of land in
watershed (kg/km2) (Chapter 9 only)
Daily dermal intake (mg/kg BW/day)
Effective diameter of contaminated area (m)
Total dry deposition for period (g/m2/s)
Total wet deposition for period (g/m2/s)
Daily intake (mg/kg BW/day)
Daily intake-adjusted (mg/kg/day)
Daily intake for the hth pollutant by the kthroute
(mg/kg/day)
Difference between receptor ground level elevation
and plume elevation (m)
Deposition (g/m2)
Distance between plants in a row (mm)
Distance between rows of plants (mm)
Depth 'of runoff in watershed (cm)
Duration of scenario (years)
Dry weight
xviii
-------
Dy
Dyd
Dyw
=10
EA
EDA
ER
ER
ERMix
ESP
EV
Fa
FaJ
FBC
FDRY
Fr
Fv
LIST OF ABBREVIATIONS (cont.)
Total yearly deposition rate of pollutant (g/m2/yr)
Yearly dry deposition rate of pollutant (g/m2/year)
Yearly wet deposition rate of pollutant (g/m2/year)
Annual average PM10 emission factor (g/m2/hr)
Excess air ,
Exposure duration adjustment (unitless)
Excess risk (unitless)
Excess risk for the hth pollutant for m exposure
routes (unitless)
Excess risk for the hth pollutant by the kth route
(unitless)
Excess risk for a mixture of n chemicals for
multiple exposure routes (unitless)
Electrostatic precipitator
Average soil evapotranspiration (cm/year)
Annual mass of contaminant in fallout per unit area
(kg/km2-year)
Fraction of the jth animal tissue group assumed to
originate,from contaminated soil (unitless)
Fluidized bed combustor
Correction factor for the fraction of the hour in
which no precipitation occurs
Fraction of time of the ith atmospheric stability
category (unitless)
Fraction of the ith plant group assumed to originate
from contaminated soil (unitless)
Froude number
Wind erosion threshold friction velocity (m/s)
Fraction of pollutant in vapor phase (unitless)
xix
-------
Fw
g
g
GIM
H
H
h
ha
Hcrit
HE
HI
HI,.
HI
HI
hfc
Mix
HMIN
Hpc
hr
Ht
Htc
I
i
la
Ias
LIST OF ABBREVIATIONS (cont.)
Fraction of wet deposition that adheres to plant
surfaces (unitless)
Acceleration of gravity (9.8 m/s2) (Chapter 3 only)
Gram
Groundwater Infiltration Model
Henry's Law constant (pa m3/mol)
Plume height (m) (Chapter 3 only)
(subscript) denotes pollutant
Hectare
Critical plume height (m)
Effective height of source emission (m)
Hazard index (unitless)
Hazard index for the hth chemical for all routes
(unxtless)
Hazard index for the hth chemical by the kth route
(unitless)
Hazard idex for a mixture of n chemicals for
multiple exposure routes (unitless)
Minimum distance between plume and ground (m)
Height of plume above critical height (m)
Hour
Height of plant above soil (m)
Height of local terrain above critical height (m)
Average annual irrigation (cm/year)
(subscript) denotes plant group
Total daily intake from consumption of animal tissue
(Mg pollutant/kg/BW/day)
Total daily intake from ingestion of jth animal
tissue group (jug pollutant/kg BW/day)
xx
-------
IP
ISCLT
ISCST
j
K
k
1E
Ke
kg
km
kp
Ko«
ks
ksg
ksl
ksv
Kt
LIST OF ABBREVIATIONS (cont.)
Total daily intake from consumption of plant tissue
pollutant/kg/BW/day)
Human daily intake of pollutant due to the
consumption of the i plant group (/zg
pollutant/kgBW/day)
Infiltration rate (cm/yr)
Industrial Source Complex Long-Term Model
Industrial Source Complex Short-Terra Model
(subsrcipt) denotes animal group
"Erodability" factor (tons/acre)
(subscript) denotes exposure route
Rate constant for zeroth order losses (year"1)
First-order loss rate (year"1)
First-order loss rate for degradation (year"1)
First-order loss rate for erosion (year"1)
First-order loss rate for infiltration (year"1)
Soil water partitioning coefficient (mL/g) or
(m3/kg)
Equilibrium coefficient
Kilogram
Kilometer
Plant surface loss coefficient (year"1)
Octanol-water coefficient (uriitless)
Soil loss constant (year"1)
Loss constant due to degradation (abiotic and
biotic)(year)
Loss constant due to leaching (year"1)
Loss constant due to volatilization (year"1)
Gas phase mass transfer coefficient (cm/s)
xx i
-------
L
L
1
Lc
LOAEL
LS
m
MB
Mg
mg
MCWI
MI
Mm
mol
"t
MUSLE
MW
MWC
MSW
N
N
n
LIST OF ABBREVIATIONS (cont.)
Liter
Mixing depth of inversion lid (m) (Chapter 3 only)
Length of the unit area (mm)
Lipid content of dietary seafood (kg/kg)
Lipid content of experimental organism (kg/kg)
Lifetime duration (years)
Length of long fruit (mm)
Lowest-observed-adverse-effect level
"Topographic or slope length" factor (unitless)
Meter
Mass burn incinerator
Megagram
Milligram
Maximum cross wind integrated concentration
Modular incinerator
Milliliter
Maximum contaminant mass per area of soil (kg/km2)
Mole
Depth of snow melt for storm event (cm)
Modified Universal Soil Loss Equation
Maximum contaminant mass per volume of water
(kg/km3)
Municipal waste combustor
Municipal solid waste
Eddy reflection number (unitless) (Chapter 3 only)
Schmidt number for gas phase (unitless)
Number of fruit per square meter
xxn
-------
NFCS
NOAEL
nr
ng
p
pa
PAH
PCB
PCDD
PCDF
Pd,
Pft
PM
PM,
10
qt
PV
LIST OF ABBREVIATIONS (cont.)
National Food Consumption Survey
No-observed-adverse-effect level
Number of plants per row
Nanogram
Average annual precipitation (cm/year)
Density of air (1190 g/m3 at 25°C)
Polycyclic aromatic hydrocarbons
Polychlorinated biphenyls
polychlorinated dibenzodioxins
polychlorinated dibenzofurans
Concentration of pollutant in the ith plant due to
direct deposition
Mass of chemical deposited directly onto the water
body (kg)
Total concentration of pollutant in the ith plant
(jug pollutant/g plant tissue DW)
Total concentration of the ith plant group eaten by
pollutant/g plant tissue
the jth animal each day
DW)
Particulate matter
Particulate matter <10
Mass of chemical entering water body in dissolved
form (kg)
Concentration of pollutant in the ith plant due to
root uptake (ng/g)
"Supporting practice" factor (unitless)
Concentration of pollutant in the ith plant due to
air-to-plant transfer
Mass of chemical entering water body in adsorbed
form (kg)
Liquid-phase vapor pressure
xxiii
-------
Q
Q
R
R
R
R10
RC
Rc
RDF
RE
RE0
Ref
RfD
RfDo
RfD,
'hk
LIST OF ABBREVIATIONS (cont.)
Emission rate of the pollutant from the soil
(Mg/m2/s)
Source emission (g/s)
Volume of runoff (km2-cm) (Chapter 9 only)
Peak runoff (m3/sec)
Quantity of the ith plant group eaten by the jth
animal each day (kg plant tissue DW/day)
Quantity of soil eaten by the jth animal each day
(kg soil/day)
Ideal gas constant (L-atm/mole-°K) (Chapter 4 only)
"Erosivity" factor (year"1) (Chapter 9 only)
Radius of the contaminated area (m) (Chapter 5 only)
Annual emission rate of the contaminant
Rotary combustor
Receptor
Refuse derived fuel
Relative effectiveness (unitless)
Relative effectiveness of the dermal route
(reciprocal od the absorption fraction for ingested
chemical) (unitless)
Interception fraction for exposed produce
Refractory-lined
Radius of individual fruit or plant (mm)
Reference dose for chronic inhalation exposure
(mg/m3)
Reference dose for chronic oral exposure (mg/kg/day)
Reference dose for chronic exposure to the hth
pollutant by the kth route
Interception fraction of long fruit (unitless)
xx iv
-------
R
lv
n
Rpg
RPl-
R,
rf
S
s
SA
SA
Sc
SCS
ST
timef
Tf
T
**
Ta
TBI
Tc
LIST OF ABBREVIATIONS (cont.)
Interception fraction for leafy
(unitless)
vegetables
Interception fraction for mature leafy vegetables
(unitless)
Number of rows of plants
Interception fraction for pasture grasses (unitless)
Interception fraction of the edible portion of plant
tissue for the ith plant group (unitless)
Interception fraction of round fruit (unitless)
Depth of total rainfall for storm event (cm)
Ratio of wind speed at 7m to friction velocity as
a function of roughness height
Second
Watershed retention parameter (cm)
Exposed skin surface area (cm2)
Starved air (Chapter 3 only)
Soil concentration of pollutant (/ig pollutant/g
soil)
Soil Conservation Service
Whitby's average total surface area of particles
(cm2/ cm3)
Time span that analysis will be carried out (years)
Time frame of the model (years)
Temperature (° K)
Environmental half-time (days)
Ambient temperature
Total background intake of pollutant (mg/kg/day)
Total time period over which deposition occurs
(years)
xxv
-------
TCDD
TDI
TER
TFC
THT
TOTCHEM
Tp,
[U]
U
U6-hr
U
USLE
ut
V
VDF
VPTG
VSx
W
W
LIST OF ABBREVIATIONS (cont.)
c
2,3,7,8-Tetrachlorodibenzo-p-dioxin
Total daily intake (mg/kg/day)
.th
Total daily intake of the hth pollutant by the k
route (mg/kg/day)
Terrain adjustment factor ranging in value from 0.0
to 1.0
Terrestrial food chain
Height of terrain obstacle above stack base
elevation (m)
Total mass of chemical in water body (kg)
Length of plants exposure to deposition per harvest
of the edible portion of the i™ plant group
Duration of storm event (hours)
Mean annual wind speed (m/s)
Stack top wind speed (m/s)
Maximum 6 hr. mean wind speed (m/s)
Mean wind speed at plume height and receptor
location (m/s)
Universal Soil Loss Equation
Threshold wind speed
Fraction of contaminated surface vegetation
Dry deposition velocity (m/s)
Vertical distribution function
Appropriate dilution volume for water body under
consideration (m )
Vertical potential temperature gradient (°K/m)
Appropriate dilution volume for watershed under
consideration over the duration of storm event (m3)
Gravitational settling velocity (m/s)
Width of the unit area (mm)
xxvi
-------
WAw
We
WW
x
Yh,
yr
z
Z
e
LIST OF ABBREVIATIONS (cont.)
Land area of watershed receiving fallout (km2)
Area of water receiving fallout (km2)
Water concentration (mg/L)
Water walls
Horizontal upwind distance of source from receptor
(m)
Sediment loss rate per unit area watershed over time
(kg/km -year)
Concentration of contaminant i in groundwater (mg/L)
Mass of sediment in the water body (kg)
Horizontal crosswind distance from plume centerline
to receptor (m)
Harvest yield of ith crop (kg DW)
Yield or standing crop biomass of the edible portion
of the i plaint group
year
Vertical coordinate of receptor (m) (Chapter 3 only)
Soil depth (m)
Gaussian plume dispersion parameter for the vertical
direction (m)
Gaussian plume dispersion parameter for the
horizontal direction (m)
Scavenging coefficient (1/s)
Viscosity of air (g/cm-s)
Microgram
Micrometer
Soil volumetric water content (ml/cm3)
XXVil
-------
LIST OF ABBREVIATIONS (cont.)
Pollutant concentration in the PM10 fraction of soil
Fraction of chemical adsorbed (unitless)
Fraction of chemical dissolved (unitless)
xxviii
-------
1. INTRODUCTION
1.1. PURPOSE
This methodology document seeks to provide risk assessors with
the guidance necessary to estimate the health risks that result
from exposure to toxic pollutants in combustor emissions by
pathways other than inhalation. Procedures for assessing human
health risks from inhalation of combustion facility emissions are
well established; this methodology sets out procedures for use in
estimating the indirect human exposures and health risks that can
result from the transfer of emitted pollutants to soil, vegetation
and water bodies.
The organization of the document reflects the four-step process
of risk assessment (hazcird identification, dose-response
assessment, exposure assessment and risk characterization)
initially described by the National Research Council (NRC, 1983).
In this document, however, exposure assessment is presented before
the other steps.
This methodology is not intended to be prescriptive; that is,
it does not comprise a set of guidelines or recommended approaches
that the U.S. EPA believes should be applied in all circumstances.
Rather, it provides a set of procedures that the risk assessor can
draw upon, where applicable, to a given assessment.
The document describes analytical procedures and computer
models that can be used to estimate exposure and risk by a variety
of environmental pathways. In addition, it serves as a
preliminary source of data for carrying out the risk calculations.
1-1
-------
The degree of scientific support or uncertainty attendant to each
calculation varies widely, as the careful reader will appreciate.
Therefore, the appropriate use of these procedures and the
discussion of uncertainties surrounding the results remain
important responsibilities of the risk assessor.,
1.2. BACKGROUND ......
Faced with increasing problems of land disposal, many
communities are actively considering incineration as a disposal
alternative. Incineration is being used as a method of disposal
for municipal solid waste (MSW), sludge and hazardous waste. A
fundamental issue attendant with the growth of incineration is the
environmental impact of pollutants in the combustor emissions. In
the past, most analyses of human health risk associated with
atmospheric emissions from combustion sources have focused only on
exposures occurring by direct inhalation. Recent studies, however,
have linked elevated levels of pollutants in soils, lake sediments
and cow's milk to atmospheric transport and deposition of
pollutants from combustion sources (Fradkin et al., 1988). These
studies indicate that deposition of atmospherically emitted
pollutants could result in indirect avenues of exposure for humans.
Therefore, this methodology was developed to examine indirect
exposures and risks to emission from a variety of ,combustion
sources. This methodology does not address exposure and health
effects from incinerator ash. A forthcoming document from the
1-2
-------
Exposure Assessment Group (U.S.. EPA) entitled Exposure Assessment
Methodology for Municipal Waste Combustor Residuals Management
considers incinerator ash exposure.
One type of incinerator that is being increasingly used is
the municipal solid waste combustor (MWC). At least 4% of the
greater than 155 million tons of MSW generated each year is
incinerated at 111 facilities in the United States. The U.S. EPA
projects that by the year 2000, >300 facilities nationwide will
incinerate >90 million tons of MSW annually. At present, 210
facilities are in the planning stages or under construction and
projected to be complete by 1993. There is a limited data base
regarding the type and guantity of pollutants that may be emitted
during the combustion of MSW.
Although many MWCs have been tested for compliance with
particulate emission standards, a limited number of tests for
specific pollutants have been conducted. The U.S. EPA's data base
on MWC pollutant emissions consists of test results from three
types of MWCs: mass burn, refuse derived fuel (RDF), and modular.
Emissions from MWCs are known to contain heavy metals and toxic
organic compounds. The metals of concern include arsenic,
beryllium, cadmium, chromium, mercury, nickel, and lead. The
organics of concern include chlorinated dioxins/furans, polycyclic
aromatic hydrocarbons (PAH), polychlorinated biphenyls (PCB),
formaldehyde, chlorobenzenes,, and chlorophenols.
Incineration is also a common method of disposal for sewage
sludge. The U.S. EPA estimates that approximately 1.7 million dry
metric tons of sewage sludge are incinerated each year, accounting
1-3
-------
for 21% of all sewage sludge disposal (Federal Register, 1989).
Currently, 169 facilities use 282 incinerators to dispose of
sludge; these facilities represent 1% of the facilities engaged in
sludge disposal. The most common type of sludge incinerator is the
multiple hearth incinerator; 231 of the incinerators in use are
this type. Emissions from the combustion of sewage sludge are
likely to contain sulfur dioxide, oxides of nitrogen, heavy metals,
toxic organic compounds, and hydrocarbons. The U.S. EPA is
proposing numerical limits on the following pollutants for
combustion of sewage sludge: arsenic, beryllium, cadmium, chromium,
lead, mercury, nickel, and total hydrocarbon. Total hydrocarbon
is the sum of all emitted organic compounds. Some specific
organics identified in sludge incinerator testing include
chlorinated benzenes, benzene, chloroform, dichloroethene,
methylene chloride, and trichloroethene.
Incineration is one method available for the disposal of
hazardous waste. The most recent data on hazardous waste
incineration was obtained from a 1981 U.S. EPA survey (described
in Freeman, 1988). in 1981, approximately 265 million metric tons
of waste was generated, consisting of mostly nonhazardous products
contaminated with smaller amounts of hazardous materials.
Approximately 5.5 million metric tons was disposed by thermal
destruction (1.7 million metric tons in 240 incineration facilities
and 3.8 million metric tons in 1300 industrial boilers and
furnaces)(Freeman, 1988).
Of the 240 hazardous waste incinerators operating in 1981, 49%
were liquid injection combustors, 19% were fixed hearth combustors,
1-4
-------
and 6% were rotary kiln combustors. Only 45% of the facilities
employed air pollution control technology in 1981. Typical systems
included combustion gas quench, venturi scrubber, acid gas
adsorber, and mist eliminator (Freeman, 1988).
The major waste streams incinerated in 1981 were spent
nonhalogenated solvents and corrosive and reactive wastes
contaminated with organics. Hydrocyanic acid and contaminated
water were also important waste streams. The typical pollutants
emitted from incineration of these wastes include acid gases (HCl,
SO2, NOX) (Freeman, 1989) and the following products of incomplete
combustion: benzene, chloroform, tetrachloroethylene, 1,1,1-
trichloroethane, toluene, and methylene chloride (Freeman, 1988,
1989).
1.3. SCOPE
In the past, the U.S. EPA has focused on the health risks
posed by inhalation of pollutants in emissions from combustion
sources. The Environmental Criteria and Assessment Office (ECAO)
of the Office of Health and Environmental Assessment, Office of
Research and Development (ORD) has developed this methodology to
expand these analytical horizons. This methodology permits the
assessment of the potential human health risks posed by indirect
exposure pathways resulting from wet and dry deposition and from
ambient air concentrations of pollutants emitted from the stacks
/'
of combustors. This methodology can be applied to all emitted
pollutants (multipollutant).
1-5
-------
Pollutants from stack emissions can be incorporated into soil
or water and, from .there, can accumulate in the terrestrial or
aquatic food chains. Humans can be exposed to these pollutants in
various media (multimedia) by dermal contact, direct ingestion
and/or ingestion of foods grown in the contaminated soil or water.
This methodology guides risk assessors through the four steps
of risk assessment to characterize the exposure and risk posed by
pollutants in various pathways. The document starts by presenting
guidelines on developing site-specific exposure scenarios (Chapter
2) . After discussing air dispersion models appropriate for
analyzing combustor emissions (Chapter 3), the document describes
assessment of exposure through terrestrial and aquatic systems.
For terrestrial systems, the document presents methods for
calculating pollutant concentration in soil (Chapter 4) and
calculating daily intake through ingestion of food (Chapter 5),
ingestion of soil (Chapter 6) , and dermal absorption from soil
(Chapter 7) . Exposure via inhalation of resuspended dust is
evaluated (Chapter 8) and found to be an insignificant pathway of
exposure to combustor emissions.
For aquatic systems, the document presents methods for
calculating pollutant concentration in surface water, rainwater,
and groundwater (Chapter 9) . Groundwater was found to be an
insignificant pathway of exposure to combustor emissions. Methods
for calculating daily intake through ingestion of water (Chapter
10) , ingestion of fish (Chapter 11) , and dermal absorption from
water (Chapter 12) are presented. The document discusses hazard
1-6
-------
identification (Chapter 13) and dose-response assessment (Chapter
14) as they apply to assessment of combustion emissions. Risk
characterization is presented in Chapter 15.
The U.S. EPA recognizes that there are uncertainties inherent
in the methodology. For example, the methodology can be applied
only to the finite number of chemicals that have actually been
identified in combustor emissions. There are potentially hundreds
of pollutants that may be emitted but have not been looked for or
measured. This document uses cadmium and benzo(a)pyrene as
examples of how these methods can be applied. Although the most
current research available on these chemicals was used in the
examples, readers should not consider this document to be a
definitive statement of exposures and risks posed by these two
chemicals. U.S. EPA (1988a) has prepared a document similar to
this methodology that focuses on exposures to and risks posed by
dioxins present in combustor emissions.
The document uses emission rates that were, for the most part,
measured during good operations of the incinerator. Therefore,
emissions resulting from perturbations, mechanical failures,
excessively wet garbage feed, shutdown and cold start of the plant
are not reflected in the exposures or risks presented.
There are other data limitations that ultimately constrain
the methodology. This document uses the most current national
average values as input into the example calculations for the
purpose of illustrating the methods. Readers should consider that
many of the factors affecting exposure vary according to region,
climate, or local habits. Risk assessors should use local data to
1-7
-------
conduct a site-specific assessment whenever possible to reduce the
uncertainty associated with applying national data to a local
community.
The methodology, therefore, is not a comprehensive
environmental audit, but is best regarded as an evolving and
emerging process that moves the U.S. EPA beyond the analysis of
potential effects on only one medium (air) and exposure pathway
(inhalation) to the consideration of other media and exposure
pathways.
1-8
-------
2. HUM&H EXPOSURE SCENARIOS
2.1. INTRODUCTION
Pollutants emitted to the atmosphere from stationary
combustion facilities may be deposited on environmental media,
i.e., soil, water and vegetation, near the combustor. Humans or
other organisms can be exposed to the emitted pollutants by various
pathways such as dermal contact with the pollutants in soil or
water, consumption of the exposed environmental media, consumption
of animals that have ingested exposed soil, water or vegetation,
or consumption of fish living in polluted waters. For humans, an
important part of a multimedia, multipollutant exposure assessment
is defining the individual and/or population in the vicinity of the
combustor and the potential pathway(s) by which the individual(s)
may be exposed to the emitted pollutants.*
Exposure scenarios are real or hypothetical situations that
define the source, individual(s) , the pathway(s) of exposure and
the variables that affect the exposure pathways. Once the exposure
scenarios are defined, quantitative exposure and risk estimates for
adverse health effects may be developed.
This chapter presents guidelines for developing site-specific
exposure scenarios. Actual exposure scenarios can be constructed
on a site-specific basis by the exposure/risk assessor using the
various pathways presented in this document (Chapters 4-12). Three
*This methodology is geared toward protection of the general
public and the environment. Occupational exposures are not
considered.
2-1
-------
hypothetical exposure scenarios will be developed in this chapter
to illustrate possible approaches. These three scenarios will be
used throughout the document to illustrate calculations for the
exposure pathways. Scenario "A" will be constructed to represent
the most likely occurrence for an exposed population anywhere
within a 50 km radius of the facility, and typical or average
values of input variables are chosen. In contrast, Scenario "C"
will illustrate the hypothetical case wherein all variables are
maximized so that the highest potential exposure to an individual
is determined. While most persons in the vicinity of a combustion
facility are not likely to have such an exposure, this possibility
may exist and should be explored. Scenario "B" will strike an
approximate midpoint between Scenarios A and C; exposures will be
higher than those in Scenario A but not as infrequent as those in
Scenario C. These scenarios will be referred to in the following
chapters and more fully developed in Section 2.6. The scenarios
constructed in this document are examples only and the risk
assessor can define/construct exposure scenarios for any site.
Uncertainties in the variables that affect the exposure pathways
will be delineated in the chapter for each pathway.
2.2. CREATING EXPOSURE SCENARIOS
Contaminants associated with atmospheric emissions from
combustion sources may be deposited on the surfaces of soil,
vegetation and surface water bodies downwind from the source.
Human exposure scenarios can be developed by defining the
individuals that could be exposed; the characteristics of the
2-2
-------
source, such as its length of operation; the environmental media
in the surrounding area; and the possible routes of exposure to
the contaminated media. Numerous exposure scenarios are,
therefore, possible. Figure 2-1 presents a decision tree that can
be used to define the exposed individual(s) and select appropriate
exposure pathways to tailor a site-specific assessment.
The exposed individual(s) must first be defined in terms of
age, location relative to the combustion source, and length of time
spent at that location. Although the age of the individual will
affect exposure routes and pathways, defining age in general terms
such as child or adult generally suffices for exposure assessment.
In most cases, location will refer to the residence of the
individual. The length of time (i.e., number of years) that the
individual spends at the defined location influences potential
pathways of exposure and exposure levels.
Various characteristics of the combustor affect human
exposure. These factors include the type and capacity of the
combustor, emission rate, emission control technology such as
electrostatic precipitators, the type of materials incinerated,
and the length of time the combustor has been in operation. These
characteristics will be further discussed in Chapter 3.
The environmental media likely to be exposed to deposited
emissions should be defined, since the environment surrounding the
source also contributes to selection of the pathways by which the
individual(s) may be exposed to the emitted pollutants.
Potentially contaminated environmental media include food
commodities, soil and water. The potential routes of exposure to
2-3
-------
c
5
U "if
e 3
11
m °
O
%
B3
« 3
|1
SO
O
__
jo |,S
cS o
0 c
111
^51
Q 0 »
w
o
•H
fc
(0
c
q
Q)
•H
En
2-4
-------
the contaminated media include ingestion, dermal absorption and
inhalation. The topography of the area and its meteorologic and
climatic conditions influence determination of relevant pathways
(also discussed in Chapter 3). For example, consumption of water
and fish may become pathways for exposure to the emitted pollutants
if a surface water .body, such as a lake or stream, are located in
the area surrounding the combustion facility. Whether the
surrounding countryside is farmland or an urban area influences the
importance of the soil pathways in constructing a site-specific
exposure scenario.
Once the individual and environment are defined and the
potential pathways of exposure chosen, mathematical models of
environmental fate and transport can be used to estimate the
concentration of pollutants in the various environmental media in
the vicinity of the combustor. In this document, "vicinity" is used
to represent the area within a 50 km radius of the source. The
remaining chapters present models used to estimate pollutant
I
concentration in food, soil and water, and to calculate daily
intake of pollutants by the various routes of exposure. By varying
the values of data entered into the models from typical to worst-
case, a range of potential exposures can be estimated. For
example, in the Terrestrial Food Chain Model, the amount of
contaminated food consumed may be varied from a minor fraction to
most of the food source being contaminated in order to determine
a range of exposure for that pathway. The risk posed by each
exposure pathway is assessed by comparing the estimated exposures
with health-based criteria. The exposure pathways are summarized
2-5
-------
in Section 2.4. and described in detail in Chapters 4-12.
Estimation of human health risks is delineated in Chapters 13-15.
2.3. DEFINING THE LENGTH OF TIME OF EMISSIONS
In this methodology, individuals are assumed to be exposed to
long-term average soil or water levels of the pollutant present in
the environment following a period of time in which there has been
continuous combustor emissions. Neither upset conditions nor
intermittent emissions are considered in the present methodology.
Varying lengths of time of operation of the combustor can be used
to estimate the pollutant concentration. The use of combustion
facility characteristics and local topography and meteorology to
model contaminant deposition rates is described in Chapter 3.
Determination of soil and water concentrations resulting from
deposition are described in Chapters 4 and 9, respectively. For
example, in the case of a municipal waste combustor (MWC), the
lifetime of the facility (or total period of deposition) could be
considered to be approximately 30 years. Since the site is already
dedicated for a MWC, it could further be assumed that the MWC will
be replaced and the lifetime of combustor emissions could be as
great as 100 years. In this document, 30, 60 and 100 years of
continual fallout from a combustion facility will be used to
construct Scenarios A, B and C, respectively.
2-6
-------
2.4. DEFINING THE EXPOSED INDIVIDUAL
2.4.1. Location of the Individual Relative to the Combustion
Source. The ground-level concentration and deposition rate of
pollutants emitted from the combustion source are affected by air
dispersion of pollutants. Numerous factors such as source
characteristics (e.g., stack height), topography, meteorological
data (e.g., wind speed and direction) and distance from the source
determine the air dispersion and resultant deposition of the
emitted pollutants. Annual ground-level concentration of emitted
pollutants decreases with distance from the source as shown in
Figure 2-2 and Table 2-1, which are based on a modeling analysis
of two sewage sludge combustors. Maximal ground-level
concentrations are usually measured at distances in very close
proximity to the source, although local meteorological conditions
and source characteristics play a role. For the combustors used in
Figure 2-2, maximal deposition is approximately 0.2 km from the
facility. At 2 km, the ground-level concentration has attenuated
to approximately 4-14%, and at 50 km from the source the
concentration is <0.02% of the maximum ground-level concentration.
In generalizing this information to other combustors, the location
of the individual relative to the combustor becomes an important
determinant of exposure. Generally, individuals located closer to
the source would be expected to have higher direct exposure to
emitted pollutants than those at further distances from the source.
In most cases location will refer to the residence of the
individual. A range of exposures for individuals representing the
2-7
-------
«">! 89
— n
i
— cu
ft f\\ pi fl
O Cn O *H
•H 4J <4-| 4-1
(0 C -H
H 0) 0) £
3 P O m
ft M C ®
o a) !3 31
ft &4-> g
•H W-H SJ
(0 nJ TJ £
<1> W C
•ft 9> o
a)
a)
§
ft 8
0 w 9
•P °
W fli S H
w S 8 a>
s
2
I
. o
^ ^H
Id's-2
« M
Cn
0)
S * § 0
ft o -S •
g'^M & a,
ft^ ^j^o
SoSS^
•H m s .71 w
i»ca8*
H -H nJ (jlj
e -p a) o
3 -H
-------
TABLE 2-1
Ground Level Concentration of Pollutants and Population
Size Surrounding Two Model Combustors
Distance from
the Plant
(km)a
Model Plant 50
1
10
5
2
0.2
Od
Model Plant 50
2
10
5
2
0.2
od
Relative
Population Size
4.0 x 106
(100%)
4.1 x 105
(10%)
8.3 X 104
(2.1%)
2.3 X 103
(0.06%)
<2.3 X 103
(<0.06%)
(<0.06%)
3.7 X 105
(100%)
4.2 X 104
(11%)
1.8 X 104
(4.9%)
3.1 X 103
(0.8%)
6.8 X 102
«o.a%>
Relative
Concentration
0.2
1.8
4.7
14
71
100
0.04
0.5
1.3
4.2
45
100
aDistance refers to radial distance from the source
Population within area of specified distance and equivalent
percent of population relative to total population within 50 km
radius of the facility
°Percent of maximum ground-level concentration at specified
distance, where ground-level concentration is the areal average
within each distance
dThis distance refers to the source location
2-9
-------
general public could be estimated by choosing several distances
from the combustor. In this document, different distances from the
facility will be used for exposure Scenarios A, B and C.
Another factor that varies with distance from the combustion
facility is the size of the population. Figure 2-2 shows that
cumulative population increases as distance from the combustion
facility increases; increasing radial distances from the source
encompass more of the population. In this document, a 50 km radius
around the combustion facility will be used to define the vicinity
around the source. The total population in this area then
represents all persons who could be exposed to the emissions,
excluding those persons that temporarily travel into the area
(e.g., for work) . At a 10 km radius of the facility, approximately
10% of the total population is found; within a 5 km radius, 2-5%
of the population resides. Therefore, the individual used in each
of the three exposure scenarios in this document can also be
described in terms of the size of the population he represents.
For example, an individual in Scenario C resides in close proximity
to the combustion facility and represents a small group of people
with high exposure because concentration of the emitted pollutants
are maximal and have not significantly attenuated over the short
distance. An individual residing at a 50 km distance from the
source will be exposed to low levels of the emitted pollutants and,
therefore, will be more representative of the general population
within the vicinity of the combustor.
In this document, a person residing at the point of maximum
deposition, a 0.2 km radius from source, is chosen for Scenario C.
2-10
-------
As shown in Figure 2-2, this individual will come from a group of
<0.3% of the total population who reside at a point where the
areally averaged ground-level concentrations are approximately 75%
of maximum ground-level concentrations. Although this is a
hypothetical situation, it is not unrealistic in that there may be
some areas in which a combustion facility is "across the street"
or "a block away" from a residential area. The exposure Scenario
B will be represented by an individual residing at 5 km and exposed
to the average concentration of pollutants within that 5 km radius
of the source. This individual represents 2-5% of the total
population. Scenario A is that of an individual living 50 km from
the facility, exposed to the areally-averaged ground-level
concentration of pollutants within that 50 km radius of the source.
Determination of the ground level concentrations for these
scenarios is presented in Chapter 3.
2.4.2. Age of Exposed Individual. The age of the individual
(child vs. adult) influences exposure to the emitted pollutants
since pathways, duration and quantities of exposure may vary with
age. The daily activities of the individual, amounts of food and
water consumed, types of food consumed and exposed skin surface
area differ for children and adults. Health-based criteria are
also different for children and adults. Therefore, for some
exposure pathways such as soil ingestion, children may have a
greater exposure and be at greater risk than adults. Human
lifespan is generally considered to be 70 years (U.S. EPA, 1986a),
and childhood represents approximately 10% of the lifespan, or 6
2-11
-------
years. In actual exposure scenarios, individuals may be exposed
only during childhood or adulthood. In other cases, exposure may
overlap these periods, such as in the case where a child grows into
adulthood and remains in the same geographical area. For the three
hypothetical exposure scenarios used in this document, a child will
represent the individual in Scenarios B and C; an adult will
represent the individual in Scenario A.
Childhood is defined in the literature differently by various
authors. Data for input into the exposure pathways described in
Chapters 4-12 may not be available for children <7 years, or may
be presented in the literature for more restrictive age groups such
as 2 year olds or children 4-6 years old. The definition of
childhood is, therefore, constrained by available data and may vary
somewhat for the following exposure pathways. Interpolations or
extrapolations of the available data are sometimes made to estimate
values for children 1-7 years old.
2.4.3. Human Body Weight (BW). The choice of body weight for use
in risk assessment depends on the definition of the individual at
risk, and that in turn depends on exposure and susceptibility to
adverse effects. In this document the individual at risk is
defined in each exposure scenario. The body weight used to
determine daily intake is that of an adult or a child, or both,
depending on the scenario. Scenario C is a child who grows to an
adult and is exposed for his 70 year lifetime. In Scenario B, the
2-12
-------
individual is exposed for a total of 30 years (as a child and then
for part of his adulthood) . In Scenario A, an adult is exposed for
16 years.
The daily intake for a pathway was defined as the dose rate
on a body-weight basis, Ingestion exposures on a body weight basis
are substantially higher for infants and toddlers than for
teenagers or adults. Certain behaviors, such as mouthing of dirty
objects or direct ingestion of soil, which could also contribute
to exposure, are also much more prevalent in children than adults.
Therefore, infants and toddlers may be at greater risk when
exposure is by food or soil ingestion. The effects, however, may
have a long latency period, in some instances approaching the human
lifespan. In these cases it may be reasonable to estimate daily
intake using the adult values of body weight. In cases where
effects have a shorter latency (<10 years) and where children are
known to be at special risk, it may be more appropriate to use a
body weight for toddlers or infants or for young children. For
chemicals that are classified as carcinogens, adult body weight is
usually assumed to derive unit risk estimates for air or water.
If exposure is lifelong, values of body weight are usually chosen
so as to be representative of adults. In this document, 70 kg is
used as the body weight for adults. The average body weight for
children 1-7 years old is 17 kg (Nelson et al., 1969). This value
is used for exposure pathways where information for the inputs in
the pathway are available for children 1-7 years old. In some
instances, information is not available for all children in this
age range and the body weight for children is constrained by the
2-13
-------
form of the data. For example, the food ingestion rate for meat
groups is presented only for 2 year olds (U.S. EPA, 1989b). The
body weight chosen, then, to determine daily intake (mg/kg/day)
via food ingestion is the mean body weight of male and female 2
year olds (Nelson et al., 1969) . The choice of body weight is made
to coincide with the available data, so that the daily intake
(mg/kg/day) is accurately determined. The form of the input data
and body weight value used for determination of daily intake (Dl)
are delineated in the chapter for each pathway.
2.4.4. Length of Time Individual Spends at Location. A person
may possibly have a lifetime exposure to emissions from a
combustion source; however, census data on population mobility
suggest that many Americans do not remain in the same area for
their 70-year lifetime (U.S. Bureau of the Census, 1986).
Therefore, a shorter period of time may also be considered to
estimate shorter duration exposure to combustor emissions.
An estimate of the number of years a person is likely to spend
in a given area, such as in the vicinity of a combustion facility,
can be derived from information about mobility rate and median time
in a residence. Mobility rate refers to the number of persons over
1 year of age (i.e., percent of the U.S. population) moving to a
different residence in the United States within a given year.
Data derived from the U.S. Bureau of the Census (1986) show that
the mobility rate in 1984 was 17% of the total population. Local
moves, those within the same county, accounted for 60% of all
moves. It could be inferred from this information that the majority
2-14
-------
of those that move would remain in the vicinity of the combustion
source. However, the likelihood that these short distance moves
will Influence exposure based on factors such as atmospheric
transport of pollutants can not accurately be predicted. Overall
rates and volume of geographical movement .(number of movers) vary
as a function of age, region of the country and type of settlement
(i.e., metropolitan area). Mobility rates have tended to decrease
with increasing age as shown in Table 2-2. However, since mobility
of the U.S. population may change over time, it cannot necessarily
be inferred that these rates will remain constant in the future,
e.g., over the next few decades. For example, the rates decrease
in farming areas; once a farm is established, its owners tend to
stay for many years.
The average number of years a person stays in his/her dwelling
also affects exposure. This median length of residence in 1980 was
30 years for homeowners, 4 years for renters, and 16 years for
owners and renters combined (Burke, 1986).
In addition to the number of years at a particular location
or residence, the amount of time spent at that location each day
directly affects exposure. For example, children that attend day
care or adults that work in a different location for part of the
day may be exposed to higher or lower contaminant levels. This
issue is further discussed under the relevant pathways in Chapters
4-12.
This methodology examines a range of exposures by varying the
number of years spent in the vicinity of the combustor. In
2-15
-------
Table 2-2
Geographic Mobility Rates as Related to Age*
Age (years)
Mobility Rate
1-4
5-19
20-29
30-44
45-64
>65
26%
17%
33%
17%
8%
5%
Source: U.S. Bureau of the Census, 1986.
2-16
-------
Scenario C, the individual is assumed to have spent an entire 70-
year lifetime in the vicinity of the combustor. In exposure
Scenario B, the individual is assumed to have lived 30 years in the
vicinity of the combustor based on the median length of residence
for homeowners. In Scenario A, the individual is assumed to have
lived 16 years near the combustor based on the median residence
time for owners and renters. For a site-specific assessment, local
data should be used, if available.
2.5. PATHWAYS OF HUMAN EXPOSURE
As explained earlier, an exposure pathway consists of exposed
environmental media and the routes by which individuals contact
these contaminated media. Figure 2-1 can serve as a guide to
choosing which pathways are relevant for a particular site-
specific exposure assessment. In pathways involving soil or water,
different routes of exposure are possible and the different ways
in which the media are used affect human exposure. Additionally,
for the food and surface water pathways, the location of these
media relative to the combustor should be defined since these could
be different from that of the corabustor.
2.5.1. Food Ingestion Pathways. Plants and animals near the
stationary combustion source may take up emitted contaminants from
the air and or deposited contaminants from the soil. Humans are
exposed to the contaminants via the food chain when these plants
and animals are produced as a food source.
2-17
-------
2.5.1.1. LOCATION OP FOOD PRODUCTION — Location of food
production relative to the combustion source is an important factor
in human exposure to emitted pollutants via the food chain. Some
of an individual's diet may come from food sources (e.g., gardens,
farmland, grazing cattle) within the vicinity (i.e., within a 50
km radius of the emission source), whereas other foods may be
imported from distant areas.
The location of the food production, i.e., distance from( the
source, can be varied to determine a range of potential exposure.
Assuming that all of a person's diet comes from produce grown at
the site of maximal deposition, whether that be a home garden or
farm, and assuming that all meat and dairy products come from
animals grazing at the maximum point of emission impact will
predict very high exposures. Other scenarios can be constructed
by assuming all foods or certain food groups (e.g., milk and dairy)
come from the areas of lesser fallout. Section 2.5.1.2 further
discusses these scenarios.
2.5.1.2. INPUT VARIABLES FOR EXPOSURE VIA FOOD
Determination of human intake of contaminants from food is based
on the types of foods consumed and the amount of food consumed per
day, the concentration of contaminant in the food and the
percentage of the diet contaminated (see Figure 2-1).
The types of foods consumed will affect exposure because
different plants and animal tissues will take up contaminants at
different rates. Therefore, the concentration of the contaminant
in the food is determined, in part, by the type of food and varies
2-18
-------
with the types of food in the diet. The types of foods consumed
vary with age, geographical region and sociocultural factors.
These issues as well as specific values for various types of foods
consumed are discussed in Chapter 5.
The amount of daily food consumption varies according to age,
sex, body weight, and geographic region and varies within these
categories as well. Values for consumption rates are available from
USDA food consumption surveys and are discussed in more detail in
Chapter 5. In this document, food consumption rates from the 1977-
1978 USDA Food Consumption Survey are used. More recent data
(1987-1988), when available, should be used. If, however, in a
site-specific assessment, there is more agricultural practice in
the area, the USDA 1966-1967 Food Consumption Survey may be more
appropriate, since the more recent rates represent mean values of
a more suburban lifestyle and may not accurately represent an
agrarian population.
The percentage of locally grown (i.e., within the 50 km radius
of the combustor facility) food consumed by the individual will
affect exposure; all of the dietary intake may not be
contaminated. The proportion of local food production will depend,
in part, on the degree of urbanization. Local food production can
refer to either a home garden or commercial farm. People living
in rural or suburban areas who can raise animals and gardens will
have a larger percentage of their food produced locally than people
living in a city. In this document, only food grown near the
combustor is assumed to be contaminated and only a proportion of
the individual's diet will be produced locally.
2-19
-------
2.5.2. Soil Exposure Pathways. As illustrated in Figure 2-1, if
there is soil exposed to deposited emissions in the area
surrounding the combustor, soil ingestion, dermal exposure to soil
and inhalation of resuspended dust are potential pathways of
exposure. Contaminant concentration in soil will vary with
distance from the source since deposition rate is related to
distance. It is assumed in this document, that soil at the
individual's location is contaminated.
2.5.2.1. USE OF THE SOIL — An individual could be exposed
to contaminants in soil by ingesting soil, absorbing the
contaminant in soil through the skin or inhaling resuspended dust.
Potential routes of exposure are determined by the way the soil is
used. Soil that is used for farming or recreation will be involved
in pathways for human exposure that are different from soil on
roadways or in urban areas. For urban populations, dermal contact
and dust resuspension may be more important pathways of exposure
than soil ingestion; whereas, for children playing outdoors in
suburban or rural areas, soil ingestion and dermal contact may be
the more significant routes of exposure.
2.5.2.2. INPUT VARIABLES FOR SOIL INGESTION — Children
and adults are directly exposed to contaminants in soil when they
consume soil that may be adhering to their hands. In addition,
young children may intentionally eat soil (i.e., a condition known
as pica). The factors that affect exposure from soil ingestion
include soil concentration, the rate of soil ingestion during time
2-20
-------
of exposure and the length of time spent in the vicinity of
contaminated soil. Soil ingestion rates in children are based on
studies that measure the quantities of nonabsorbable tracer
minerals in the feces of young children. Ingestion rates for
adults are based on assumptions about exposed surface area and
frequency of hand to mouth contact. Both indoor dust and outdoor
soil may contribute to total daily ingestion. Specific values for
ingestion rates are presented in Chapter 6. The amount of time the
individual spends in the vicinity of soil exposed to deposition of
emitted pollutants (both indoors and outdoors) also influences
exposure levels. The worst-case scenario for soil ingestion is
undoubtedly a child with pica in close proximity to the combustor.
An average scenario may be an adult with a small home garden.
2.5.2.3. INPUT VARIABLES FOR DERMAL EXPOSURE TO SOIL —
Humans will be exposed to contaminants by absorption through the
skin when they come in contact with contaminated soil. Surface
area, contact time, contact amount, amount of time spent near the
combustion source and fraction of contaminants absorbed through
skin are some of the factors that affect dermal exposure.
In general, the larger the surface area, the more contaminant
that can be absorbed through the skin. Surface area is affected
by age and body weight; children have less total surface area than
adults. The amount of surface area available for exposure to soil
is also affected by the amount of clothing worn. An adult working
in the garden in long sleeves and pants will have a smaller exposed
surface than an adult working in shorts and a short-sleeved shirt.
2-21
-------
For dermal exposure from soil, the exposed surface area dictates
the contact amount, the amount of soil that can adhere to exposed
skin.
The longer contaminated soil stays in contact with the skin,
the greater the amount of contaminant that can be absorbed.
Contact time refers to the duration of time each day that contact
with soil is possible. The amount of time each day spent in the
vicinity of the combustion source where soil is likely to be
exposed to emitted pollutants also affects dermal exposure. Both
indoor dust and outdoor soil may contribute to daily contact hours.
Seasonal exposure can also be considered since regional climate
will influence this variable.
The amount of contaminant that can be absorbed through the
skin (i.e., absorption fraction) depends on the chemical properties
of the contaminant, properties of the soil matrix, and dermal
pharmacokinetics. If a compound cannot readily be absorbed through
the skin, the daily intake of the compound may be small even if
other exposure characteristics, such as contact time, are
favorable. Specific values for these variables are discussed in
Chapter 7.
2.5.2.4. INPUT VARIABLES FOR EXPOSURE BY RESUSPENSION OF
DUST — Pollutants in the contaminated soil can be resuspended as
particulates in the air by wind erosion. The amount resuspended
is dependent on the moisture content of the soil, the fraction of
vegetation cover, the wind velocity, soil particle size, the
2-22
-------
pollutant concentration in the soil and the size of the
contaminated area. As a Consequence to resuspension of the dust,
persons may inhale the pollutant particles.
Methodologies have been developed to assess the exposure to
pollutants resuspended by wind erosion for landfills (U.S. EPA,
1985a) and Superfund sites (U.S. EPA, 1988b). Application of these
methodologies to the deposited combustor emissions indicates that
dust resuspension by wind erosion is not a significant pathway.
This is further discussed in Chapter 8.
2.5.3. Water Exposure Pathways. The water exposure pathways can
be used to determine the concentration of emitted contaminants in
groundwater, surface .water bodies, or collected precipitation, such
as in a cistern. Estimation of daily exposure of individuals using
these water sources for various purposes, such as fishing, swimming
and drinking water, is possible using various models. Possible
routes of human exposure to contaminants via water are shown in
Figure 2-1.
2.5.3.1. LOCATION — Contaminants emitted from the combustor
are subject to dissolution from precipitation events and then
either move over the ground into a surface water body or infiltrate
into the ground and recharge the groundwater. Since annual ground-
level of contaminants decreases with distance from the source (see
Section 2.3.1), the location of the precipitation collection
apparatus, surface water body on which emitted pollutants are
deposited, and the soil concentration (affects leachate
2-23
-------
concentration) are important determinants of the water
concentration. Additionally, the location and size of the
watershed affect the concentration of contaminants suspended in
runoff for the Surface Water Model.
2.5.3.2. USE OF THE WATER — The way in which the water is
used, whether it be collected precipitation or groundwater, or a
surface water body such as a lake, farm pond or city reservoir,
will affect possible exposure routes. Use of the surface water
body for recreation or as a drinking water source will introduce
dermal contact and water ingestion as possible exposure pathways.
Commercial and/or recreational fishing with subsequent use of the
fish and shellfish as a food source, make the food chain an
important route of exposure for communities with a surface water
body in the vicinity of a combustion source. Ingestion of the
water is an important route of exposure if the surface water body
is a reservoir; whereas, if the surface water body is used
recreationally for swimming and boating, exposure to pollutants by
dermal contact with the water is possible. The relevant exposure
pathways can be chosen in a site-specific assessment to construct
the scenario most suitable for the individual(s) of interest.
2.5.3.3. INPUT VARIABLES FOR EXPOSURE VIA SURFACE WATER —
A surface water body can be a lake, river, pond or reservoir, and
can be used for fishing, swimming and boating, or as a source of
drinking water.
2-24
-------
Exposure to the contamirumt if the surface water body is used
as a drinking water source is affected by the concentration of the
contaminant in the water, daily amount of water ingested and the
percentage of time the individual spends in the area serviced by
that water supply system* It is assumed in this methodology that
treatment processes for drinking water do not alter the deposited
contaminants. Chapter 10 further discusses determination of
exposure from water ingestion.
Factors that affect human exposure by ingestion of fish from
a contaminated surface water body include the sediment and water
contaminant concentration, the types of fish/shellfish consumed
and the ingestion rates for the various fish and shellfish groups,
the bioconcentration factor for these groups, and the percent of
the dietary fish that is caught in the surface water body near the
source. The types of fish consumed will affect exposure because
different types of fish and shellfish take up contaminants at
varying rates. For example, fatty fish will tend to accumulate
organic contaminants more readily than lean fish.
The amount of fish consumed also affects exposure as people
who eat large amounts of fish will tend to have higher exposures.
Fish consumption rates vary greatly depending on geographic region
and social/cultural factors. Because 100% of an individual's
dietary fish may not originate from the surface water body near the
combustion facility, the percentage of locally caught fish is also
a variable for exposure. Specific values for fish consumption
rates are discussed in Chapter 11.
2-25
-------
If the surface water body is used for recreational purposes
such as swimming and boating, dermal absorption of contaminated
water becomes another possible route for human exposure. The
surface area of exposed skin, the chemical concentration in the
water, the permeability of the skin to the chemical, and the length
of time the individual is in contact with the water are the
variables that contribute to dermal exposure. Incidental swallowing
of water when using the water body for recreation also contributes
to exposure. Dermal exposure via water is discussed in more detail
in Chapter 12.
The input variables for various pathways of exposure via
surface water can be changed to determine a range of exposures. An
average exposure scenario might be represented by an individual
that fishes, swims and obtains drinking water from the same water
source (e.g., a large river or lake), while a worst-case
possibility could be a person whose drinking water source is a
cistern holding collected precipitation (see Section 2.5.3.4) and
who fishes and swims in a small farm pond.
2.5.3.4. INPUT VARIABLES FOR PRECIPITATION FOR DRINKING
WATER — In several regions, rain and/or snow fall are collected
and stored as a source of drinking water. Individuals in these
areas may be exposed to pollutants deposited on the surface area
from which water is collected. The input variables for this
exposure pathway include: total annual deposition (both wet and dry
deposition), annual precipitation, and the daily amount of water
consumed. Section 9.3 further discusses this exposure pathway.
2-26
-------
2.5.3.5. INPUT VARIABLES FOR GROUNDWATER INFILTRATION
MODEL -"As a consequence of emitted pollutants infiltrating into
the groundwater, the groundwater may be a source of exposure. The
methodology developed to calculate risks from the groundwater
pathway was originally intended to evaluate impacts from the
landfilling of municipal sludge. Application of the methodology
to deposited combustor emissions, showed it not to be a significant
exposure pathway. This analysis is presented in Section 9.4.
2.6. DEFINING EXAMPLE EXPOSURE SCENARIOS
In this document, three example scenarios will be constructed
and used to illustrate calculation of exposures. Risk assessors
can vary the set of conditions to create exposure scenarios for a
site-specific assessment. A hypothetical municipal waste combustor
located in western Florida will be used as an example combustor
facility. Scenario "A" is constructed to represent the most likely
occurrence for an exposed population anywhere within a 50 km radius
of the facility; typical or average values of input variables are
chosen. In contrast, Scenario "C" illustrates the hypothetical
case wherein all variables are maximized so that the highest
potential exposure to an individual is determined. While most
persons in the vicinity of a combustion facility are not likely to
have such an exposure, this possibility may exist and should be
explored. Scenario "B" strikes an approximate midpoint between
Scenarios A and C; exposures will be higher than those in Scenario
A but not as but infrequent as those in Scenario C.
2-27
-------
For illustrative purposes in this document, certain
assumptions will be made about the various routes of exposure.
For example, for the food ingestion pathways, all milk will be
assumed to be produced within the 50 km radius of the combustor
facility. Depending on the exposure scenario, some milk may also
be produced at distances closer to the facility. For all three
hypothetical exposure scenarios, meat and food crops will be
assumed to come from outside the vicinity of the combustor, unless
produced by the individual. General descriptions of the three
scenarios are provided below and in Table 2-3. Further details and
example calculations for the scenarios are presented in each
exposure pathway chapter.
2.6.1. Exposure Scenario A. Scenario A represents the most likely
occurrence for an exposed population anywhere within a 50 km radius
of the facility. This scenario is represented by an adult living
in a metropolitan area (i.e., central city) where a MWC has been
operating continuously for 30 years. Ground level concentrations
of the emitted pollutants are relatively low, being represented as
the average concentration for a 50 km (radial distance) ring around
the source. The individual resides in this area for 16 years, but
works (8 hours/day) away from the area receiving fallout.
In this scenario, a majority of the food is produced outside
the vicinity with the exception of milk, which comes from a nearby
dairy also in the area of average deposition. The individual grows
some of his food in a small home garden, so that some of the diet
is contaminated by the deposited emissions,, The proportion
2-28
-------
TABLE 2-3
Description of Exposure Scenarios
Scenario A
Scenario B
Scenario C
Length Time of
Combustor
Emissions (years)
Location of
Individual (km)
Age
BW* (kg)
Years at
Location
Food Production
(meat and
vegetables)
Milk
Production
30
50
(central city)
Adult
70
16
Percent home
produced for
central city
dweller;
remainder from
outside 50 km
radius of
facility
100% within 50
km radius of
facility
Food
Consumption
Rate (percentile)
Dermal Exposure
from Soil
(percentile)
50
Average
60
(suburban)
Child then
adult
17 then 70
30
Percent home
produced for
suburban
dweller;
remainder from
outside 50 km
radius of
facility
Percent home
produced for
suburban
dweller (5 km);
remainder
within 50 km
of facility
70-75
Upper end of
average range
100
0.2
(non-
metropolitan)
Child then
adult
17 then 70
70
Percent home
produced for
non-
metropolitan
dweller;
remainder from
outside 50 km
radius of
facility
Percent home
produced for
non-
metropolitan
dweller (0.2
km) ; remainder
within 50 km
of facility
90-95
High-end of
range
2-29
-------
TABLE 2-3 (cont.)
Scenario A
Scenario B
Scenario C
Soil Ingestion « Average
Water Source
Fish Ingestion
Water Ingestion
Rate
Dermal Exposure
from Water
Lake within
50 km
« Average
consumption for
U.S. population
« Average U.S.
population
w Upper-end of
average range
Lake within
50 km
* 95th
percentile for
U.S. population
« 70-75
percentile for
U.S. population
Pica child;
adult high-
end of range
Residential
cistern
« 75th
percentile for
recreational
fisherman
« 90-95
percentile for
U.S. population
Not quantified Not quantified Not quantified
*BW of children may vary and is further discussed in Section 2.4.3.
2-30
-------
of the meat and vegetables home grown will be estimated using the
rates for center city dwellers as determined in the 1977-1978 USDA
Food Consumption Survey, with the remainder of these food groups
coming from outside the 50 km deposition radius, and therefore not
contaminated by emitted pollutants. Specific inputs for exposure
via food ingestion are discussed in Chapter 5. In general, the
50th percentile of consumption for all food groups will be used for
this scenario.
The dermal exposure and soil ingestion pathways are relevant
in this scenario since this person gardens. Average rates of soil
ingestion and dermal exposure are used in this scenario. Specific
inputs are detailed in the respective chapters.
A lake located in the vicinity provides drinking water for the
metropolitan area and is used for fishing and recreation. This
surface water body receives; runoff and pollutants directly
deposited from the MWC. While this person consumes average amounts
of fish (average for general U.S. population), none are caught from
the contaminated lake. Water ingestion rates for this individual
are equal to the U.S. population average. Specific inputs for
determining exposure via drinking and recreational usage of this
water body are described in Chapters 10, 11 and 12.
2.6.2. Exposure Scenario B. Scenario B strikes a midpoint between
Scenarios A and C and is represented by a child living in the
suburbs, 5 km from the combustion facility, where the areal average
ground-level concentrations of emissions (average for the 5 km ring
surrounding the source) have been deposited for the 60 years of
2-31
-------
operation of the MWC. The child grows up and remains in this area
for a total of 30 years. During childhood, the individual stays
at home during the day, but as an adult works 8 hours/day away from
the residence.
Some of this individual's food, i.e., milk, meats and
vegetables, will be assumed to be home-grown and will therefore
originate within the 5 km radius of the facility where the
individual resides. Rates for the suburban area, as delineated in
the 1977-1978 USDA Food Consumption Survey, will be used to
estimate the proportion home-grown. The remainder of the milk will
originate from within the 50 km radius of deposition, but the
remaining proportion of meat and vegetable in the diet will be
imported into the area. The upper end of the range of mean values
for food consumption rates (approximately 70-75th percentile) of
all food groups will be assumed.
Dermal exposure and soil ingestion pathways are also means of
exposure since this person gardens. The upper end of the average
rates, if available, of soil ingestion and dermal exposure will be
used in this scenario for both the childhood and adulthood portions
of this scenario. Specific inputs are detailed in the respective
chapters.
A surface water body (lake) located in the area of moderate
fallout is used as a drinking water source as well as for
recreational swimming and fishing. All of the drinking water and
some of the fish consumed by this individual are contaminated. It
is assumed that the same drinking water source serves the
individual's residence and place of work. The water ingestion rate
2-32
-------
for this individual is the 70-75th percentile for the U.S.
population. The fish ingestion rate for this individual is the
upper-end of the range for the U.S. population.
Specific input values used for this exposure scenario are
described in the appropriate pathways.
2.6.3. Exposure Scenario G. Scenario C illustrates a case in
which all variables are maximized so that the highest potential
exposure can be determined. In exposure Scenario C, the individual
is a child living in close proximity to the facility (0.2 km away)
in a non-metropolitan area after the combustion source has been
operating for 100 years. The child grows up and spends his
lifetime (70 years) in this area. It is assumed the individual's
daily activity is confined to the area of maximal deposition of
pollutants.
In this scenario, a proportion of the dairy products will be
come from the area of maximal deposition within the 0.2 km radius
of the facility. Most of the meat and vegetables consumed by this
individual will be home grown, with the remainder of the diet being
imported into the area. The proportion of the diet grown by this
individual will be estimated using rates for non-metropolitan area
as stated in USDA Food Consumption Survey (1977-1978). Food
consumption rates for this individual are assumed to be in the
upper 90-95th percentile for all food groups.
2-33
-------
As a youngster, this person has pica, and therefore, has high
ingestion and dermal contact with soil. As an adult, the
individual also has the high range of soil ingestion and dermal
contact.
The individual obtains drinking water from a residential
cistern which has collected precipitation. The water ingestion
rate for this individual is the 90-95th percentile for the U.S.
population. A pond, located near the residence, is the source of
all fish consumed by this person, and is also used for swimming.
It is assumed that this person consumes the mid-range values for
fish for recreational fishermen. Chapters for each pathway further
elaborate this scenario.
The remainder of this document presents a methodology for
determining human exposure by various pathways, and then approaches
for assessing if the exposure poses a risk to human health.
Example calculations for cadmium and benzo(a)pyrene are presented
for the hypothetical Scenario B.
2-34
-------
3. AIR DISPERSION AND DEPOSITION OP EMITTED POLLUTANTS
3.1. BACKGROUND AND PURPOSE
Combustion of materials produces residual amounts of pollution
that may be released to the environment. Estimation of potential
human health risks associated with these releases requires
knowledge of atmospheric pollutant concentrations and annual
deposition rates in the areas around the combustion facility.
Values for these quantities are most often estimated through the
use of atmospheric dispersion models. To achieve the most
realistic answers possible, these models should contain state-of-
the-art algorithms to account for the effects of complex terrain
and the process of pollutant deposition. In many instances,
however, models that have very basic means of accounting for these
two processes are used in risk assessments of municipal waste
combustors. For example, the ISCST model provides an adequate
treatment of flat terrain, but not for estimating the effects of
deposition (U.S. EPA, 1986b). Other risk assessments for
incinerators located in complex terrain (U.S. EPA, 1987b) have used
the ISCLT and LONGZ models. In order to determine how appropriate
the use of these simple models in situations where complex terrain
effects and pollutant deposition are considered, it was necessary
to compare results from these models with those generated by a
complex terrain, deposition model. No such model is currently
available through the U.S. EPA (1986c). Therefore, an appropriate
complex terrain atmospheric dispersion model had to be selected
that could be modified to realistically account for depositional
3-1
-------
effects. Two existing models, RTDM and COMPLEX I were altered to
meet this requirement. This chapter evaluates the impact that the
new models' treatment of complex terrain and depositional effects
has on values used in risk assessments, compares their results with
those from an existing model, and presents the areal-average
deposition used in the three exposure scenarios A, B, and C in this
document.
Deposition values for pollutants are the initial input for all
models to estimate human exposure and risk. The choice of the
appropriate model for air dispersion and deposition of emitted
pollutants can, therefore, have a significant impact on the
exposure scenario. The approaches shown here are illustrative.
The risk assessor should evaluate the terrain, meteorologic
conditions and other conditions for a site-specific assessment and
determine the most appropriate model for the site.
3.2. MODEL SELECTION
Selection of appropriate complex terrain atmospheric
dispersion models involved meeting three criteria: (1) account for
the effects of complex terrain; (2) simulate yearly time periods
(ideally five years or more); and (3) calculate both pollutant
concentration and atmospheric deposition. A number of existing
atmospheric models were investigated for possible use. These
models generally fell into two categories.
The first group includes Gaussian puff and numerical
simulation models. While many of these models contain algorithms
for estimating pollutant deposition as well as concentration, they
3-2
-------
are usually very complex and sophisticated. The chief advantage
of these models is their ability to calculate realistic wind
fields, which in theory enable them to more effectively account for
the effects of complex terrain. Further, many of these models are
formulated such that these wind fields allow pollutant
concentrations to be calculated at a specific time and location;
thus, they are often used for simulating exposures during emergency
situations. Generating wind fields, however, requires extensive
computing time; consequently, these models are usually only viable
for simulating time periods ranging from hours to a few days.
The second group contains less sophisticated models based on
the Gaussian plume algorithm. Instead of calculating detailed wind
fields, these models make only crude assumptions regarding complex
terrain effects and are therefore better suited for determining
average, time-independent concentrations at a desired location.
Since these models are also less computer intensive than the first
group of models, they may be used to simulate periods of a year or
more. Like the more complicated models of the first group, many
of these models also include provisions for estimating particle
3
deposition.
The U.S. EPA Guideline on Air Quality Models and accompanying
Supplement A (U.S. EPA, 1986c) contain recommendations for
selecting models for use in complex terrain. No models capable of
generating their own wind fields are recommended. Instead, three
categories of screening models, initial, level two, and level
three, are identified. The recommended level two and level three
models in the Guideline and Supplement A were COMPLEX I and RTDM,
3-3
-------
respectively. COMPLEX I is a Gaussian plume model that
incorporates a number of traditional algorithms for accounting for
complex terrain effects. RTDM is a somewhat more advanced model
that employs partial plume reflection algorithms, which prevent
overestimates of pollutant concentrations near sloping terrain.
Both of these models were selected for use in this analysis. Both
of these models are Gaussian plume and are therefore well-suited
for simulating time periods of one to five years. Algorithms used
in COMPLEX I and RTDM to estimate air concentrations of pollutants
are derived based on the assumption that no deposition occurs.
However, for this analysis, it was also necessary to calculate
pollutant deposition. Since concentration is dependent on
deposition (i.e., plume depletion), accurate representations of
both these quantities cannot be obtained unless the model used to
calculate them contains concentration algorithms that consider
depositional effects. Therefore, concentration algorithms
explicitly formulated to account for the effects of deposition
(U.S. EPA, 1982a) were substituted for the concentration algorithms
originally found in COMPLEX I and RTDM to form two new models,
COMPDEP and RTDMDEP. In addition, algorithms for estimating the
effects of building wakes were obtained from a fourth model, ISCST,
and used in the COMPDEP model. Details of each model are discussed
in the following sections.
3-4
-------
3.3. MODEL DESCRIPTIONS
3.3.1. General Model Development. Changes to existing models and
new features incorporated in the COMPDEP and RTDMDEP models are
given in these sections. For a more detailed description of
existing models not included in this report, readers are referred
to the user's guides for previously mentioned models: Turner
(1986), U.S. EPA (1979), U.S., EPA (1980a), U.S. EPA (1982a), and
ERT (1987).
3.3.2. COMPDEP Model. COMPDEP is a modification of COMPLEX I to
account for both wet and dry deposition. The capability to
estimate pollutant deposition was added to COMPLEX I by the
inclusion of concentration algorithms from U.S. EPA (1982a). The
algorithms from U.S. EPA (1982a) estimate pollutant deposition
based on a single deposition velocity that is independent of
particle size or atmospheric conditions. Therefore, to provide a
more realistic value, these algorithms were altered to allow
calculation of deposition velocity based on particle .size and
atmospheric conditions (GARB, 1986). Additional routines to enable
the model to estimate pollutant concentration and deposition during
periods of precipitation were also included. COMPDEP uses hourly
meteorologic data as described in Section 3.4.2 to estimate
atmospheric concentrations and pollutant deposition rates at
desired receptor locations.
3-5
-------
3.3.2.1. CONCENTRATION ALGORITHMS — Hourly pollutant
concentrations were calculated using algorithms taken from the
MPTER-DS model. The equation used to calculate concentrations
during time periods of no precipitation is given by:
= FDRY.R.Q • [exp -h'(y/0y)2] • VDF
(Equation 3-1)
where:
Cd,
FDRY
Q
U
y
VDF
Pollutant concentration associated with particles of
the ith size category during the portion of the
during the portion of the hour in which no
precipitation occurs (g/m3)
Correction factor for the fraction of the hour in
which no precipitation occurs. Equal in magnitude to
1.0 - FWET
Fraction of pollutant attached to particles in the ith
size category
Source emission (g/s)
Mean wind speed at plume height and receptor location
(m/s)
Horizontal crosswind distance from plume centerline to
receptor (m)
Gaussian plume dispersion parameter for the horizontal
direction (m)
Vertical distribution function given by one of the
following equations:
For stable conditions or unlimited mixing:
VDF - 1 • [exp{-2W«(z - H)-x - x^
"
[exp{-(z - H)2} + exp{-(z + H)2}
(Equation 3-2)
For unstable or neutral conditions with CTZ < 1.6L
VDF =__!_ . S [exp{-2W.(z - HO-x - xt2} •
[exp{-(£ - Hj)2} + exp{-(z
{1 - 475f«V1.x.exp(^12)'
3-6
(Equation 3-3)
-------
For stable or neutral conditions with uniform vertical mixing
az > 1.6L and V2 = 0 = 0
VDF = !«[(! + 2x,2)-erfc(x1) - (2xv
L
For V2 i* 0.0
VDF = i_ . |V1/V2).exp(4V(,.V2.x2)»(2V,.x)
L
where:
(Equation 3-4)
(Equation 3-5)
H -
H, * HE + 2N-L
H, = H,/,/2a.
x =
x, = x«W
z = 2jj2at
W = W/U
V, «
V, = Vd - W/2
V2 » Vd - W
f = z + H + 2V,-J
f, = z + H, + 2Vt.J
(Equation 3-6)
(Equation 3-7)
(Equation 3-8)
(Equation 3-9)
(Equation 3-10)
(Equation 3-11)
(Equation 3-12)
(Equation 3-13)
(Equation 3-14)
(Equation 3-15)
(Equation 3-16)
(Equation 3-17)
3-7
-------
and
HE
N
L
X
z
W
V.,
» Effective height of source emission (m)
= Eddy reflection number
= Mixing depth of inversion lid (m)
= Horizontal upwind distance of source from receptor (m)
= Vertical coordinate of receptor (m)
= Gravitational settling velocity (m/s)
= Dry deposition velocity (m/s)
Effects of dry deposition and gravitational settling are
accounted for in the vertical distribution function through the use
of deposition and gravitational settling velocities. For the
fraction of the hour during which there is no precipitation,
concentrations are calculated using deposition and settling
velocities described in Section 3.3.2.2.
Concentrations are also calculated for periods in which
precipitation occurs. For the fraction of the hour during which
precipitation takes place (identified according to methods
described in Section 3.4.2.2), concentrations are calculated with
deposition and gravitational velocities set equal to zero. This
is based on the assumption that the effects of dry deposition and
gravitational settling are negligible compared with precipitation
scavenging. Effects of precipitation scavenging during these
periods are approximated by the addition of an exponential loss
term to the equation used to calculate pollutant concentrations:
Cw, - FWET.F..Q . [exp{-3s(y/ay)2}] . VDF . exp(-A.*/U) (Equation 3-18)
JffK \J*0y
3-8
-------
where:
Cw- = Pollutant concentration in the portion of the hour in
1 which precipitation does occur associated with particles
in the ith particle size category (g/m )
A = Scavenging coefficient (1/s)
The model uses a different scavenging coefficient for each particle
size and precipitation intensity.
The total hourly concentration of pollutant is the sum of the
dry and wet pollutant concentrations (Cd, and Cw,., respectively)
over all particle size categories.
3.3.2.2. POLLUTANT DEPOSITION -- Both dry and wet deposition
are calculated by the model. During periods in which no
precipitation occurs, dry deposition of pollutant associated with
each particle size is calculated as the product of concentration
(as determined by Equation 3-1) and a particle size dependent
deposition velocity. This velocity is automatically calculated by
the model from routines developed by the California Air Resources
Board (GARB, 1986). These algorithms were selected based on
results from an evaluation of four deposition velocity models
(Travis and Yambert, 1989), which showed them to be superior in
estimating deposition velocities of particulates to grasses and
small crops, the type of surfaces in risk assessments for which
deposition estimates are often desired. Total dry deposition for
the period, DEPd^ (g/m2/s) is obtained by summing the depositions
for each different particle size.
Using methods given in Slinn (1984), wet deposition is
estimated by integrating the product of Equation 3-17 and the
3-9
-------
scavenging coefficient, A, over the range of 0.0 < z < oo. The
resulting equation used to calculate wet deposition for particles
in each size category is:
DEPwi =, AjWET-.Fi.Q . expB(y/ay)2] . exp^A-x/U) (Equation 3-19)
where:
DEPWj = Hourly wet deposition value for associated particles in
the ith size category (g/m/s)
Total wet deposition for each hour is the sum of the values for
each particle size.
3.3.2.3. TERRAIN REPRESENTATION — COMPLEX I contains five
different methods for estimating the effects of complex terrain.
The first of these algorithms was retained for use in the COMPDEP
model (U.S. EPA, 1986b). Effects of complex terrain are accounted
for in the following manner. First, the effective plume height is
equivalent to the maximum of the following calculated values:
where:
HE = H - THT-(1.0 - TER)
HE = H-TER
HE = HMIN
(Equation 3-20)
(Equation 3-21)
(Equation 3-22)
H
THT
TER
HMIN
Plume height (sum of stack height and plume rise) (m)
The height of the terrain obstacle above stack base
elevation (m)
Terrain adjustment factor ranging in value from 0,0 to
1.0
Minimum distance between plume and ground (m) (usuallv
set to 10.0 m) *
3-10
-------
For receptors whose ground level elevation exceeds the plume
elevation (the sum of stack, base elevation and plume height) ,
pollutant concentrations and depositions are multiplied by the
following correction factor:
CORK = (400.0 - DIFF)/400.0 0.0 < DIFF < 400.0 m
(Equation 3-23)
= 0.0
DIFF > 400.0 m
(Equation 3-24)
where:
DIFF
The difference between receptor ground level elevation
and plume elevation (m)
3.3.2.4. BUILDING WAKE EFFECTS — COMPLEX I does not contain
a methodology for estimating building wake effects. To provide
the model with this capability, algorithms from the ISCST model
were used.
3.3.3. RTDMDEP. RTDMDEP is a modification of RTDM to account for
both wet and dry deposition. RTDM is recognized as a third-level
screening model for use in complex terrain. Like COMPLEX I, it was
modified to account for the effects of pollutant deposition by the
inclusion of concentration algorithms described in U.S. EPA
(1982a). The modifications performed are identical to those
described in Sections 3.3.2.1 and 3.3.2.2 for the COMPDEP model.
The new model, RTDMDEP, calculates pollutant concentrations and
depositions during periods of precipitation and no precipitation
using deposition velocities that are particle size dependent (CARB,
1986). With the exception of these modifications, RTDMDEP is very
3-11
-------
similar in form to the original RTDM model. The chief difference
between it and COMPDEP is the way in which the model accounts for
the effects of complex terrain. This difference is discussed
briefly in the next section.
3.3.3.1. TERRAIN REPRESENTATION — Like COMPDEP, RTDMDEP
modifies the effective plume height to account for the effects of
terrain; however, it refines this approach with the additional
consideration of critical height defined as:
where:
Hcrit
Fr
Hcrit = THT • (l.o - Fr)
(Equation 3-25)
Critical plume height defined such that only air above
this level will pass over a terrain feature (m)
Froude number defined as:
Fr - u/(THT«7(g- VPTG/Ta))
(Equation 3-26)
where:
u » Stack top wind speed (m/s)
g = The acceleration of gravity (9.8 m/s2)
VPTG = Vertical potential temperature gradient (K/m)
Ta = Ambient temperature (K)
The effective plume height is calculated according to:
a
HE
HE - TER-Hpc
Hpc - (1.0 - TER)'Htc
H < THT
H > THT
(Equation 3-27)
(Equation 3-28)
where:
Hpc = The height of the plume above the critical height (m)
Htc = The height of the local terrain above the critical
height (m)
3-12
-------
Additionally, to avoid artificial plume depression in cases
where the plume is above the critical height, RTDMDEP assumes a
level plume by setting; the value for TER equal to zero. RTDMDEP
also adjusts the mixing height in an analogous manner to that used
in calculating effective plume height.
RTDMDEP .contains an algorithm used to modify pollutant
concentrations to account for the effects of partial plume
reflection from sloping terrain. Conventional Gaussian plume
models model the ground surface through the use of "image source"
routines. These routines can result in unrealistically high
estimated pollutant concentrations for cases in which a plume
approaches rapidly sloping terrain. The partial plume reflection
algorithm used in RTDMDEP prevents these inaccuracies by
multiplying pollutant concentrations by a reflection factor, R,
based on terrain slope and plume growth. This factor is the ratio
of the minimum value of the maximum cross wind integrated
concentration (MCWI) to the cross wind integrated; concentration
evaluated at the point of impact of closest plume approach to
terrain.
3.3.4. ISC8T. The Industrial Source Complex Short Term Model
(ISCST) is a preferred model for use in flat or moderately rolling
terrain when modeling sources such as stacks from MWCs or other
industrial source complexes (U.S. EPA, 1986c). It is a steady
state Gaussian plume model that contains limited provisions for
estimating the effects of terrain on pollutant transport. ISCST
is sometimes used to estimate pollutant transport in complex
3-13
-------
terrain. However, concentrations estimated for points above stack
top elevation are subject to extreme uncertainty as the model
substitutes stack top elevation for the point's actual elevation.
Pollutant deposition is calculated through the use of surface
reflection coefficients and gravitational settling velocities.
3.4. ANALYSIS PROCEDURES
The modeling analysis conducted was designed with two primary
goals. The first of these was a quantification of how atmospheric
pollutant concentrations predicted by the two models detailed in
this paper differ from those calculated by existing flat terrain
models for sources located in complex terrain. Second, an
evaluation of the deposition velocity based algorithms used by
COMPDEP and RTDMDEP was desired. In order to accomplish this,
model runs were made for a hypothetical, complex terrain
incinerator using COMPDEP, RTDMDEP, and the existing flat terrain
model, ISCST. ISCST was chosen for the analysis since it is
probably the most widely used and accepted flat terrain model for
estimating pollutant concentrations due to stack emissions.
Whereas terrain elevation is the chief aspect of complex
terrain that affects predictions made by COMPDEP, concentration
predictions made by RTDMDEP are affected by terrain slope and hill
height as well. Because of this difference, two receptor patterns
were considered. Concentrations predicted by COMPDEP and ISCST
were calculated for a variety of terrain elevations, with no
consideration given to terrain slope. RTDMDEP model runs analyzed
3-14
-------
selected COMPDEP and ISCST receptor locations but incorporated
terrain slope and hill height into model calculations in addition
to terrain elevation.
Each model was run in two modes. The first of these
determined atmospheric pollutant concentrations assuming no
deposition occurred, while the second calculated both atmospheric
pollutant concentrations and pollutant deposition rate. Results
from these analyses are presented in Sections 3.5.0 - 3.5.3.2.
Model input is described in the following sections.
3.4.1. Source Characteristics. The source parameters for a
hypothetical incinerator in the Northeastern United States used for
comparison of the models study, given in Table 3-1, are taken from
an existing mass burn incinerator. Unit emission rates of
pollutant were used for all model runs. For model runs that
considered pollutant deposition, emitted pollutants were assumed
to fall into one of three particle size categories: (1) less than
2 Mm, (2) 2 to 10 jura, and (3) greater than 10 Mm. Representative
sizes and fraction of emissions in each of these categories were
based on data for organic pollutants found in U.S. EPA (1986b) and
are, respectively, 1.0 /m and 0.875 for category 1, 6.78 Mm and
0.095 for category 2, and 20.0 /Ltm and 0.030 for category 3.
3.4.2. Meteorologic Data. In order to realistically consider
possibly occurring meteorologic conditions, model runs were made
using one year of actual meteorologic data for Albany, New York.
The Source Receptor Analysis Branch of the U.S. EPA at Research
3-15
-------
TABLE 3-1
Source Characteristics for MWC Analysis
Parameter
Val
Stack diameter (m)
Stack exit velocity (m/s)
Stack gas temperature (K)
Building Dimensions (m)
Height
Width
Length
1.04
15.24
327.60
11.00
48.80
73.20
3-16
-------
Triangle Park, NC provided this data in standard, preprocessed
format containing hourly values for mixing height, stability class,
temperature, wind direction, and wind speed.
3.4.3. Receptor Locations. Receptor locations used by ISCST and
COMPDEP model runs were chosen to reflect how model predictions are
influenced by terrain both near to and far away from the source.
Whereas terrain elevation is the chief aspect of complex terrain,
which affects predictions made by COMPDEP and ISCST, predictions
made by RTDMDEP are also influenced by hill height. Because of
this difference, two receptor patterns were considered.
Predictions made by COMPDEP and ISCST were calculated for a variety
of terrain elevations, with no consideration given to terrain
slope. Receptor distances for these model runs varied from 200 to
50,000 meters, while elevations ranging from that of the stack base
to 300 meters above the top of the stack were considered. The
most frequent wind direction for all stability classes during 1970
was due north (U.S. EPA, 1987b). Thus, to ensure exposure to the
widest possible range of meteorologic conditions, all receptors
were located along this direction.
Selected receptor locations analyzed by COMPDEP and ISCST were
also analyzed by RTDMDEP. In order to investigate the influence
of terrain slope on model predictions, three sets of RTDMDEP runs
were made assuming terrain slopes of 30.0°, 45.0°, and 60.0°
respectively.
The coordinate systems used to express receptor locations are
defined as follows. Coordinate systems for both locations are
3-17
-------
rectangular with the "+x" direction corresponding to east, and the
»+y" direction corresponding to north. Origins for all systems
correspond to the stack location. A complete listing of receptor
locations used by ISCST and RTDMDEP is given in Table 3-2. Receptor
locations used by RTDMDEP for each of the three terrain slopes
considered are found in Table 3-3.
3.4.4. Model Run Descriptions. For each model, a concentration-
only run, and a deposition run were made. For all deposition runs,
only dry deposition was calculated. In its original form, ISCST
does not determine wet deposition. Consequently, a comparison of
wet depositions between flat and complex terrain models could not
be attained without its modification. Hence, for purposes of this
analysis, wet deposition was not determined.
The following were assumed by all model runs: exponents for
power-law wind increase with height of 0.07, 0.07, 0.10, 0.15,
0.35, 0.55, anemometer height of 10.0 meters, surface roughness of
0.10 meters, no pollutant decay. Model options specific to each
of the three models used in the analysis are presented in Tables
3-4 to 3-6.
3.4.5. Reflection Coefficients Used by ISCST. The algorithms used
by COMPDEP and RTDMDEP to estimate pollutant deposition require an
estimate of deposition velocity for each particle size analyzed.
This is done automatically via the inclusion of the CARB (1986)
algorithms. Unlike these models, ISCST requires that a reflection
coefficient and gravitational settling velocity be input manually
3-18
-------
TABLE 3-2
Receptor Locations Used by COMPDEP and ISCST
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
X Coordinate
(m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Y Coordinate
(m)
200.0
400.0
500.0
600.0
800.0
1000.0
1250.0
1500.0
1750.0
2000.0
2500.0
3000.0
4000.0
5000.0
10000.0
20000.0
30000.0
40000.0
50000.0
200.0
400.0
500.0
600.0
800.0
1000.0
1250.0
1500.0
1750.0
2000.0
2500.0
3000.0
4000.0
5000.0
10000.0
20000.0
30000.0
40000.0
50000.0
200.0
400.0
500.0
600.0
800.0
1000.0
1250.0
1500.0
1750.0
Elevation
(m)
169.0
169.0
169.0
169.0
169.0
169.0
169.0
169 . 0
169.0
169 o 0
169.0
169.0
169.0
169.0
169.0
169.0
169.0
169.0
169.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
194.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
3-19
-------
TABLE 3-2 (cont.)
Receptor
Number
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
X Coordinate
(m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Y Coordinate
(m)
2000.0
2500.0
3000.0
4000.0
5000.0
10000.0
20000.0
30000.0
40000.0
50000.0
200.0
400.0
500.0
600.0
800.0
1000.0
1250.0
1500.0
1750.0
2000.0
2500.0
3000.0
4000.0
5000.0
10000.0
20000.0
30000.0
40000.0
50000.0
200.0
400.0
500.0
600.0
800.0
1000.0
1250.0
1500.0
1750.0
2000.0
2500.0
3000.0
4000.0
5000.0
10000.0
20000.0
30000.0
40000.0
50000.0
Elevation
(m)
220.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
220.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
270.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
470.0
3-20
-------
TABLE 3-3
Receptor Locations Used by RTDMDEP
for Terrain Slope of 30, 45, and 60'
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
X Coordinate
(m)
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
Y Coordinate
(m)
500.0
500.0
500.0
500.0
1000.0
1000.0
1000.0
1000.0
5000.0
5000.0
5000.0
5000.0
Elevation
(m)
194.0
220.0
270.0
470.0
194.0
220.0
270.0
470.0
194.0
220.0
270.0
470.0
3-21
-------
TABLE 3-4
COMPDEP Model Options
Building Wake Effects
Terrain Effects
Gradual Plume Rise
Buoyancy Induced Dispersion
Building Wake Effects
No Calm Processing
No Stack Downwash
3-22
-------
TABLE 3-5
RTDMDEP Model Options
Terrain Effects
Gradual Plume Rise
Buoyancy Induced Dispersion
No Stack Downwash
Pasquill-Gifford Dispersion Coefficients
Partial Plume Penetration (VPTG 0.0060 Deg K/m)
Unlimited Mixing for Stable Conditions
Plume Path Coefficient for Stability Classes 1-6: 0.50
Default Vertical Potential Temperature Gradients for Stability
Classes 5 & 6
Stability Class-Dependent oy and az
No Wind Direction Shear
Partial Plume Reflection v
Off Centerline Horizontal Distribution Function for
All Stabilities
Constant Emission Rate
3-23
-------
TABLE 3-6
ISCST Model Options
Terrain Effects
Rural Option
Gradual Plume Rise
Buoyancy Induced Dispersion
Building Wake Effects
No Stack Downwash
No Calm Processing
Default Vertical Potential Temperature Gradients
3-24
-------
for each particle size. Values of these quantities used for all
model runs were taken from U.S. EPA (1986b) and are listed in Table
3-7.
3.5. COMPARISON OF MODEL RESULTS
3.5.1. Calculated Air Concentrations Neglecting Deposition
Effects.
3.5.1.1. COMPDEP AND ISCST RESULTS — An examination of Table
3-8 results shows that for receptors 1-19, which correspond to
flat terrain, COMPDEP and ISCST predict essentially equivalent
atmospheric pollutant concentrations. This agrees with anticipated
results since the concentration algorithms used by both models are
similar for conditions in which the effects of terrain and
pollutant deposition are neglected.
Once terrain elevations are introduced, differences in model
predictions increase. For receptors with elevations of half the
stack height, receptors 20 - 38, ISCST predictions exceed those of
COMPDEP by a factor of 1.6 on an average basis and individually by
\.
as much as a factor of 5.2 (at receptor 1) at distances less than
or equal to 10 km from the siource. At distances greater than 10
km, both models again predict nearly identical concentrations.
A similar pattern is seen for receptors that are equal to the
stack height in elevation (receptors 39 - 57). ISCST predictions
for receptors within 10 km of the stack exceed those of COMPDEP by
an average factor of 1.7. The maximum deviation between the two
3-25
-------
TABLE 3-7
Reflection Coefficients and Gravitational Settling Velocities
Used in ISCST Model Runs*
Particle
Diameter
Reflection
Coefficient
Settling
Velocity (cm/s)
D £ 2.0
2.0 < D < 10.0
D £ 10.0
0.961
0.861
0.714
5.55E-03
2.55E-01
2.22E+00
*Source: U.S. EPA, 1986b
3-26
-------
TABLE 3-8
Average Atmospheric Concentrations (/itg/m3) , Neglecting
Deposition Effects, Predicted by COMPDEP and ISCST
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
COMPDEP
Concentration
0.848E-02
0.112E+00
0.165E+00
0.212E+00
0.295E+00
0.350E+00
0.355E+00
0.342E+00
0.322E+00
0.300E+00
0.258E+00
0.223E+00
0.171E+00
0.137E+00
0.640E-01
0.284E-01
0.174E-01
0.124E-01
0.949E-02
0.404E-01
0.271E+00
0.381E+00
0.474E+00
0.589E+00
0.622E+00
0.581E+00
0.532E+00
0.484E+00
0.441E+00
0.367E+00
0.310E+00
0.231E+00
0.181E+00
0.795E-01
0.332E-01
0.199E-01
0.139E-01
0.106E-01
0.117E+01
0.132E+01
0.148E+01
0.156E+01
0.153E+01
0.139E+01
0.118E+01
0.101E+01
0.870E+00
ISCST
Concentration
0.849E-02
0.112E+00
0.165E+00
0.212E+00
0.296E+00
0.350E+00
0.355E+00
0.342E+00
0.322E+00
0.300E+00
0.258E+00
0.223E+00
0.171E+00
0.137E+00
0.640E-01
0.284E-01
0.174E-01
0.124E-01
0.949E-02
0.212E+00
0.829E+00
0.101E+01
0.109E+01
0.107E+01
0.965E+00
0.814E+00
0.694E+00
0.602E+00
0.528E+00
0.419E+00
0.343E+00
0.248E+00
0.191E+00
0.809E-01
0.334E-01
0.199E-01
0.139E-01
0.106E-01
0.128E-4-02
0.650E+01
0.500E+01
0.401E+01
0.282E+01
0.212E+01
0.161E+01
0.129E+01
0.106E+01
3-27
-------
TABLE 3-8 (cont.)
Receptor
Number
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
COMPDEP
Concentration
0.757E+00
0.589E4-00
0.473E+00
0.330E+00
0.247E+00
0.966E-01
0.377E-01
0.219E-01
0.152E-01
0.114E-01
0.851E+01
0.569E+01
0.489E+01
0.422E+01
0.320E+01
0.248E+01
0.189E+01
0.149E+01
0.121E+01
0.100E+01
0.733E+00
0.564E+00
0.374E+00
0.271E+00
0.992E-01
0.374E-01
0.216E-01
0.148E-01
0.111E-01
0.403E+01
0.270E+01
0.231E+01
0.199E+01
0.151E+01
0.117E+01
0.896E+00
0.702E+00
0.570E+00
0.473E+00
0.346E+00
0.266E+00
0.176E+00
0.128E+00
0.468E-01
0.177E-01
0.102E-01
0.702E-02
0.527E-02
ISCST
Concentration
0.890E+00
0.663E+00
0.518E+00
0.350E+00
0.259E+00
0.981E-01
0.379E-01
0.220E-01
0.152E-01
0.114E-01
0.128E+02
0.650E+01
0.500E+01
0.401E+01
0.282E+01
0.212E+01
0.161E+01
0.129E+01
0.106E+01
0.890E+00
0.663E+00
0.518E+00
0.350E+00
0.258E+00
0.981E-01
0.379E-01
0..220E-01
0.152E-01
0.114E-01
0.128E+02
0.650E+01
0. 500E+01
0.401E+01
0.282E+01
0.212E+01
0.161E+01
0.129E+01
0. 106E+01
0.890E+00
0.663E+00
0.518E+00
0.350E+00
0.258E+00
0.981E-01
0.379E-01
0.220E-01
0.152E-01
0.114E-01
3-28
-------
models' predictions occurs 200 meters from the source where the
ISCST concentration is a factor of 10.9 times that of COMPDEP. As
with the previous two groups of receptors, for distant points
(those in excess of 10 km from the source) concentration
predictions are essentially identical. In summary, for the
elevations considered so far, it is apparent that while average
concentrations predicted by ISCST and COMPDEP agree rather well
even in complex terrain, individual differences between ISCST and
COMPDEP differ by as much as a factor of ten.
An examination of the remaining two groups of receptors
considered, those that are 100 and 300 meters above stack height,
yields results that differ somewhat from those obtained for the
first three groups investigated. While concentrations predicted
for receptors 100 meters above stack height and less than 600
meters from the source are 1.0 to 1.5 times higher than COMPDEP1s,
ISCST predictions are less than COMPDEP's, by an average factor of
0.9, between 600 and 20,000 meters from the stack. This
discrepancy may be explained by the fact that, for terrain above
the stack, ISCST substitutes the stack height for the actual
terrain elevation in concentration calculations. Conseguently,
this approach yields almost random values, which should be
considered highly uncertain. Concentrations predicted by COMPDEP
are greater at this elevation than at stack height. This indicates
that the terrain correction factor described in Section 3.3.2.3 has
little effect at this elevation as the plume height (the sum of
3-29
-------
stack height and plume rise) is still above terrain elevation.
ISCST concentrations beyond 20 km are again equal to those of
COMPDEP.
In contrast to results for receptors 100 meters above stack
elevation, concentrations predicted by ISCST at elevations 300
meters above stack top are greater on average than those of COMPDEP
by a factor of 2.1 for all locations. The maximum deviation occurs
at receptor 77 where ISCST overpredicts COMPDEP by a factor of 3.2.
While concentrations predicted by ISCST are identical to those for
the previous two groups, those calculated by COMPDEP are
significantly less than at the previous elevation. Thus, for this
elevation, COMPDEP's terrain correction factor has a considerable
influence on model predictions.
3.5.1.2. RTDMDEP Results — An examination of concentrations
predicted by the third model used in the analysis, RTDMDEP,
provides additional interesting results (See Table 3.9). As stated
in Section 3.3.1, RTDMDEP employs a partial plume penetration
algorithm that, for elevated terrain, tends to prevent erroneous
magnification of pollutant concentrations. It is thus expected
that concentrations predicted by RTDMDEP should be less than those
of COMPDEP in areas of elevated terrain. For the most part, this
is indeed the case. Excluding receptor 2, where processes that
occur close to the source such as building wake effects, tend to
dominate pollutant concentrations, values predicted by COMPDEP for
receptors equal to and above stack height exceed those of RTDMDEP
for all terrain slope values. For these points, COMPDEP,
3-30
-------
TABLE 3-9
Average Atmospheric Concentrations (Mg/m3) , Neglecting
Deposition Effects, Predicted by RTDMDEP, COMPDEP and ISCST
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
RTDMDEP
30" Slope
0.395E+00
0.197E+01
0.242E+01
0.179E+01
0.662E+00
0.132E+01
0.143E+01
0.115E+01
0.176E+00
0.175E+00
0.173E+00
0.107E+00
RTDMDEP
45° Slope
0.395E+00
0.197E+01
0.242E+01
0.179E+01
0.662E+00
0.132E+01
0.143E+01
0.104E+01
0.176E+00
0.178E+00
0.173E+00
0.108E4-00
RTDMDEP
60° Slope
0.395E+00
0.220E+01
0.242E+01
0.174E+01
0.662E+00
0.133E+01
0.143E+01
0.990E+00
0.176E+00
0.182E+00
0.174E+00
0.109E+00
COMPDEP
0.381E+00
0.148E+01
0.489E+01
0.231E+01
0.622E+00
0.139E+01
0.248E+01
0.117E+01
0.181E+00
0.247E+00
0.271E+00
0.128E+00
ISCST
O.IOIE+OI
Oo500E+01
0.500E+01
0.500E+01
0.965E+00
0.212E+01
0.212E+01
0.212E+01
0.191E+00
0.259E+00
0.258E+00
0.258E+00
3-31
-------
predictions exceed those of RTDMDEP by a factor of 1.1 at receptor
6 to a value of 2.0 at receptor 3. These results are consistent
with previous evaluations of COMPLEX I and RTDM (Paine and Egan,
1986) which show that both models have a tendency to overestimate
pollutant concentrations in complex terrain, and that the
atmospheric concentrations predicted by RTDM are less than those
determined by COMPLEX I.
For receptors 1 and 5, which are lower than stack height and
near the source, the COMPDEP concentration are less than that of
RTDMDEP. This is most likely due to the fact that the partial
plume reflection algorithm of RTDMDEP does not affect terrain below
stack elevation.
Of further interest is the fact that for all receptors above
stack height, differences in values predicted by the two models
decrease with increasing receptor elevation. This, and the fact
that the algorithms RTDMDEP uses to compute terrain effects are
much more sophisticated than those in COMPDEP, suggests that the
simple terrain correction factor (Section 3.3.1) used by COMPDEP
works reasonably well for the range of elevations considered.
Variation of terrain slope seems to have a minimal effect on
concentrations predicted by RTDMDEP for the conditions modeled.
The greatest differences are seen at locations of large downwind
distances and high elevations (Receptors 8, 10, 11, and 12,). The
highest change occurs at Receptor 8 where a 13.9% decrease in
concentration occurs as terrain slope varies from 30° to 60°. It
is possible that additional variation in hill height used in model
calculations could magnify the differences seen.
3-32
-------
For all locations, concentrations predicted by ISCST are
greater than those of RTDMDEP. ISCST predicts concentrations that
range from 1.1 to 2.8 times those of RTDMDEP.
3.5.2. Calculated Air Concentrations Including Deposition
Effects. ISCST is designed to provide calculated values for either
atmospheric concentrations or pollutant deposition rate. However,
both COMPDEP and RTDMDEP detearmine both quantities when the option
to calculate pollutant deposition is selected. Tables 3-10 and 3-
11 list atmospheric concentrations calculated by these two models
when they are run in deposition mode. It can be seen that these
concentrations are almost the same as those obtained when
deposition effects are not considered. Only at distances greater
than 5 km do the COMPDEP results differ significantly than those
in Table 3-8, while RTDMDEP results are nearly identical for all
locations. This is understandable when one considers that, over
a period of time, deposition results in a depleted plume. Plume
depletion is reflected by an initial decrease in pollutant
concentration at the top of the plume that gradually extends
downward as plume travel time increases. Plume travel time to
receptors near the source pollutant is relatively short. Thus,
while air concentrations near the top of the plume may be somewhat
reduced, concentrations at ground level receptors, like those used
in the analysis, are similar to when no deposition effects are
considered. As seen by the results of this analysis, when plume
3-33
-------
TABLE 3-10
Average Atmospheric Concentrations (/ig/m3) Including
Dry Deposition Effects Predicted by COMPDEP
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
COMPDEP
Concentration
0.851E-02
0.112E+00
0.165E+00
0.212E+00
0.295E+00
0.349E+00
0.353E+00
0.340E+00
0.320E+00
0.298E+00
0.255E+00
0.220E+00
0.169E+00
0.134E+00
0.615E-01
0.262E-01
0.156E-01
0.108E-01
0.882E-02
0.405E-01
0.271E+00
0.381E+00
0.474E+00
0.589E+00
0.620E+00
0.579E+00
0.529E+00
0.481E+00
0.438E+00
0.363E+00
0.306E+00
0.227E+00
0.177E+00
0.751E-01
0.300E-01
0.174E-01
0.118E-01
0.874E-02
0.117E+01
0.132E+01
0.148E+01
0.156E+01
0.153E+01
0.139E+01
0.117E+01
0.100E+01
0.859E+00
3-34
-------
TABLE 3-10
Receptor
Number
48
49
52
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
(cont.)
COMPDEP
Concentration
0.745E+00
0.576E+00
0.460E+00
0.317E+00
0.234E+00
0.888E-01
0.330E-01
0.186E-01
0.124E-01
0.908E-02
0.847E-02
0.564E+01
0.483E+01
0.416E+01
0.314E+01
0.243E+01
0.184E+01
0.144E+01
0.117E+01
0.965E+00
0.701E+00
0.536E+00
0.351E+00
0.252E+00
0.888E-01
0.318E-01
0.176E-01
0.117E-01
0.850E-02
0.401E+01
0.267E+01
0.228E+01
0.197E+01
0.148E+01
0.115E+01
0.867E+00
0.681E+00
0.550E+00
0.455E+00
0.331E+00
0.253E+00
0.165E+01
0.119E+01
0.419E-01
0.150E-01
0.834E-02
0.553E-02
0.403E-02
3-35
-------
TABLE 3-11
Average Atmospheric Concentrations (/Lig/m3) , Including
Dry Deposition Effects, Predicted by RTDMDEP
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
RTDMDEP
30° Slope
0.396E+00
0.197E+01
0.238E+01
0.178E+01
0.661E+00
0.133E+01
0.140E+01
0.116E+01
0.174E+00
0.171E+00
0.169E+00
0.104E+00
RTDMDEP
45° Slope
0.396E+00
0.197E+01
0.238E+01
0.178E+01
0.661E+00
0.133E+01
0.140E+01
0.105E+01
0.174E+00
0.174E+00
0.170E+00
0.106E+00
RTDMDEP
60° Slope
0.396E+00
0.221E+01
0.238E+01
0.174E+01
0.661E+00
0.133'E+01
0.140E+01
0.988E+00
0.174E+00
0.178E+00
0.171E+00
0.106E+00
3-36
-------
travel time increases (i.e., distance from the source increases),
the reduction in pollutant concentration due to depletion becomes
apparent even at ground level.
3.5.3. Calculated Values of Pollutant Deposition
3.5.3.1. COMPDEP AND ISCST RESULTS — Unlike atmospheric
concentrations that were identical, predicted values for dry
deposition made by COMPDEP (Table 3-12) are different than those
of ISCST in flat terrain (Receptors 1 - 19) . Deposition rates for
receptors within 3000 meters of the source predicted by ISCST are
on average 2.3 times higher than those of COMPDEP. Maximum
deposition rates predicted by ISCST are as much as 10.0 times those
of COMPDEP. Beyond this distance, however, the opposite is true
as deposition rates predicted by COMPDEP are 1.1 to 4.2 times those
of ISCST. Examination of the remaining receptors yields similar
results. For points near the source, ISCST deposition rates vary
from 1.0 to 37.6 and average 3.0 times those of COMPDEP. As
downwind distance increases, values predicted by COMPDEP average
2.5 times higher than those of ISCST. Furthermore, for receptors
25, 50 and 100 meters above stack height, the point at which
COMPDEP predictions exceed those of ISCST shifts toward the stack
as terrain height increases. The fact that this trend is not seen
for the receptors 300 meters above the stack is probably due to the
increasing influence of the terrain correction factor used by
COMPDEP.
3-37
-------
TABLE 3-12
Average Annual Dry Deposition (g/m2/yr)
Predicted by COMPDEP and ISCST
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
COMPDEP
Deposition
0.816E-03
0.108E-01
0.158E-01
0.203E-01
0.279E-01
0.322E-01
0.318E-01
0.300E-01
0.277E-01 '
0.254E-01
0.212E-01
0.180E-01
0.134E-01
0.104E-01
0.420E-02
0.145E-02
0.770E-03
0.500E-03
0.359E-03
0.404E-02
0.267E-01
0.372E-01
0.455E-01
0.545E-01
0.555E-01
0.505E-01
0.451E-01
0.402E-01
0.360E-01
0.291E-01
0.239E-01
0.170E-01
0.127E-01
0.460E-02
0.153E-02
0.806E-03
0.520E-03
0.369E-03
0.916E-01
0.119E+00
0.132E+00
0.137E+00
0.129E+00
0.113E+00
0.926E-01
0.769E-01
0.646E-01
ISCST
Deposition
0.814E-02
0.626E-01
0.782E-01
0.872E-01
0.956E-01
0.917E-01
0.703E-01
0.562E-01
0.454E-01
0.371E-01
0.256E-01
0.185E-01
0.105E-01
0.693E-02
0.188E-02
0.475E-03
0.225E-03
0.139E-03
0.983E-04
0.150E+00
0.335E+00
0.331E+00
0.297E+00
0.217E+00
0.156E+00
0.981E-01
0.699E-01
0.522E-01
0.404E-01
0.257E-01
0.177E-01
0.962E-02
0.611E-02
0.156E-02
0.402E-03
0.197E-03
0.125E-03
0.899E-04
0.344E+01
0.908E+00
0.574E+00
0.394E+00
0.215E+00
0.134E+00
0.787E-01
0.533E-01
0.383E-01
3-38
-------
TABLE 3-12 (cont.)
Receptor
Number
48
49
52
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
COMPDEP
Deposition
0.549E-01
0.411E-01
0.318E-01
0.209E-01
0.148E-01
0.485E-02
0.157E-02
0.816E-03
0.520E-03
0.367E-03
0.535E+00
0.384E+00
0.334E+00
0.288E+00
0.216E+00
0.164E+00
0.122E+00
0.944E-01
0.752E-01
0.613E-01
0.434E-01
0.325E-01
0.204E-01
0.142E-01
0.449E-02
0.144E-02
0.744E-03
0.472E-03
0.333E-03
0.252E+00
0.181E+00
0.156E+00
0.134E+00
0.101E+00
0.768E-01
0.571E-01
0.441E-01
0.351E-01
0.286E-01
0.203E-01
0.151E-01
0.953E-02
0.661E-02
0.209E-02
0.671E-03
0.348E-03
0.221E-03
0.156E-03
ISCST
Deposition
0.287E-01
0.173E-01
0.115E-01
0.613E-02
0.386E-02
0.105E-02
0.307E-03
0.161E-03
0.108E-03
0.798E-04
0.344E+01
0.908E+00
0.574E+00
0.394E+00
0.215E+00
0.134E+00
0.787E-01
0.533E-01
0.383E-01
0.287E-01
0.173E-01
0.115E-01
0.613E-02
0.386E-02
0.105E-02
0.307E-03
0.161E-03
0.108E-03
0.798E-04
0.344E+01
0.908E+00
0.574E+00
0.394E+00
0.215E+00
0.134E+00
0.787E-01
0.533E-01
0.383E-01
0.287E-01
0.173E-01
0.115E-01
0.613E-02
0.386E-02
0.105E-02
0.307E-03
0.161E-03
0.108E-03
0.798E-04
3-39
-------
TABLE 3-13
Average Annual Dry Deposition (g/m /yr)
Predicted by RTDMDEP, COMPDEP and ISCST
Receptor
Number
1
2
3
4
5
6
7
8
9
10
11
12
RTDMDEP
30° Slope
0.382E-01
0.162E+00
0.179E+00
0.158E+00
0.584E-01
0.105E-01
0.973E-01
0.950E-01
0.129E-01
0.118E-01
0.104E-01
0.735E-02
RTDMDEP
45° Slope
0.382E-01
0.162E+00
0.179E+00
0.158E+00
0.584E-01
0.105E-01
0.972E-01
0.836E-01
0.129E-01
0.120E-01
0.104E-01
0.745E-02
RTDMDEP
60° Slope
0.382E-01
0.180E+00
0.179E+00
0.154E+00
0.584E-01
0.105E-01
0.972E-01
0.793E-01
0.129E-01
0.122E-01
0.105E-01
0.750E-02
COMPDEP
0.372E-01
0.132E+00
0.334E+00
0.156E+00
0.555E-01
0.113E+00
0.164E+00
0.768E-01
0.127E-01
0.148E-01
0.142E-01
0.661E-02
ISCST
0.331E+00
0.574E+00
0.574E+00
0.574E+00
0.156E+00
0.134E+00
0.134E+00
0.134E+00
0.611E-02
0.386E-02
0.386E-02
0.386E-02
3-40
-------
3.5.3.2. RTDMDEP RESULTS — Deposition rates were also
calculated by RTDMDEP (Table 3-13) for terrain 25, 50, 100, and 300
meters above stack height. Differences in deposition rates
predicted by the two models are less than for concentrations.
COMPDEP predictions range from 0.8 to 1.9 times those of the
RTDMDEP. Like COMPDEP, RTDMDEP predicted deposition rates 2.7
times lower on average and as much as 8.7 times lower than those
of ISCST near the source. At larger downwind distances, RTDMDEP
values were 1.9 to 3.1 times higher than those of ISCST.
3.5.4. Summary of Air Dispersion Modeling
Two complex terrain models, COMPDEP and RTDMDEP were developed
to account for pollutant transport in complex terrain. Atmospheric
deposition components were added to two existing, complex terrain
models (U.S. EPA, 1986c), COMPLEX I and RTDM, so that the models
could be used to estimate uptake of pollutants through the
terrestrial food chain. Both COMPDEP and RTDMDEP contain
algorithms taken from the MPTER-DS model (U.S. EPA, 1982a), which
calculate deposition rates based on a dry deposition velocity and
account for plume depletion due to deposition. Additional, state-
of-the-art algorithms (GARB, 1986), were incorporated that
automatically calculate particle size-dependent dry deposition
velocities based on atmospheric and surface conditions. Finally,
the capability to estimate wet deposition was incorporated into
these models through the use of scavenging coefficients.
To evaluate the impact of the new models' treatment of complex
terrain and deposition, comparisons were made between their results
3-41
-------
and those of the widely used ISCST model. Comparisons between
predicted concentrations showed, on the average, ISCST generally
estimated marginally higher annual average atmospheric pollutant
concentrations that were roughly twice those of COMPDEP and three
times those of RTDMDEP. A few isolated cases existed in which
ISCST actually gave slightly unconservative predictions with
respect to those made by COMPDEP. These occurred primarily at
large downwind distances and at receptor elevations of 100.0 meters
above stack elevation. Many risk assessments, such as those
weighted according to population density, are more a function of
overall average pollutant concentration near the source than a
maximum concentration at a single point. Based on the results of
this analysis, use of ISCST in these risk assessments would be
expected to give results that would range from about 0.9 to 3.0
times those obtained if either COMPDEP or RTDMDEP were used.
Hence, for such assessments, use of ISCST in complex terrain
appears to be valid.
Somewhat different conclusion can be drawn when considering
maximum point concentrations. Whereas differences in average
concentrations predicted by flat and complex terrain models were
small, differences in maximum point concentrations were much more
significant, differing in some instances by an order of magnitude.
In all cases, the maximum point concentrations were predicted by
ISCST. Thus in some cases, such as for risk assessments that base
their exposures on maximum point concentrations of pollutant, use
of ISCST in complex terrain must be questioned. If the user can
tolerate a factor of ten conservatism, this model can be applied.
3-42
-------
However, if this is not the case, consideration should be given to
either COMPDEP or RTDMDEP. Based on the experience gained during
this study, COMPDEP is recommended over RTDMDEP for use in these
situations. It is easier to use, faster, and can be applied to
more physical scenarios than RTDMDEP. Only for cases in which
concentrations estimated by COMPDEP appear to be too conservative
is RTDMEP recommended for use.
It should be noted that in spite of their technical
improvements, currently recommended complex terrain screening
models (U.S. EPA, 1986c) are still rather limited in their ability
to model atmospheric transport in complex terrain. Being of their
Gaussian plume formulation, their treatment of terrain is limited
to a few algorithms that modify plume height. In the future it is
possible that a refined complex terrain model, capable of
realistically simulating flow in complex terrain, will be developed
that can be used to obtain estimates of annual average atmospheric
pollutant concentrations.
Comparisons between predicted deposition rates showed that
those predicted COMPDEP and RTDMDEP were are as much as 39 times
lower than those of ISCST. The deposition algorithms incorporated
into COMPDEP and RTDMDEP are, in theory, more advanced than those
contained in ISCST and should provide more realistic estimates of
pollutant deposition rate. In addition unlike ISCST, which require
the user input highly subjective values for reflection coefficient,
they offer a systematic and automated means by which deposition
velocities can be estimated. However, accurate conclusions
regarding which of these algorithms provides the best estimates of
3-43
-------
dry deposition velocity cannot be obtained without reference to
quantitative evaluations between predicted and measured values.
Unfortunately such evaluations are rare. Bowers and Anderson
(1981) found that ISCST produced results within a factor of two of
measured values. In some instances calculated values were lower
than measured values. Based on these results, deposition rates
predicted by COMPDEP and RTDMDEP appear to grossly underestimate
pollutant deposition rates. However, these deposition velocities
were calculated under a limited set of atmospheric conditions and
downwind distances and for particle sizes that ranged from 50 to
200 microns, much larger than those addressed in this analysis.
In contrast to these results Tesche et al. (1987) indicated that
under certain conditions ISCST can significantly overestimate dry
deposition. Although no quantitative estimates of the degree of
overprediction by ISCST were reported, the Tesche et al. (1987)
results indicate that COMPDEP and RTDMDEP may provide better
estimates of deposition rate under certain conditions. Thus,
current evaluations of ISCST performance in estimating deposition
rate appear contradictory. Until more detailed evaluations of the
performance of models such as ISCST in estimating deposition rates
are available, meaningful recommendations regarding whether the
deposition algorithms developed for this study are indeed superior
to those found in ISCST cannot be made.
3.6. DETERMINATION OP AREAL-AVERAGE DEPOSITION
In this document, deposition values were derived from the
COMPDEP model described in Chapter 3 using unit (1 g/s) emission
3-44
-------
rate. The MWC modeled was a hypothetical facility in Western
Florida. Both wet and dry deposition was determined. These values
can be used to determine annual average deposition for various
areas (areal averages) surrounding the combustor. These areas are
defined as concentric rings of varying radii with the source at the
center. The maximum radius is 50 km surrounding the combustion
facility. Appendix A illustrates the concentric rings and lists
the receptors within each radius. In this document, an annual
average deposition value is calculated for each of the three
exposure scenarios. As discussed in Chapter 2, a hypothetical MWC
will be used as the combustion facility and examples shown for
cadmium and benzo(a)pyrene. The output for the COMPDEP model, run
at unit emission rates for cadmium and benzo(a)pyrene is shown in
Tables 3-14 - 3-18.
To support development of regulations for MWCs under the Clean
Air Act, Section 111, pollutant emission rates representative of
a variety of facility types were estimated based on available MWC
emission test reports (Myers, 1989). Myers (1989) estimated these
rates from both measured particulate matter and PCDD/PCDF emissions
and developed metal to particulate ratios and organic compound to
PCDD/PCDF ratios so that the emission rates of specific metals and
organic compounds could be quantified. Emission rate estimates for
cadmium and benzo(a)pyrene have been selected from this source for
use in developing the three exposure scenarios (A, B and C) in this
document. The selections are somewhat arbitrary and are chosen
primarily to illustrate the range in emission rates observed for
facilities of different designs and sizes.
3-45
-------
TABLE 3-14
Annual Dry Deposition for Benzo(a)Pyrene
Re
1
5
9
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
101
105
109
113
117
121
125
129
133
137
141
145
149
153
157
Dp
0.3406E-02
0.3870E-02
0.2259E-02
0.4720E-02
0.1816E-02
0.2019E-02
0.1158E-02
0.2580E-02
0.1340E-02-
0.1320E-02
0.7473E-03
0.1765E-02
0.1301E-02
0.1201E-02
0.7574E-03
0.1702E-02
0.6974E-03
0.6950E-03
0.5009E-03
0.1137E-02
0.3530E-03
0.3812E-03
0.2947E-03
0.7194E-03
0.1530E-03
0.1687E-03
0.1409E-03
0.3532E-03
0.8797E-04
0.9512E-04
0.8170E-04
0.2018E-03
0.5844E-04
0.6248E-04
0.5361E-04
0.1320E-03
0.4233E-04
0.4486E-04
0.3832E-04
0.9480E-04
* Maximum annual
Re -
Dp »
Receptor
Deposition
Re
2
6
10
14
18
22
26
30
34
38
42
46
50
54
58
62
66
70
74
78
82
86
90
94
98
102
106
110
114
118
122
126
130
134
138
142
146
150
154
158
Dp
0.1726E-02
0.3619E-02
0.4180E-02
0.3602E-02
0.1098E-02
0.1895E-02
0.2197E-02
0;1988E-02
0.1036E-02
0.1142E-02
0.1410E-02
0.1497E-02
0.9607E-03
0.1094E-02
0.1378E-02
0.1491E-02
0.4789E-03
0.7882E-03
0.9432E-03
0.9431E-03
0.2605E-03
0.5065E-03
0.5883E-03
0.5754E-03
0.1188E-03
0.2372E-03
0.2901E-03
0.2766E-03
0.6839E-04
0.1329E-03
0.1672E-03
0.1589E-03
0.4589E-04
0.8775E-04
0.1090E-03
0.1054E-03
0.3358E-04
0.6344E-04
0.7762E-04
0.7645E-04
dry deposition =
Re
3
7
11
15
19
23
27
31
35
39
43
47
51
55
59
63
67
71
75
79
83
87
91
95
99
103
107
111
115
119
123
127
131
135
139
143
147
151
155
159
Dp
0.2525E-02
0.3240E-02
0.5416E-02
0.3048E-02
0.1607E-02
0.1752E-02
0.3081E-02
0.1662E-02
0.1534E-02
0.1072E-02
0.2140E-02
0.1276E-02
0.1396E-02
0.1008E-02
0.2078E-02
0.1267E-02
0.6448E-03
0.7073E-03
0.1495E-02
0.7631E-03
0.3291E-03
0.4590E-03
0.9935E-03
0.4446E-03
0.1455E-03
0.2188E-03
0.5186E-03
0.2147E-03
0.8325E-04
0.1223E-03
0.3037E-03
0.1260E-03
0.5633E-04
0.8048E-04
0.1986E-03
0.8364E-04
0.4182E-04
0.5821E-04
0.1416E-03
0.6036E-04
0.6874E-02 at receptor:
Re
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
104
108
112
116
120
124
128
132
136
140
144
148
152
156
160
12
Dp
0.6839E-02
0.2324E-02
0.6874E-02*
0.2686E-02
0.3914E-02
0.1290E-02
0.3806E-02
0.1477E-02
0.3166E-02
0.8627E-03
0.2609E-02
0.1082E-02
0.3014E-02
0.8650E-03
0.2553E-02
0.1046E-02
0.1505E-02
0. 6547E-03
0. 1845E-02
0. 6420E-03
0.7448E-03
0.4476E-03
0.1212E-02
0.3748E-03
0.3134E-03
0.2220E-03
0.6221E-03
0.1785E-03
0. 1769E-03
0.1274E-03
0.3628E-03
0. 1035E-03
0. 1181E-03
0.8491E-04
0.2375E-03
0.6862E-04
0.8662E-04
0.6166E-04
0. 1696E-03
0.4965E-04
3-46
-------
TABLE 3-15
Annual Dry Deposition for Cadmium
(g/m2)
Re
1
5
9
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
101
105
109
113
117
121
125
129
133
137
141
145
149
153
157
Dp
0.7892E-02
0.8985E-02
0.5246E-02
0.1076E-01
0.4195E-02
0.4652E-02
0.2657E-02
0.5787E-02
0.3091E-02
0.3035E-02
0.1717E-02
0.3979E-02
0.2995E-02
0.2763E-02
0.1750E-02
0.3868E-02
0.1600E-02
0.1597E-02
0.1160E-02
0.2598E-02
0.8043E-03
0.8723E-03
0.6812E-03
0.1648E-02
0.3423E-03
0.3758E-03
0.3195E-03
0.7864E-03
0.1931E-03
0.2052E-03
0.1807E-03
0.4315E-03
0.1260E-03
0.1310E-03
0.1155E-03
0.2715E-03
0.8990E-04
0.9179E-04
0.8065E-04
0.1886E-03
RC
2
6
10
14
18
22
26
30
34
38
42
46
50
54
58
62
66
70
74
78
82
86
90
94
98
102
106
110
114
118
122
126
130
134
138
142
146
150
154
158
Dp
0.3955E-02
0.8406E-02
0.9643E-02
0.8257E-02
0.2496E-02
0.4326E-02
0.4996E-02
0.4496E-02
0.2359E-02
0.2603E-02
0.3210E-02
0.3397E-02
0.2189E-02
0.2513E-02
0.3165E-02
0.3403E-02
0.1091E-02
0.1818E-02
0.2176E-02
0.2158E-02
0.5933E-03
0.1171E-02
0.1359E-02
0.1318E-02
0.2634E-03
0.5290E-03
0.6552E-03
0.6160E-03
0.1467E-03
0.2832E-03
0.3652E-03
0.3410E-03
0.9588E-04
0.1801E-03
0.2301E-03
0.2188E-03
0.6866E-04
0.1262E-03
0.1587E-03
0.1545E-03
RC
3
7
11
15
19
23
27
31
35
39
43
47
51
55
59
63
67
71
75
79
83
87
91
95
99
103
107
111
115
119
123
127
131
135
139
143
147
151
155
159
Dp
0.5800E-02
0.7526E-02
0.1234E-01
0.6981E-02
0.3667E-02
0.3996E-02
0.6899E-02
0.3772E-02
0.3503E-02
0.2442E-02
0.4813E-02
0.2904E-02
0.3185E-02
0.2314E-02
0.4726E-02
0.2896E-02
0.1469E-02
0.1633E-02
0.3422E-02
0.1745E-02
0.7493E-03
0.1065E-02
0.2282E-02
0.1014E-02
0.3225E-03
0.4903E-03
0.1163E-02
0.4795E-03
0.1784E-03
0.2611E-03
0.6565E-03
0.2737E-03
0.1177E-03
0.1648E-03
0.4125E-03
0.1768E-03
0.8576E-04
0.1152E-03
0.2834E-03
0.1245E-03
RC
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
104
108
112
116
120
124
128
132
136
140
144
148
152
156
160
Dp
0.1579E-01*
0.5376E-02
0.1564E-01
0.6188E-02
0.9011E-02
0.2918E-02
0.8537E-02
0.3365E-02
0.7273E-02
0.1954E-02
0.5876E-02
0.2466E-02
0.6913E-02
0.1981E-02
0.5809E-02
0.2391E-02
0.3442E-02
0.1513E-02
0.4223E-02
0.1469E-02
0.1695E-02
0.1041E-02
0.2781E-02
0.8559E-03
0.6965E-03
0.4993E-03
0.1394E-02
0.3977E-03
0.3826E-03
0.2742E-03
0.7840E-03
0.2236E-03
0.2503E-03
0.1760E-03
0.4939E-03
0.1440E-03
0.1807E-03
0.1237E-03
0.3406E-03
0.1016E-03
* Maximum annual dry deposition = 0.1579E-01 at receptor: 4
Re = Receptor
Dp = Deposition
3-47
-------
TABLE 3-16
Annual Wet Deposition for Benzo(a)Pyrene
(g/m2)
Re
1
5
9
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
101
105
109
113
117
121
125
129
133
137
141
145
149
153
157
—
Dp
0.7370E-01
0.5229E-01
0.7245E-01
0.8744E-01
0.2208E-01
0.1690E-01
0.2147E-01
0.2477E-01
0.8540E-02
0.7105E-02
0.7742E-02
0.8865E-02
0.3401E-02
0.3001E-02
0.2857E-02
0.3344E-02
0.1041E-02
0.9100E-03
0.8035E-03
0.1015E-02
0.3906E-03
0.3223E-03
0.2899E-03
0.3751E-03
0.1220E-03
0.9150E-04
0.8766E-04
0.1119E-03
0.5349E-04
0.3787E-04
0.3739E-04
0.4720E-04
0.2714E-04
0.1862E-04
0.1839E-04
0.2322E-04
0.1497E-04
0.1013E-04
0.9818E-05
0.1250E-04
=====
* Maximum annual
Re =
Dp -
Receptor
Deposition
Re
2
6
10
14
18
22
26
30
34
38
42
46
50
54
58
62
66
70
74
78
82
86
90
94
98
102
106
110
114
118
122
126
130
134
138
142
146
150
154
158
=s
Dp
0.5439E-01
0.4445E-01
0.8588E-01
0.8513E-01
0.1657E-01
0.1364E-01
0.2555E-01
0.2434E-01
0.5868E-02
0.5249E-02
0.9756E-02
0.8847E-02
0.2376E-02
0.2017E-02
0.3874E-02
0.3414E-02
0.7418E-03
0.5973E-03
0.1241E-02
0.1014E-02
0.2730E-03
0.2196E-03
0.4795E-03
0.3651E-03
0.8024E-04
0.6763E-04
0.1498E-03
0.1054E-03
0.3333E-04
0.2967E-04
0.6412E-04
0.4320E-04
0.1614E-04
0.1515E-04
0.3153E-04
0.2066E-04
0.8555E-05
0.8459E-05
0.1685E-04
0.1079E-04
======^==
wet deposition = 0.
Re
3
7
11
15
19
23
27
31
35
39
43
47
51
55
59
63
67
71
75
79
83
87
91
95
99
103
107
111
115
119
123
127
131
135
139
143
147
151
155
159
Dp
0.4516E-01
0.4787E-01
0.7578E-01
0.1021E+00
0.1385E-01
0.1564E-01
0.2185E-01
0.3133E-01
0.5195E-02
0.6497E-02
0.7707E-02
0.1248E-01
0.2106E-02
0.2774E-02
0.3023E-02
0.5037E-02
0.6586E-03
0.8922E-03
0.9495E-03
0.1525E-02
0.2397E-03
0.3333E-03
0.3545E-03
0.5593E-03
0.6800E-04
0.9911E-04
0.1058E-03
0.1678E-03
0.2740E-04
0.4161E-04
0.4403E-04
0.7128E-04
0.1302E-04
0.2050E-04
0.2117E-04
0.3517E-04
0.6832E-05
0.1112E-04
0.1108E-04
0.1892E-04
========
1129E+00 at receptor:
Rc
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
104
108
112
116
120
124
128
132
136
140
144
148
152
156
160
12
Dp
0.6277E-01
0.5710E-01
0.1129E+00*
0. 1123E+00
0.2098E-01
0.1836E-01
0. 3545E-01
0.3634E-01
0.8787E-02
0. 7178E-02
0.1409E-01
0. 1410E-01
0.3681E-02
0.2950E-02
0.5958E-02
0.5715E-02
0. 1140E-02
0.9517E-03
0.1895E-02
0. 1744E-02
0. 4250E-03
0. 3688E-03
0. 6974E-03
0. 6437E-03
0. 1321E-03
0 . 1154E-03
0.2047E-03
0.1959E-03
0. 5877E-04
0 . 4945E-04
0 . 8529E-04
0.8473E-04
0.3055E-04
0 . 2431E-04
0.4154E-04
0. 4268E-04
0. 1734E-04
0. 1296E-04
0. 2217E-04
0.2348E-04
-
3-48
-------
TABLE 3«17
Annual Wet Deposition for Cadmium
(g/m2)
RC
1
5
9
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
101
105
109
113
117
121
125
129
133
137
141
145
149
153
157
Dp
0.1202E+00
0.8496E-01
0.1213E+00
0.1460E+00
0.3159E-01
0.2330E-01
0.3100E-01
0.3549E-01
0.1027E-01
0.8183E-02
0.9276E-02
0.1051E-01
0.3406E-02
0.2978E-02
0.2843E-02
0.3311E-02
0.9640E-03
0.8463E-03
0.7404E-03
0.9409E-03
0.3619E-03
0.3003E-03
0.2685E-03
0.3485E-03
0.1141E-03
0.8632E-04
0.8240E-04
0.1053E-03
0.5067E-04
0.3612E-04
0.3562E-04
0.4500E-04
0.2605E-04
0.1794E-04
0. 1775E-04
0.2241E-04
0.1456E-04
0.9859E-05
0.9593E-05
0.1222E-04
Re
2
6
10
14
18
22
26
30
34
38
42
46
50
54
58
62
66
70
74
78
82
86
90
94
98
102
106
110
114
118
122
126
130
134
138
142
146
150
154
158
Dp
0.8744E-01
0.7355E-01
0.1447E+00
0.1395E+00
0.2292E-01
0.1958E-01
0.3703E-01
0.3407E-01
0.6933E-02
0.6311E-02
0.1157E-01
0.1028E-01
0.2368E-02
0.2030E-02
0.3813E-02
0.3357E-02
0.6904E-03
0.5515E-03
0.1150E-02
0.9428E-03
0.2535E-03
0.2035E-03
0.4459E-03
0.3388E-03
0.7528E-04
0.6357E-04
0.1409E-03
0.9878E-04
0.3168E-04
0.2821E-04
0.6109E-04
0.4112E-04
0.1555E-04
0.1456E-04
0.3046E-04
0.1996E-04
0.8351E-05
0.8212E-05
0.1650E-04
0.1058E-04
RC
3
7
11
15
19
23
27
31
35
39
43
47
51
55
59
63
67
71
75
79
83
87
91
95
99
103
107
111
115
119
123
127
131
135
139
143
147
151
155
159
Dp
0.7260E-01
0.8140E-01
0.1285E+00
0.1684E+00
0.1903E-01
0.2260E-01
0.3171E-01
0.4478E-01
0.6011E-02
0.7662E-02
0.9147E-02
0.1482E-01
0.2086E-02
0.2749E-02
0.2976E-02
0.5013E-02
0.6152E-03
0.8285E-03
0.8816E-03
0.1418E-02
0.2229E-03
0.3107E-03
0.3294E-03
0.5197E-03
0.6366E-04
0.9366E-04
0.9943E-04
0.1573E-03
0.2600E-04
0.3980E-04
0.4195E-04
0.6762E-04
0.1253E-04
0.1983E-04
0.2047E-04
0.3379E-04
0.6671E-05
0.1087E-04
0.1087E-04
0.1841E-04
RC
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
104
108
112
116
120
124
128
132
136
140
144
148
152
156
160
Dp
0.1039E+00
0.9575E-01
0.1895E+00*
Ool861E+00
0.2982E-01
0.2659E-01
0.5080E-01
0.5262E-01
0.1038E-01
0.8549E-02
0.1654E-01
0.1703E-01
0.3669E-02
0.2917E-02
0.5883E-02
0.5759E-02
0.1058E-02
0.8828E-03
0.1763E-02
0.1621E-02
0.3950E-03
0.3430E-03
0.6482E-03
0.5955E-03
0.1242E-03
0.1085E-03
0.1924E-03
0.1832E-03
0. 5587E-04
0.4712E-04
0.8131E-04
0.8037E-04
0.2934E-04
0.2349E-04
0.4015E-04
0.4103E-04
0.1682E-04
0.1270E-04
0.2171E-04
0.2285E-04
Maximum annual wet deposition = 0.1895E+00 at receptor: 12
Re = Receptor
Dp = Deposition
3-49
-------
TABLE 3-18
Average Hourly Concentrations for Benzo(a)Pyrene
' _3 \
Rc
1
5
9
13
17
21
25
29
33
37
41
45
49
53
57
61
65
69
73
77
81
85
89
93
97
101
105
109
113
117
121
125
129
133
137
141
145
149
153
157
Dp
0.4043E-01
0.4649E-01
0.3550E-01
0.6635E-01
0.2137E-01
0.2421E-01
0.1782E-01
0.3633E-01
0.1616E-01
0.1590E-01
0.1107E-01
0.2429E-01
0.1635E-01
0.1484E-01
0.1114E-01
0.2309E-01
0.9254E-02
0.8919E-02
0.7522E-02
0.1531E-01
0.4939E-02
0.5016E-02
0.4488E-02
0.9487E-02
0.2341E-02
0.2503E-02
0.2335E-02
0.5248E-02
0.1458E-02
0.1596E-02
0.1509E-02
0.3494E-02
0.1035E-02
0.1151E-02
0.1094E-02
0.2584E-02
0.7901E-03
0.8893E-03
0.8482E-03
0.2030E-02
* Maximum average
Rc »
Dp -
Receptor
Deposition
Rc
2
6
10
14
18
22
26
30
34
38
42
46
50
54
58
62
66
70
74
78
82
86
90
94
98
102
106
110
114
118
122
126
130
134
138
142
146
150
154
158
Dp
0.2242E-01
0.4867E-01
0.6131E-01
0.5002E-01
0.1418E-01
0.2561E-01
0.3224E-01
0.2745E-01
0.1339E-01
0.1530E-01
0.2031E-01
0.2027E-01
0.1275E-01
0.1449E-01
0.1950E-01
0.2015E-01
0.6497E-02
0.1034E-01
0.1338E-01
0.1282E-01
0.3527E-02
0.6477E-02
0.8251E-02
0.7753E-02
0.1794E-02
0.3536E-02
0.4494E-02
0.4196E-02
0.1164E-02
0.2348E-02
0.2974E-02
0.2776E-02
0.8489E-03
0.1734E-02
0.2192E-02
0.2045E-02
0.6605E-03
0.1361E-02
0.1720E-02
0.1603E-02
hourly concentration
Rc
3
7
11
15
19
23
27
31
35
39
43
47
51
55
59
63
67
71
75
79
83
87
91
95
99
103
107
111
115
119
123
127
131
135
139
143
147
151
155
159
= 0
Dp
0.3095E-01
0.4430E-01
0.7950E-01
0.4320E-01
0.1969E-01
0.2413E-01
0.4574E-01
0.2321E-01
0.1905E-01
0.1463E-01
0.3109E-01
0.1740E-01
0.1790E-01
0.1356E-01
0.2922E-01
0.1739E-01
0.8487E-02
0.9373E-02
0.2066E-01
0.1080E-01
0.4334E-02
0.5826E-02
0.1343E-01
0.6379E-02
0.2139E-02
0.3210E-02
0.7786E-02
0.3358E-02
0.1386E-02
0.2146E-02
0.5295E-02
0.2189E-02
0.1016E-02
0.1596E-02
0.3969E-02
0.1598E-02
0.7956E-03
0.1260E-02
0.3149E-02
0.1245E-02
.9888E-01 at
Rc
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
64
68
72
76
80
84
88
92
96
100
104
108
112
116
120
124
128
132
136
140
144
148
152
156
160
Dp
0.7807E-01
0.3720E-01
0.9888E-01*
0.3588E-01
0.4450E-01
0.2040E-01
0.5532E-01
0.1961E-01
0.3688E-01
0.1323E-01
0.3721E-01
0.1422E-01
0.3646E-01
0.1287E-01
0.3534E-01
0.1403E-01
0.1893E-01
0.9370E-02
0.2531E-01
0.8942E-02
0.9690E-02
0.6056E-02
0.1637E-01
0.5267E-02
0.4550E-02
0.3444E-02
0.9345E-02
0.2753E-02
0.2852E-02
0.2334E-02
0.6307E-02
0.1790E-02
0.2044E-02
0.1748E-02
0.4704E-02
0.1305E-02
0.1575E-02
0.1387E-02
0.3719E-02
0.1016E-02
receptor: 12
3-50
-------
Cadmium emissions were found to be related to both facility
type and particulate matter (PM) emission rate. Highest emissions
per unit PM emission rate (5539 /zg Cd/g PM) were for mass burn (MB)
facilities with PM control but without acid gas control, where as
lowest rates were for refuse-derived fuel (RDF) combustors with
only PM control (395 /xg Cd/g PM) or for all combustors with acid
gas control (390 jug Cd/g PM) .
Benzo(a)pyrene emissions were assumed to be related to total
PCDD/PCDF concentrations, in spite of the lack of an observed
correlation between these rates. The ratio of B(a)P:PCDD/PCDF was
assumed to be 11.15 for all MWC facility types. Based on the
emission rate ratios of both cadmium to particulate matter and
benzo(a)pyrene to PCDD/PCDF, Myers (1989) modeled representative
pollutant emission rate estimates for a variety of facility types.
Cadmium emissions for a large MB facility were estimated to be 1843
kg Cd/yr and will be used to estimate deposition. Since
benzo(a)pyrene emissions were assumed proportional to PCDD/PCDF
emissions, and since the latter differ from particulates in
emission characteristics, the type of facility selected to
represent the various cases will differ for these two chemicals.
The highest benzo(a)pyrene emissions (6.5 kg B(a)P/yr) were chosen
from a MB facility. Table 3-19 lists the emission estimated for
the different facility types.
The modeled wet and dry deposition rates (from Chapter 3) were
adjusted by the above emission rates, and areal-average deposition
values determined for the three exposure scenarios (See Table 3-
20). Areal-average deposition was determined for Scenario B by
3-51
-------
TABLE 3-19
Emission Rates (kg/yr) for Planned MWCs
Plant Type
Small MB/WW
Medium MB/WW
Large MB/WW
MB/Ref
MB/RC
RDF
RDF (Cofired)
MI/EA
MI/SA No heat rec
MI/SA
Bubbling Bed FBC
Circulating Bed FBC
Particulate
Matter8
29610.36
118386.99
332708.26
73935.17
155127.78
364232.77
322956.08
56825.76
9071.80
18869.34
33000.00
33000.00
Cadmium6
164.0
655.75
1842.87
409.53
859.25
143.87
63.78
314.76
12.05
35.64
13.04
6.52
PCDD/
PCDFa'c
10.00050
0.00323
0.00909
0.00303
0.00633
0.00996
0.00883
0.00097
0.00016
0.00061
0.00045
0.00895
Benzo(a) -
Pyreneb
0.36
2.31
6.5
2.17
4.53
7.13
6.32
0.69
0.14
0.43
0.32
6.41
a Measured emissions
b Predicted emissions
Particulate emission is expressed as TCDD equivalents. TCDD equivalents
for MWC with PM control were determined by multiplying the emission rate
by the factor 0.0156.
3-52
-------
o
CM
CO
9
3
EH
oi '
SL
o
N
a
_
*d
q
<0
•H
B
•a
cO
O
J^
fi
01
-p
03
§
•H
4J
-H
01
O
ft
01
Q
o
a
g
^
c _
o
•d o 77
o >r\ rv
-p -p <|
S (fl~B
•0 O \
T! &CJ1
c
o —
•rH 4) U
01 JJ \
01 (0 O!
•H « »"
1
C Xj
o t><
H'M^
c 01 B
£3 -H g;
'1-
r*
«
•d o
a) TH "P
•p -P ii
m-H ^
3 Olfv/^
•r-i O s
•a a\
< t
•H Olry—
C 01 B
6 O1
W "*~*"
c
1^
(D -H M
-p -p >1
rj (Q00^
*n O ^
ril 0)
Q ^
O ,_,
•H 01 ",7
to -p-;?
01 10 «;
-H « S
B
c rr
o cL
•P-H C|
•^4 raj^
BUB
w ~^
^j
'o
X
•3"
H
'o
i-l
• X ''
O
• (Mt
ro
'o
. ,X
r>
co
*
,. VO
1^
'o
^
in
M
**
0
. H
VO
o
CM
ro
'o
rH
01
•H
»
i— J
CO
'o
X
r-
,
CO
-*
'o
,_(
X
VO
o
CM
'o
H
X
CO
H
„
5 Q
CQ
o
i-4
X
CO
CO
CM
'o
H
X
O
CM
^
'o
i-4
X
CO
H
*
H
^_
'o
t-4
X
f>
VO
**
O
H
X
O
CM
ro
'o
i-l
X
CM
•
CO
CO
'o
H
X
r>
CM
-»
'o
H
X
0
CM
'o
H
X
CO
CO
H
I
,
'o
H
X
CM
O
Cv]
'o
"X
o
CM
*
(M
'o
X
CO
>
01
*
•o
'o
1-1
X
CM
CO
ro
*f
O
H
X
vo
o
CM
*
'o
H
X
VO
•
H
*
'o
X
o
H
VO
-»
O
H
X
vo
O
•
CM
*
ro
'o
H
X
VO
01
CM
O
'o
i-4
X
H
Ol
CO
in
o
1NJ
'o
i-4
X
CO
in
•
H
-*
'o
H
X
C!
in
H
CO
in
o
•
ro
'o
,-1
X
r-
•
CM
'o
H
X
•*
CO
CM
CO
in
o
**
'o
H
CO
O
•*
* *
O
'o
X
H
i-l
CO
in
0
T-
'o
rH
X
o
01
•
i-4
*T
'o .
H
X
p^
i-l
CM
CO
in
o
•
ro
'o
i-4
X
^"
I-
•
CO
*o
'o
X
VO
VO
CO
in
o
•
-*
'o
•H
X
H
H
Q
ro
B
\
=i
4J
C
a
rH
f-\
o
Q.
o
c
o
v>
id
J_t
_(J
f3
a>
o
c
o
0
Vj
•rH
a)
4J
0)
•H
i
10
II
o
*
3-53
-------
using deposition for radii <5 km, whereas Scenario A was calculated
from the radii £50 km. The point of maximal deposition (at 0.2 km)
was used for Scenario C. These values are the initial input for
the determination of soil and water concentrations. Appendix A
shows the formula for calculation of the areal-average deposition.
3-54
-------
4, CALCULATING SOIL CONCENTRATION
4.1. INTRODUCTION
Contaminants associated with combustor emissions are subject
to deposition on soil surfaces in the vicinity of the point source.
Following deposition, contaminants may be incorporated into the
upper layers of soil where crops or other vegetation are grown.
This chapter focuses on the estimation of the cumulative
concentration of a given contaminant in the soil following
deposition from a combustor. The concentrations will be used to
estimate the risk to humans who may ingest soil directly or consume
vegetation and animals that have been exposed to contaminated soil.
Contaminant levels in soil are the result of both wet and dry
particulate deposition onto soil and the loss of contaminant due
to leaching, abiotic and biotic degradation and volatilization.
Determination of annual deposition was discussed in Chapter 3.
Figure 4-1 illustrates the number of factors required to determine
the cumulative soil concentration. Soil conditions such as pH,
soil structure and characteristics, organic matter content and
water content affect the distribution and mobility of contaminants
after deposition onto the soil. Loss of contaminants from the
soil is modeled using rates that depend on site-specific data about
the physical and chemical characteristics of the soil.
4-1
-------
I
5
o
m
o>
4-2
-------
4.2. OVERVIEW OF SOIL CONCENTRATION CALCULATIONS
The cumulative soil concentration of a pollutant is derived
from the wet and dry deposition rates over the lifetime of the
combustor and the contaminant loss rate from the soil.
Contaminants may be lost from soils as a result of numerous
factors, including leaching, abiotic and biotic degradation and
volatilization. The equations used to estimate contaminant
concentrations and losses in soil are adapted from Travis et al.,
(1983) . A summary of the general assumptions for the soil
concentration estimates is presented in Table 4-1.
4.2.1. Calculating Cumulative Soil Concentration. The cumulative
concentration of a pollutant (mg pollutant/g soil) is calculated
from the wet and dry deposition rates and the loss constant:
Sc = (Dyd + Dyw) • [1.0-exp(-ks • Tc)] • 0.1
BD
ks
(Equation 4-1)
where:
Sc = soil concentration of pollutant after total time period
of deposition (mg pollutant/g soil)
Dyd = yearly dry deposition rate of pollutant
(g pollutant/m2/yr-)
Dyw = yearly wet deposition rate of pollutant
(g pollutant/m2/yr)
ks = soil loss constant (yr~1)
Tc = total time period over which deposition occurs (yrs)
0.1 = units conversion factor
Z = soil depth (cm)
BD = soil bulk density (g/cm3)
4-3
-------
TABLE 4-1
Assumptions for Soil Concentration Calculations
Assumptions
Ramifications/Limitations
If soil incorporation of
contaminant is assumed,
incorporation depth is 20 cm,
and the upper 20 cm soil
layer has a dry mass of 2.7
x 103 Mg/ha.
Soil background concentra-
tion is not considered.
Degradation of organic con-
taminants is first-order.
Trace metal contaminants are
assumed to be conserved
indefinitely in the upper
layer of soil unless loss
constants are available.
Erosion is not considered as
a loss mechanism.
The fact that loss processes
occur simultaneously is not
taken into account.
By this assumption, a soil concentration
of 1 jug/g corresponds to a pollutant
application of 2.7 kg/ha. If actual depth
(and mass) is less, contaminant
concentration could be underpredicted (and
vice versa).
The calculated soil concentration is only
the soil increment due to combustion
fallout, and could underpredict the soil
concentration.
Could over- or underpredict degradation
rate, which is complex and not necessarily
first-order.
Although most heavy metals are tightly
bound to the soil, measurements show that
this assumption often overpredicts
concentrations (i.e., at low pH metals
become increasingly mobile and may be
leached out of the upper soil layers).
Could overestimate the soil contaminant
concentration because significant losses
can occur by this process.
Could overpredict losses because
contaminant concentration available for
each loss process would be overp'redicted.
4-4
-------
The cumulative soil concentration, Sc, represents the
concentration increment due to accumulation of contaminant
deposited onto soil from the combustor. The cumulative soil
concentration does not represent the total concentration, because
it does not take into account background concentrations of the
contaminant that may already be present, whether natural or from
other pollution sources.
4.2.2. Calculating the Soil Loss Constant. The soil loss constant
is determined from the loss due to leaching and the loss due to
degradative and volatilization processes:
ks = ksl + ksg + ksv
(Equation 4-2)
where:
ks
ksl
ksg
soil loss constant due to all processes (yr~1)
loss constant due to leaching (yr"1)
loss constant due to degradation (abiotic and biotic)
~1
ksv = loss constant due to volatilization (yr )
Losses due to degradation (ksg) are empirically determined
from field studies and should be available in literature. Soil
losses due to leaching (ksl) and volatilization (ksv) can be
calculated as shown below.
The soil loss constant of contaminants due to leaching
(yr"1) is obtained by calculating the water availability in the
soil and the influence of soil properties on water and contaminant
movement by the following equation:
4-5
-------
ksl =
P + I - Ev
(Equation 4-3)
6 • Z • [1.0 + (BD •
where:
ksl -
P =
I -
Ev =
e =
z -
BD »
K,, =
~1
loss constant due to leaching (yr~)
average annual precipitation (cm/yr)
average annual irrigation (cm/yr)
average annual evapotranspiration (cm/yr)
soil volumetric water content (mL/cm )
soil depth from which leaching removal occurs (cm)
soil bulk density (g/cm3)
soil water partitioning coefficient (mL/g)
The soil loss constant due to volatilization (ksv) is based
on gas equilibrium coefficients and gas phase mass transfer:
ksv = Ke
Kt
(Equation 4-4)
where:
ksv
Ke
Kt
soil loss constant due to volatilization (yr~1)
equilibrium coefficient
gas phase mass transfer coefficient (cm/s) •
The equilibrium coefficient is calculated using the following
equation:
where:
Ke
H
Z
*d
R
T
BD
3.1536x10
10
H
Ke =
R
BD
(Equation 4-5)
= equilibrium coefficient
= Henry's law constant (atm-m3/mole)
= soil depth (cm)
= soil-water partitioning coefficient (mL/g)
= ideal gas constant (L-atm/mole-°K)
= temperature (°K)
= bulk density (g/cm3)
4-6
-------
The gas phase mass transfer can be calculated by the following
equation:
Kt = 0.482 • u°'78 • N"°-67 • d"°-11 (Equation 4-6)
where:
Kt = gas phase mass transfer coefficient (cm/s)
u = wind speed (m/s)
N = Schmidt number for gas phase (unitless)
de = effective diameter of contaminated area (m)
To calculate the Schmidt number for gas phase, the following
equation can be used:
N
(Equation 4-7)
pa
where:
N
pa
D_
Schmidt number for gas phase (unitless)
viscosity of air (g/cm-s)
air density (g/cm )
diffusion coefficient of pollutant in air (cmz/s)
The rate constant due to degradation (ksg) is based on field
observations or empirical data and is discussed in more detail in
section 4.3.5.2.
4.3. DESCRIPTION OF INPUT VARIABLES
Calculating cumulative soil concentration requires the input
of site-specific data, which must be calculated or derived from the
available literature. This section defines the input variables
required for the calculations, discusses the factors that affect
the value of each variable, and outlines the typical range of
values and sources of each variable.
4-7
-------
4.3.1. Dry and Wet Deposition Rates (Dyd and Dyw). Site-specific
dry and wet deposition rates are determined by air dispersion
modeling. Several methods for determining deposition rates are
described in Chapter 3.
4.3.2. Time Period of Deposition (Tc). In this methodology,
individuals are assumed to be exposed to contaminants in the soil
following a time period in which there has been continuous
combustor emissions. The lifetime of a combustion facility, such
as a MWC, could be considered to be > 30 years. However, since
the site would already be dedicated to incineration, the combustor
lifetime could be assumed to be as great as 100 years. In this
document, 30, 60, and 100 years of continuous deposition will be
used to construct Scenario A, Scenario B and Scenario C,
respectively.
4.3.3. Soil Depth (Z) . When modeling exposures to pollutants
found in soils, the depth of assumed contamination is important in
calculating the appropriate soil concentration. Contaminants
deposited onto soil surfaces may be incorporated into lower soil
profiles by tilling, whether done manually in a garden, or
mechanically in a large field. In general, if the area under
consideration is likely to be tilled (e.g., agricultural soils),
soil depth may be assumed to be 10 - 20 cm, depending on local
conditions and the equipment used. This methodology uses an
incorporation depth of 20 cm for tilled soils. If soil
incorporation does not occur, contaminants are assumed to be
4-8
-------
retained in a shallower, uppermost soil layer. While the actual
depth of this uppermost layer is unknown, a value of 1 cm will be
assumed in this methodology. If more specific information is
available on soil mixing depth for a particular location, a
different value can be substituted.
Soil concentrations based on soil depth are used to calculate
exposure via several pathways: ingestion of plants contaminated
by root uptake and by volatilization from the soil, direct
ingestion of the soil by humans or herbivores, surface runoff into
water bodies and dermal contact with the soil. The assumed soil
depth varies according to the pathway under consideration. When
calculating exposures due to uptake through plant roots, the
average concentration of pollutant over the depth of the plant root
determines plant uptake. For agricultural soils, the root depth
is assumed to equal the tilling depth. For untilled soils (e.g.,
grazing pastures) the root zone does not directly reflect a tilling
depth. However, this methodology assumes that tilling depth is an
adequate substitute for root zone depth. If site-specific
information is available for root zone depth, it can be used as the
depth of contamination.
Plant uptake due to volatilization is assumed to occur from
the uppermost soil layer. Any volatile contaminants are likely to
be desorbed from particles soon after emission from the combustor,
prior to deposition onto the soil. However, semivolatile compounds
and volatiles emitted in sufficiently high concentrations may be
deposited in particulate form and exhibit volatilization losses
from soils. Pollutants subject to volatilization losses may be
4-9
-------
incorporated to 20 cm by tilling and will not readily volatilize
from this depth. The volatilization rate will reflect the
contaminant concentration at the soil surface.
For direct ingestion of soil by humans and herbivores, a depth
of 1 cm is used. Children who either intentionally or incidentally
eat soil are most likely to be exposed to contaminated soil in
gardens, lawns, landscaped areas, parks and recreational areas.
Herbivorous animals are most likely to be exposed to contaminated
soils while grazing pastures. Most areas where soil exposure could
occur are unbilled or are tilled irregularly (e.g., pastures,
lawns, parks), so the appropriate depth assumption ordinarily is
1 cm.
For dermal exposure, the conditions and areas of likely
exposure are similar to those of the soil ingestion pathway.
Therefore, as in the soil ingestion pathway, the dermal exposure
model usually assumes a depth of 1 cm in estimating risk due to
contact with contaminated soil. This assumption overestimates soil
concentrations where incorporation occurs to any depth greater than
1 cm, and need not be used if site-specific information is
available.
For exposure through the surface water, contaminants are
assumed to travel from a watershed into a surface water body
dissolved in runoff water and adsorbed to eroded soil particles.
The assumptions for soil depth for the watershed area are site-
specific. If the area under consideration is an agricultural area
4-10
-------
that is likely to be tilled, soil depth is assumed to be 20 cm.
In all other cases, soil depth is assumed to be 1 cm, to be
conservative.
4.3.4. Bulk Density of the Soil (BD). The soil bulk density (BD)
is the ratio of the mass of the soil to its total volume. Bulk
density is affected by the soil structure (e.g., the looseness or
compaction of the soil, depending on the water and clay content of
the soil) (Hillel, 1980). The values range from 0.93 - 1.84 g/cm3
depending on soil type (Hoffman and Baes, 1979). This document
assumes a bulk density of 1.5 g/cm3. However, bulk density of the
soil at the study site should be used, if available.
4.3.5. Soil Loss Constant (ks). Organic and some inorganic
contaminants may be lost from the soil by many processes, which
may occur simultaneously and/or at different rates. The rate at
which a contaminant is lost from the soil is known as the "soil
loss constant" and is determined by the physical, chemical and
biological characteristics of the soil. The processes that remove
contaminants from the soil include leaching, biotic and abiotic
degradation, and volatilization.
Organics are subject to loss by leaching, degradation and
volatilization. Inorganics, such as metals, may be lost by
leaching and in some cases, by volatilization. This section
discusses the kinetics of contamination loss, including first-
4-11
-------
order processes, which this methodology generally assumes, and
other types of kinetics that may govern loss of contaminants from
soil.
First-order reaction rates depend on the concentration of one
reactant (Bohn et al., 1985). The loss of a contaminant by a
first-order process depends only on the concentration of the
contaminant in the soil, and a constant fraction of the compound
is removed from the soil over time.
Those processes that apparently exhibit first-order kinetics
without implying a mechanistic dependence on a first-order loss
rate are termed "apparent first-order" loss rates (Sparks, 1989).
Some higher-order kinetic processes appear to be first-order
because of the relative abundance of some reactants, such as water
and oxygen, in comparison with one reactant. This type of reaction
is limited by the concentration of only one reactant under the
conditions at which the apparent rate law is applicable (Sparks,
1989).
The assumption that contaminant loss follows first-order
kinetics may be an oversimplification since at various
concentrations or environmental conditions, the loss rates from
soil systems will resemble different kinetic expressions. However,
at low concentrations, a first-order loss constant may be adequate
to describe the loss of the chemical from the soil.
In addition to first-order kinetics, contaminant loss in soil
could follow zero-order or second-order kinetics. Zero-order loss
rates are independent of reactant concentrations (Bohn et al.,
1985). Zero-order rates describe some processes in which the
4-12
-------
reactants are in very high concentration. Under zero-order
kinetics, a constant amount of a contaminant is lost from the soil
over time, independent of its concentration.
Processes that follow second-order kinetics are dependent on
the concentrations of two reactants or on the concentration of one
reactant squared (Bohn et al., 1985). The loss constant of
contaminant disappearance following a second-order process can be
contingent on its own concentration, or on both its concentration
and the concentration of another reactant, such as an enzyme or
catalyst.
Because contaminant loss from soil is dependent on many
complex factors, it may be difficult to model the overall rate of
loss. Combined soil loss rates by these processes can be derived
experimentally and values for certain compounds are presented in
U.S. EPA (1986d) . If field data where soil concentrations have
been followed over time are available, these data should be used
to calculate the soil concentration. Loss of the contaminant can
be measured as the total loss by the three processes. In this case
it would not be necessary to calculate the constant as described
in Section 4.2.2. If the loss constant must be calculated, the
input variables described below should be used. It should be noted
that physical phenomena that may cause losses of contaminants from
soils, such as leaching, volatilization and degradation can
simultaneously occur, causing an overestimation of loss rates for
each process (Valentine, 1986). When possible, the common
occurrence of all loss processes should be taken into account.
4-13
-------
4.3.5.1. SOIL LOSS DUE TO LEACHING (ksl) — Losses of soil
contaminants due to leaching are dependent on the amount of water
available to generate leachate and on soil properties such as bulk
density, soil porosity, and sorptive properties. Equation 4-3 in
section 4.2.2. shows the calculation of ksl.
4.3.5.1.1. Available Water (P + I - Ev) — The average
annual volume of water available to generate leachate is the sum
of the average annual precipitation (P) and the average annual
irrigation (I) minus the average annual evapotranspiration (Ev).
Precipitation values range from 18.06 to 164.19 cm/year (data for
69 selected cities; U.S. Bureau of the Census, 1987). Irrigation
values range from 0 to 100 cm/year (data for 69 selected cities;
Baes et al., 1984) Evapotranspiration values range from 35 to 100
cm/year (Baes et al., 1984). The available water as calculated by
(P + I - Ev) is assumed to be a mass balance of all water inputs
and outputs from the area in consideration. Other mechanisms of
removal or retention in the soil are not considered in this
methodology.
4.3.5.1.2. Soil Properties — Three variables take into
account the soil properties that affect leachate formation: bulk
density (BD), soil water content (6), and the soil-water
partitioning coefficient (Kd) . Bulk density is described in
Section 4.3.4. The soil water content (9) depends on the available
water and on the soil structure. Values for 6 range from 0.03 to
0.40 mL/cm3 depending on soil type (Hoffman and Baes, 1979). The
4-14
-------
equilibrium partitioning coefficient (Kd) , used to calculate the
infiltration rate, reflects the sorption/desorption kinetics of the
soil. Mean values for Kd for many chemicals can be found in Baes
et al. (1984). However, variations in soil type greatly effect Kd,
which can vary up to three orders of magnitude in soils ranging
from pH 4.5 - 9.0. Therefore, it is advisable to use site-
specific data for Kd when available. Partitioning coefficients
for inorganic pollutants used in modeling may be derived from soil-
to-plant transfer coefficients (Baes et al., 1984). Partitioning
coefficients for organics may be estimated from Koc/ the Kd
correlated to the soil organic carbon content, if the contaminant
concentrations are in the appropriate range (
-------
1986). First-order loss rates often do not account for the high
variability of these parameters in a single soil system. However,
the use of simple rate expressions may be appropriate at low
chemical concentrations (ng range) where a first-order dependence
on chemical concentration may be reasonable.
The rate of biological degradation of contaminants is chemical
specific, depending on the complexity of the chemical and the
usefulness of the chemical to the microorganisms. Some substrates
are co-metabolized, not used by the organisms as a nutrient or
energy source, but simply degraded along with other similar
chemicals, which can be further utilized. Environmental and
chemical factors that may limit the biodegradation of chemicals in
the soil environment include availability of the chemical, nutrient
limitations, toxicity of the compound, and inactivation or
nonexistence of the enzyme capable of degrading the compound
(Valentine and Schnoor, 1986).
4.3.5.2.2. Losses by Abiotic Degradation — Chemical
degradation of organic pollutants can be a significant mechanism
of removal of these compounds from soil systems. Hydrolysis and
oxidation-reduction (redox) reactions are the primary chemical
transformation processes occurring in the upper layers of soils
(Valentine, 1986). General rate expressions describing
transformation of some contaminants by all nonbiological processes
are available, and are helpful when division into component
reactions is not possible.
4-16
-------
Hydrolysis in aqueous systems is characterized by three
processes: acid-catalyzed, bcise-catalyzed and neutral reactions
(i.e., H+, OH" and H2O catalyzed hydrolysis). The overall rate of
hydrolysis is the sum of the first-order rates of these processes
(Valentine, 1986). In soil systems, sorption of the chemical can
increase, decrease or not affect its rate of hydrolysis as
numerous studies cited in Valentine (1986) have shown. The total
rate of hydrolysis in soil can be predicted by adding the rates in
the soil and water phases, which are assumed first-order at fixed
pH (Valentine, 1986). Methods for estimating these hydrolysis
constants are described by Lyman et al. (1982).
Organic and inorganic compounds also undergo redox reactions
in the soil (Valentine, 1986). Organic redox reactions involve
exchange of oxygen and hydrogen atoms by the reacting molecules.
For example, glucose is oxidized to CO2 and H2O by oxygen; and the
oxygen is reduced to H2O. Inorganic redox reactions may involve
the exchange of atoms or of electrons by the reactants.
In soil systems where the identities of oxidant and reductant
species are not specified, a first-order rate constant can be
obtained for describing loss by redox reactions (Valentine, 1986).
Redox reactions involving metals do not constitute a degradative
process, but may promote losses by making metals more mobile.
However, the effect of each redox reaction depends on the metal and
on the soil conditions.
4-17
-------
4.3.5.2.4. Losses Due to Volatilization (ksv)
Semivolatile and volatile compounds emitted in high concentrations
may become adsorbed to soil particles and exhibit volatilization
losses from the soil. The loss of a chemical from the soil by
volatilization is dependent on the rate of movement of the chemical
to the soil surface, the chemical vapor concentration at the soil
surface (due to soil conditions) and the rate at which the vapor
is carried away by the atmosphere (due to atmospheric conditions)
(Jury, 1986).
Polar molecules are less volatile than slightly polar or
nonpolar molecules due to greater absorption to soil constituents
and greater water solubility (Poe, 1988) . Volatilization loss
rates for specific chemicals may be found in the literature.
Volatilization rates for hexachlorobenzene have been modeled by
Spencer et al. (1982). Laboratory measured volatilization rates
for soil incorporated chemicals adequately estimate the field
measured rates, because the rates are primarily dependent on soil
factors that can be mimicked in the laboratory (Spencer et al.,
1982). Such experimental values may be used if site-specific data
are unavailable.
4.4. EXAMPLE CALCULATIONS
In determining soil concentration, the time of deposition
(i.e., the number of years of continual combustor emissions) and
the distance of the individual from the incinerator differ for each
exposure scenario. In Scenario A the duration of deposition is 30
years, the annual deposition value is calculated as the areal
4-18
-------
average over, the 50 km "ring" surrounding the incinerator, and the
exposed individual is within this 50 km radius. In Scenario B, the
individual is 5 km from the combustion source exposed to the areal-
v
averaged annual deposition for the 5 km ring and the duration of
deposition is 60 years. Scenario C defines the individual as
residing at the point of maximal deposition at.a distance of 0.2
km from the incinerator and the duration of deposition as 100
years. This section illustrates the soil concentration for the
three exposure scenarios (A,B,C) for both cadmium and
benzq(a)pyrene (B(a)P).
4.4.1. Cadmium. The values chosen for input variables to
determine soil concentration of cadmium are found in Table 4-2.
In determining soil concentrations, the soil loss constant of
cadmium due to leaching is first calculated. For surface depth (Z
= 1 cm), the soil loss constant (ksl) can be calculated according
to Equation 4-3:
ksl =
120.0 cm/yr + 25.0 cm/yr - 100.0 cm/yr
0.22 mL/cm3 • 1 cm • [1.0 + (1.5 g/cnr • 500 mL/g/0.22 mL/cnr1) ]
-1
= 0.06 yr
The soil loss constant, due to all processes, for cadmium at
surface depth (Z = 1 cm) can be determined by using Equation 4-2:
ks = 0.06 yr"1 + 0 yr"1 + 0 yr"
= 0.06 yr
-1
4-19
-------
TABLE 4-2
Input Variables for Determination of Cadmium
Soil Concentrations
Exposure Scenario
Variable
P (cm/yr)
I (cm/yr)
Ev (cm/yr)
6 (mL/cm3)
Z (cm)
Surface depth
Till depth
BD (g/cm3)
Kd (mL/g)
ksg (yr'1)
ksv (yr~1)
Dyd (g/m2/yr)
Dyw (g/m2/yr)
Tc (yrs)
A
120.0
25.0
100.0
0.22
1
20
1.5
500
0
0
2.34X10'5
0.66X10"5
30
B
120.0
25.0
100.0
0.22
1
20
1.5
500
0
0
1.59X10"4
2.17X10"4
60
C
120.0
25.0
100.0
0.22
1
20
1.5
500
0
0
0.09X10"2
1.10X10"2
100
4-20
-------
The soil loss constant for cadmium due to leaching at till
depth (Z = 20 cm) can be determined using Equation 4-3:
120.0 cm/yr + 25.0 cm/yr - 100.0 cm/yr
ksl =
0.22 mL/cm3 • 20 cm • [1.0 + (1.5 g/cm3 • 500 mL/g/0.22 mL/cm ) ]
= 0.003 yr
-1
The soil loss constant, due to all processes, for cadmium at
till depth (Z =20 cm) can be calculated according to Equation 4-
2:
ks = 0.003 yr'1 + 0 yr"1 + o yr"1
= 0.003 yr"1
In calculating the soil concentration in Scenario A, it is
assumed that the incinerator has been operating continuously for
30 years (Tc = 30 yrs) . The soil concentrations at the surface and
till depths can be determined by using Equation 4-1. For surface
depth (Z = 1 cm), soil concentration is:
-1
Sc =
(2.34X10"5 + 0.66x10° ) • [ 1.0-exp (-0.06 yr"' • 30 yrs)] • 0.1
1 cm • 1.5 g/cm3 • 0.06 yr"1
= 2.78x10 mg Cd/g soil
4-21
-------
Soil concentration for till depth (Z = 20 cm) is:
Sc
(2.34x10 3 + 0.66x10°) • [1.0-exp(-0.003 yr"1 • 30 yrs)] • 0.1
20 cm • 1.5 g/cm • 0.003 yr
-1
Sc
» 2.87xlO"6 mg Cd/g soil
In Scenario B, it is assumed that the incinerator has been
operating continuously for 60 years. The soil concentration for
both surface depth and till depth can be calculated using Equation
4-1. For surface depth (Z= l cm) soil concentration is:
(1.59x10"* + 2.17X10"4) • [1.0-exp(-0.06 yr"1 • 60 yrs)] • 0.1
1 cm • 1.5 g/cm3 • 0.06 yr"1
4.06xlO"4 mg Cd/g soil
For till depth (Z = 20 cm), soil concentration is:
Sc =
(1.59X10"4 + 2.17X10"4) • [1.0-exp(-0.003 yr"1 • 60 yrs)] -0.1
20 cm • 1.5 g/cm • 0.003 yrs
-1
= 6.88X10"5 mg Cd/g soil
In Scenario C, the combustion source has been operating for
100 years. The soil concentration of cadmium in Scenario C for
surface and till depth can be calculated using Equation 4-1. For
surface depth (Z = 1 cm), soil concentration is:
4-22
-------
.-2
-1
Sc =
(0.09x10"* '+'1.10x10"*) • [1.0-exp(-0.06 yr"1 • 100 yrs) ] • 0.1
1 cm • 1.5 g/cm3 • 0.06 yr"1
,-2
= 1.32x10 mg Cd/g isoil
For till depth (Z = 20 cm), soil concentration is:
Sc =
(0.09xlO"2 + l.lOxlO"2) • [1.0-exp(-0.003 yr"1 • 100 yrs)] • 0.1
20 cm • 1.5 g/cm3 • 0.003 yr"1
= 3.43x10 mg Cd/g soil
4.4.2. Benzo(a)pyrene (B(a)P). The values chosen for input
variables to determine soil concentration of benzo(a)pyrene are
found in Table 4-3.
The soil loss constant of benzo(a)pyrene due to leaching for
surface depth (Z = 1 cm) can be calculated according to Equation
4-3:
ksl =
120.0 cm/yr + 25.0 cm/yr - 100.0 cm/yr
0.22 mL/cm3 • 1 cm • [1.0 + (1.5 g/cm3 • 1.2xl05 mL/g/0.22 mL/cm3) ]
= 2.5xlO"4 yr"1
The soil loss constant of benzo(a)pyrene due to leaching at till depth
(Z = 20 cm) can be determined using Equation 4-3:
ksl =
120.0 cm/yr +25.0 cm/yr - 100.0 cm/yr
0.22 mL/cm3 • 20 cm • [1.0 + (1.5 g/cm3 • 1.2xl05 mL/g/0.22 mL/cm3)]
= 1.25X10"5 yr"1
4-23
-------
TABLE 4-3
Input Variables for Determination of Benzo(a)Pyrene
Soil Concentrations
Variable
P (cm/yr)
I (cm/yr)
Ev (cm/yr)
0 (mL/cm3)
Z (cm)
Surface depth
Till depth
BD (g/cm3)
Kd (mL/g)
ksg (yr'1)
Dyd (g/m2/yr)
Dyw (g/m2/yr)
Tc (yrs)
Ma (g/cm-s)
pa (g/cm3)
D, (cm2/s)
u (m/s)
-------
The loss constant of benzo(a)pyrene due to volatilization
(ksv) requires that two variables be calculated, the gas phase mass
transfer coefficient and the equilibrium coefficient. The Schmidt
number (N) , needed as an input for the equilibrium coefficient, can
be calculated using Equation 4-7:
184 g/cm-sec
N =
1.19X10"3 g/cm3 • 0.043 cm2/s
= 3.59x10°
The gas phase mass transfer (Kt) can be determined using
Equation 4-6:
Kt = 0.482 • (3.9 m/s)°-78 • (3.59xl06) "°-67 • (200 m)"0'11
= 3.16X10"5 cm/s
The equilibrium coefficient (Ke) for surface depth (Z = 1 cm)
can be determined using Equation 4-5:
Ke =
3.1536X1010 « 1.55x10"° (atm-m3/mole)
1 cm • 1.2X105 mL/g • 0.08205 (L-atm/mole-°K) • 298°K • 1.5 g/cm3
= 1.11x10
-2
The equilibrium coefficient (Ke) at till depth (Z = 20 cm) can
be calculated using Equation 4-5:
Ke =
3.1536X1010 • 1.55x10"° (atm-m3/mole)
20 cm • 1.2xl05 mL/g • 0.08205 (L-atm/mole-°K) • 298°K • 1.5 g/cm3
= 5.5x10
-4
4-25
-------
The soil loss of benzo(a)pyrene due to volatilization (ksv)
at surface depth (Z = 1 cm) can be calculated using Equation 4-4:
ksv = l.llxlO"2 • 3.16X10"5
= 3.51xlO"7 yr"1
The soil loss of benzo(a)pyrene due to volatilization (ksv)
at till depth (Z = 20 cm) can be calculated using Equation 4-4:
ksv = 5.5X10"4 • 3.16X10"5
= 1.74X10"8 yr"1 >
The soil loss constant of benzo(a)pyrene due to degradation
(ksg) is equal to 0.803 yr"1, which is experimentally derived based
on the half-life of benzo(a)pyrene in soil (U.S. EPA, 1988c) . The
amount of benzo(a)pyrene lost to volatilization and leaching is
minimal relative to the amount lost due to degradation.
The soil loss constant of benzo(a)pyrene due to all processes
(ks) at surface depth (Z = l cm) can be calculated using Equation
4-2:
ks - 2.5X10"4 yr"1 + 0.803 yr"1 + S.SlxlO"7- yr"1
= 0.803 yr"1
The soil loss constant of benzo(a)pyrene due to all processes
(ks) at till depth (Z = 20 cm) can be calculated using Equation 4-
2: ' • • •• -
"5 yr"1 + 0.803 yr"1 + l.74xlO"8 yr
= 0.803 yr
ks » 1.25xlO"5 yr
-1
-1
4-26
-------
In calculating the soil concentration in Scenario A, it is
assumed that the incinerator has been operating continuously for
30 years (Tc = 30 yrs) . The soil concentrations at the surface and
till depths can be determined by using Equation 4-1. For surface
depth (Z = 1 cm), soil concentration is:
Sc =
(3.77X10"8 + 2.74X10"8) • [ 1.0-exp(-0.803 yr"1 • 30 yrs) ] • 0.1
1 cm « 1.5 g/cm3 • 0.803 yr"1
5.43X10"9 mg B(a)P/g soil
For till depth (Z = 20 cm), soil concentration is:
Sc =
(3.77xlO"8 + 2.74X10"8) * [l.0-exp(-0.803 yr"1 • 30 yrs)] • 0.1
20 cm • 1.5 g/cm3 • 0.803 yr"1
,-10
= 2.70xlO"lu mg B(a)P/g soil
In Scenario B, it is assumed that the incinerator has been
operating continuously for 60 years. The soil concentration for
both surface depth and till depth can be calculated using Equation
4-1. For surface depth (Z= 1 cm), soil concentration is:
Sc =
(2.45X10"7 + 6.74X10"7) '• [1.0-exp(-0.803 yr"1 • 60 yrs)] • 0.1
1 cm • 1.5 g/cm3 • 0.803 yr"1
= 7.66xlO"8 mg B(a)P/g soil
4-27
-------
For till depth (Z = 20 cm), soil concentration is:
Sc
(2.45X10"7 + 6.74X10"7) • [1.0-exp(-0.803 yr"1 • 60 yrs)] • 0.1
20 cm • 1.5 g/cm • 0.803 yrs
-1
Sc
= 3.81xlO"y mg B(a)P/g soil
In Scenario C, the combustion source has been operating for
100 years. The soil concentration of cadmium in Scenario C for
surface and till depth can be calculated using Equation 4-1. For
surface depth (Z = 1 cm), soil concentration is:
(0.142X10"5 + 2.33X10"5) • [1.0-exp(-0.803 yr"1 • 100 yrs)] • 0.1
1 cm • 1.5 g/cm • 0.803 yr
-1
1-6
2.06x10° mg B(a)P/g soil
For till depth (Z = 20 cm), soil concentration is:
Sc
(0.142x10"' + 2.33xlO"s) • [1.0-exp(-0.803 yr"1 • 100 yrs)] • 0.1
20 cm • 1.5 g/cm • 0.803 yr
-1
- 1.02X10"7 mg B(a)P/g soil
The soil concentrations determined above and shown in Table
4-4 are used in subsequent chapters as input variables for
determination of exposure by various pathways.
4-28
-------
TABLE 4-4
Summary of Soil Concentrations
(mg/g)
Cadmium
Benzo(a)pyrene
Scenario A:
1 cm depth
20 cm depth
Scenario B:
1 cm depth
20 cm depth
Scenario C:
2.78x10
2.87x10
-5
-6
4.06x10
6.88X10
-4
-5
5.43x10
2.70x10
-9
-10
7.66X1Q"8
3.81X10"9
1 cm depth
20 cm depth
1.32X10"2
3.43X10"3
2.06X10"6
1.02X10"7
4-29
-------
-------
5. DETERMINING EXPOSURE THROUGH
THE TERRESTRIAL FOOD CHAIN
5.1. INTRODUCTION
The previous chapter describes how contaminants associated
with combustor emissions are incorporated into soil. This chapter
describes how contaminants may enter the terrestrial food chain
(TFC) from the soil or directly from the atmosphere. It also shows
how daily exposure to food chain contaminants is calculated and
used to estimate human risk. The model assumes that contaminants
enter the food chain through either uptake by plants or direct soil
ingestion by animals. Humans may also directly ingest soil;
however, this pathway of exposure is discussed in Chapter 6. The
remaining sections of this chapter discuss methods and important
i
parameters for calculating pollutant concentrations in plant and
animal tissues and determining human daily intake from foods.
Example calculations for benzo(a)pyrene and cadmium are also
presented.
5.2. CALCULATING CONCENTRATION OF POLLUTANT IN PLANTS
This chapter examines three mechanisms by which pollutants
can bioaccumulate in plants: uptake by roots, direct deposition on
exposed plant tissues, and air-to-plant transfer of vapor-phase
pollutants. The relative importance of each mechanism depends in
part on which of three broad categories a plant belongs to: leafy
vegetables, exposed produce, or protected produce. Available
research on mechanisms of pollutant uptake by the three routes is
limited. Questions exist as to whether pollutants accumulated by
5-1
-------
direct deposition and air-to-plant transfer remain on the plant
surface or are internalized by plants. In addition, the degree to
which pollutants are transported between plant organs through the
vascular system is uncertain. Thus, uptake via direct deposition
and air-to-plant transfer is calculated for leafy and exposed
produce but not protected produce since the latter is not in direct
contact with the air.
Yang and Nelson (1986) describe leafy vegetables, such as
lettuce, as having a "broad flat leaf for the direct interception
of atmospherically deposited material." Exposed produce, such as
fruits, has a smaller exposed surface area per unit mass than leafy
produce although it still intercepts atmospherically deposited
particles. Protected produce, such as potatoes or vegetables in
pods, is not directly exposed to atmospherically deposited
material. For this plant category, accumulation is most likely to
occur by root uptake.
The food chain model calculates the pollutant concentration
in each plant due to the appropriate mechanism and sums them to
obtain the total pollutant concentration:
where:
Pr.-
Pd,
Pvf
Pf = Pr,- + Pd,- + Pv,
(Equation 5-1)
total concentration of pollutant in the ith plant
group (jug/g)
concentration of pollutant in i plant group due
to root uptake (jiig/g)
concentration of pollutant in ith plant group due
to direct deposition (/xg/g)
concentration of pollutant in ith plant group due
to air-to-plant transfer
5-2
-------
Section 5.5 presents calculations of Pf for seven plant groups
consumed by humans (grains, legumes, potatoes, root vegetables,
fruiting vegetables, leafy vegetables, fruits) and two plant groups
consumed by animals (grains and forage). Grains, potatoes, and
root vegetables are considered to be protected produce; fruits,
fruiting vegetables, and legumes are considered to be exposed
produce; and leafy vegetables and forage are considered leafy
produce. Risk assessors may want to consider site-specific data
when developing appropriate plant groups. Note that a separate P?
must be calculated for each pollutant in each plant group to be
considered in the risk assessment.
5.2.1. Plant Pollutant Concentration Due to Root Uptake (Pr{).
The concentration of pollutant in plant tissue due to root uptake
is determined from the soil concentration and the plant-soil
bioconcentration factor for each plant group:
where:
Pr, = Sc
Br,-
(Equation 5-2)
Pr{ = concentration of pollutant in i plant group due
to root uptake (/Ltg pollutant/g plant tissue, dry
weight [DW])
Sc = soil concentration of pollutant after the total
period of deposition (/xg pollutant/g soil)
Br,
th
= plant-soil bioconcentration factor for the i
plant group (|>g pollutant/g plant tissue DW]/[/ig
pollutant/g soil])
5.2.1.1. SOIL CONCENTRATION (Sc) — Methods for calculating
soil concentration are presented in Chapter 4. When considering
root uptake, soil concentration should be calculated assuming an
5-3
-------
incorporation depth of 20 cm since the ground is usually tilled to
this depth to plant food and grain crops. Even for crops, such as.
pasture grasses grown on untilled soil where pollutants are assumed
to be within the upper 1 cm of soil, plant roots penetrate well
beyond this depth into relatively uncontaminated soil. Therefore,
a soil depth of 20 cm is also used in these cases to estimate .the
average soil concentration in a 20-cm root zone.
5.2.1.2. BIOCONCENTRATION FACTORS AND PLANT UPTAKE (Br,-) --
The plant-soil bioconcentration factor is a measure of a
pollutant's ability to accumulate in plant tissue and is defined
as the pollutant concentration in the plant . divided by the
pollutant concentration in the soil. Bioconcentration factors may
be derived differently for inorganic and organic pollutants, but
they are generally dependent on the bioavailability , of the
pollutant in the soil.
5.2.1.2.1. Inorganics -- Data on uptake of inorganics by
plants may be available in the form of bioconcentration factors,
which are discussed in Baes et al. (1984). However, in some cases,
accumulation of inorganics is measured by "uptake response slopes"
rather than by bioconcentration factors.
The term "uptake response" is used to denote the difference
between pre- and post-exposure plant tissue concentrations.
Uptake response slopes can be derived from any data set where
changes in tissue concentration and pollutant application rate have
been determined; the units are often expressed as
5-4
-------
DW)/(kg/ha). Uptake response slopes can be converted to
bioconcentration factors by multiplying by a conversion factor of
2.7xl03 Mg/ha, which is the dry mass of soil in the upper 20 cm if
a typical bulk density of 1.35 g/cm3 is assumed.
Uptake rates of metals by plants, reviewed by CAST (1980),
Ryan et al. (1982), and Logan and Chaney (1983), are influenced by
application rate, soil pH and soil organic matter content. Ryan
et al. (1982) used linear regression of plant tissue cadmium
concentration against cadmium application rate to derive uptake
response slopes for various crops. The authors stated that, given
constant soil conditions, cadmium plant tissue concentration had
a highly significant positive linear correlation with the amount
of cadmium added to the soil.
More recent work, as reviewed by Page et al. (1986) , has shown
that plant response to metals from sludge-amended soil is
curvilinear, approaching a plateau concentration in tissue as
sludge application rate increases. However, metal-adsorptive
materials present in the sludge matrix are thought to be
responsible for this effect. Since no such uptake-limiting effect
has been demonstrated for deposited metals, uptake response slopes
will be assumed to be linear for this methodology. Therefore,
plant concentration and resulting dietary intake of a contaminant
are assumed to increase continually with deposition if some or all
of the diet originates from contaminated soils. This assumption
is conservative but reasonable given the lack of data regarding the
saturability of plant tissues.
5-5
-------
Soil pH is probably the most important characteristic
influencing metal retention, and thus availability, by soils
(LaBauve et al., 1988). The pH of soils varies greatly, from 2 or
less in acid soils to 9 or more in alkaline soils (Hillel, 1980).
Agricultural soils are usually limed to maintain the pH at
approximately 6.5, the pH at which plant nutrients are most readily
available for plant uptake (Bohn et al., 1985).
Metal cations exhibit a pH-dependent adsorption behavior,
probably due to several different adsorption mechanisms (Elliott
et al., 1986). At pH less than 6.0, some metal cations tend to be
more mobile. For example, cadmium mobility tends to increase as
soil pH decreases; lead is less pH-dependent (Harter, 1983).
As soil pH increases, generally above 6.0, metal cations are
adsorbed more strongly and become less available for uptake by
plants. Several studies have demonstrated that, generally, plant
uptake of metals decreases with increasing pH (Lagerwerff, 1971;
Ryan et al., 1982; Giordano et al., 1983). However, the effects
of liming acid soils to increase pH are unclear. While Giordano
et al. (1983) and Ryan et al. (1982) report decreased cadmium
concentration in plants after soil has been limed to pH 6.5,
several studies cited in CAST (1980) indicate that using lime to
raise soil pH from 5.0 to 6.6 had little or no effect on metal
uptake by plants.
Ideally, the risk assessor should use site-specific data
because local soil pH and organic matter content greatly influence
the uptake response. If literature data are used, the data should
be from studies in which the pH and organic matter content of the
5-6
-------
experimental soils are similar to those of soils at the site.
Where evidence suggests that response may be other than linear, a
slope derived from application rates (or soil concentrations)
similar to those predicted for the site in question should be
preferred. Also, uptake response data should be derived from
studies of deposited emissions, if possible. For contaminants
lacking such data, uptake response slopes derived from other types
of chemical application (such as sludge or pesticide additions) can
be assumed to apply to deposited contaminants as well.
5.2.1.2.2. Organics — A linear uptake response is also
assumed for organic chemicals, but because most of the compounds
of concern are xenobiotics, tissue concentration can be assumed to
be zero when soil concentration is zero. Therefore, a
bioconcentration factor, which can be derived from a single data
pair, is used more often for organics than for inorganics. Such
a factor can be derived from any data set that includes both tissue
concentration (ng/g DW) and soil concentration (Mg/g) •
As for inorganics, however, the most accurate factor would be
derived from interpolation of values for soil concentration similar
to those predicted for the site of concern. Ideally, factors
determined using local soil data should be used. However, if
literature data are used, the experimental soils should be of
similar type, pH, and organic matter as soils on the site to the
greatest extent possible.
For organics, uptake from soil and transport to above-ground
plant parts is dependent on the solubility of a chemical in water,
5-7
-------
which is inversely proportional to the octanol-water partition
coefficient (KOH) (Travis and Arms, 1988) . Thus, the
bioconcentration factor can be calculated from the following
equation if literature data are unavailable:
log Br, = 1.588 - 0.578 log K0
(Equation 5-3)
This equation is based on bioconcentration factors for 29 chemicals
in vegetative tissues (Travis and Arms, 1988).
The bioconcentration factors for cadmium and benzo(a)pyrene
used in the example calculations are presented in Belcher and
Travis (1989). These factors are available for calculating uptake
into potatoes, leafy vegetables, legumes, root vegetables, fruiting
vegetables, grain and forage. The bioconcentration factor for
fruiting vegetables was assumed to apply to fruits also because
both plant groups belong to the same broad group of exposed
produce. However, if site-specific data concerning uptake into
fruits is available, this data should be used.
5.2.2. Plant Pollutant Concentration Due to Direct Deposition
(Pdj) . Pollutants are deposited on exposed plant .surfaces by both
dry and wet deposition. Pollutant concentration due to deposition
is calculated for leafy plants and exposed produce but not for
protected produce (Yang and Nelson, 1986), which is not directly
exposed to deposition. Pd contributes more to the total
concentration of leafy produce than exposed produce because leafy
produce has a larger surface area with which to intercept direct
5-8
-------
deposition (Yang and Nelson, 1986; Shor et al., 1982).
Concentration of pollutant in plant tissue due to direct deposition
onto plant surfaces is calculated as:
Pd,- = 1000 • [Dyd + (Fw • Dyw) ] • Rp,. • [i.o - exp(-kp • Tp,-)]
YPi • kp
(Equation 5-4)
where:
Pd; =
1000
Dyd
Fw
Dyw
HP,
kp
TPi
Ypf =
concentration of pollutant due to direct
deposition in the ith plant group (/zg pollutant/g
plant tissue DW)
conversion factor, kg/103 g and 106 /ig/g pollutant
yearly dry deposition rate (g pollutant/m2/yr)
fraction of wet deposition that adheres to plant
surfaces (unitless)
yearly wet deposition rate (g pollutant/m2/yr)
interception fraction of the edible portion of
plant tissue for the ith plant group (unitless)
plant surface loss coefficient (yr )
length of plant's exposure to deposition per
harvest of the edible portion of the ith plant
group (yrs)
yield or standing crop biomass of the edible
portion of the ith plant group (kg DW/m2)
5.2.2.1. YEARLY DEPOSITION RATES (Dyd and Dyw) — Refer to
Chapter 3 to find methods for determining both dry and wet yearly
deposition averaged over the area surrounding the combustor.
5.2.2.2. FRACTION OF WET DEPOSITION ADHERING (Fw) — All of
the pollutants that come in contact with plant surfaces as a result
of dry deposition are assumed to remain on the plant surfaces until
removed by weathering. However, only a fraction of pollutant
material that intercepts plant surfaces as a result of wet
5-9
-------
deposition adheres; the remainder is assumed to wash off almost
immediately. Fw is a measure of the amount of wet deposition that
adheres to the plant surface. Fw is assumed to be a constant of
2% for all pollutants and plant groups (Belcher, 1989).
5.2.2.3. INTERCEPTION FRACTION (Rp,-) — Baes et al. (1984)
define the interception fraction as a "factor which accounts for
the fact that not all of the airborne material depositing within
a unit area will initially deposit on edible vegetation surfaces."
Models developed for the Nuclear Regulatory Commission assumed the
interception fraction was a constant of 0.2 for dry and wet
deposition of particles (Boone et al., 1981). However, Shor et al.
(1982) suggest that the diversity of plant growth indicates the
need for vegetation-specific interception fractions. Baes et al.
(1984) presents equations for interception fractions for pasture
grasses, leafy vegetables, and exposed produce. Based on an
empirically derived relationship to standing crop biomass,
interception fractions for pasture grasses can be calculated as:
Rpg = l-exp(-2.88 • Ypj) (Equation 5-5)
where:
Rpg = interception fraction for pasture grasses
(unitless)
Yp? = standing crop biomass for pasture grasses (kg
DW/m2)
Interception fractions for mature leafy vegetables can be
calculated as:
nr rn 7rrf2
[(np-l)d + 2rf][(rn-l)dr + 2rf]
(Equation 5-6)
5-10
-------
where:
n
= interception fraction for mature leafy vegetables
(unitless)
= number of plants per row
= number of rows of plants
= radius of individual fruit or plant (mm)
= distance between plants in a row (mm)
= distance between rows of plants (mm)
The average interception fraction for mature leafy vegetables was
calculated to be 0.3. Interception fractions for exposed produce
is based on the five most commercially important non-citrus fruit
and field crops (apples, snap beans, tomatoes, peaches, cherries;
Baes et al., 1984) and is calculated by determining the total fruit
cross-sectional area per unit area:
Rrf = n7rrf2
Iw
(Equation 5-7)
where:
R,
n
rf
w
= interception fraction of round fruit (unitless)
= number of fruit per unit area
= radius of the fruit (mm)
= length of the unit area (mm)
= width of the unit area (mm)
For long rather than round exposed produce (i.e., snap beans), the
above equation is modified to:
where:
R
if
Rlf = n2rflf
Iw
(Equation 5-8)
interception fraction of long fruit (unitless)
length of the long fruit (mm)
5-11
-------
Interception fractions for leafy vegetables and exposed produce
can also be calculated based on an empirical relationship to
standing crop biomass (Yp{) as for pasture grasses:
R
lv
= 1-exp(-0.0846 • Yp,)
Re = 1-exp (-0.0324 • Ypf)
(Equation 5-9)
(Equation 5-10)
where:
R
tv
= interception fraction for leafy vegetables
(unitless)
= interception fraction for exposed produce
(unitless)
Risk assessors may wish to choose the appropriate set of equations
for interception fraction based on the availability of site-
specific data. In addition, values for number .of plants per row,
number of fruit per square meter, etc. should be selected from
site-specific data.
5.2.2.4. PLANT-SURFACE LOSS COEFFICIENT (kp) — Once
contaminants are deposited onto plant surfaces, several
environmental processes, including wind removal, water removal,
and growth dilution, work to reduce the amount of contamination on
the plant surface (Miller and Hoffman, 1983). The plant-surface
loss coefficient is a measure of the amount of contaminant lost to
these processes over time. Miller and Hoffman (1983) give the
following equation to calculate kp:
kp = In2
• 365 days/yr
(Equation 5-11)
5-12
-------
where:
kp = plant-surface loss coefficient (yrs"1)
t.% = environmental half-time (days)
In radiologic assessments, a half-time value of 14 days has
generally been assumed to represent all radionuclides and plant
types (Miller and Hoffman, 1983). However, Miller and Hoffman
(1983) report half-time values ranging from 2.8 to 34 days for all
contaminants on herbaceous vegetation. These half-time values
correspond to kp values ranging from 7.44 to 90.36 yrs"1. The half-
time of benzo(a)pyrene in air is reported to range from 1-6 days
(U.S. EPA, 1986d); values of kp calculated from those half lives
-1
-1
range from 42-253 yr . The value of 126.5 yr was selected from
this range to represent a typical value (Belcher and Travis, 1989).
5.2.2.5. LENGTH OP PLANT EXPOSURE (Tp) — This variable is
defined as the amount of time the edible part of the plant is
exposed to direct deposition. Tp is treated as a constant and is
based on an average period between successive hay harvests (60
days) and an average period between successive grazing by cattle
(30 days) (Belcher and Travis, 1989). For animal forage, these
two time periods are averaged (45 days) and then divided by 365
days to give a Tp of 0.123 yrs. For the remaining animal foods,
time between successive hay hetrvests is divided by 365 days to give
a Tp of 0.164 (Belcher and Travis, 1989). For lack of appropriate
data, a Tp of 0.164 is also used in this document for foods
5-13
-------
consumed by humans. However, risk assessors may want to consider
local growing seasons to develop site-specific Tp values for foods
consumed by humans.
5.2.2.6. STANDING CROP BIOMASS (Yp{) — As may be seen from
Equations 5-4 through 5-10, plant pollutant concentration due to
direct deposition is influenced by the standing crop biomass, Yp,-.
The best estimate of Yp,- is productivity (Baes et al., 1984; Shor
et al., 1982), which is defined as:
Yh5
(Equation 5-12)
YP,
P =
^i
Ah,-
where:
th
P- = productivity of the i crop (kg DW/m )
Yh. « harvest yield of ith crop (kg DW)
Ah- = the area planted to crop i that is harvested
(m2)
Belcher and Travis (1989) have calculated average productivity
values for several animal and human foods; 'these values are
presented in Table 5-1. In addition, Shor et al. (1982) present
average productivity values of leafy vegetables, exposed produce,
protected produce, grain for human consumption, honey, pasture,
hay, silage, and animal grain for all U.S. counties. The data in
Shor et al. (1982) are presented as fresh weight and can be
converted to dry weight using the water content values presented
in Baes et al. (1984).
5-14
-------
TABLE 5-1
Average Values of Productivity (kg DW/m2)
Plant
Group
Animal Forage
Animal Grain
Potatoes
Leafy Vegetables
Root Vegetables
Garden Fruits
Legumes
Range
0.02-0.75
0.14-0.45
0.41-0.56
0.09-0.35
0.09-0.44
0.01-0.25
0.08-0.13
Arithmetic
Mean
0.31
0.30
0.48
0.18
0.34
0.09
0.10
Source: Belcher and Travis, 1989
5-15
-------
5.2.3. Plant Pollutant Concentration Due to Air-to-Plant
Transfer (Pv,,). In addition to the root uptake and deposition
routes, plants can accumulate vapor-phase contaminants by air-to-
plant transfer. Air-to-plant transfer is likely to be important
for leafy plants and, to a lesser extent, exposed produce.
However, this pathway need not be calculated for protected produce.
Two parameters are necessary to estimate air-to-plant
transfer: the air-to-plant biotransfer factor and the atmospheric
concentration of pollutant at ground level. The latter is the sum
of the concentration of pollutant in the vapor phase due to
emissions from the combustor and the atmospheric concentration due
to volatilization of pollutant deposited on the soil:
Pv,. = [ (Fv • Cy) + Cv] • Bv,- (Equation 5-13)
where:
Pv, =
Fv —
Cy =
Cv =
BV{
pa
pa
concentration of pollutant due to air-to-plant
transfer in the ith plant group (fj.g pollutant/g
plant tissue DW)
fraction of pollutant air concentration present
in the vapor phase (unitless)
concentration of pollutant in air due to direct
emissions (/ig pollutant/m3 air)
concentration of pollutant in air due to
volatilization from soil (jug pollutant/m3 air)
air-to-plant biotransfer factor for the ith plant
group [ (Atg pollutant/g plant tissue DW)/(/xg
pollutant/g air)]
density of air (1190 g/m3 at 25° C)
The example calculations in Section 5.5 demonstrate that Cv can
be up to seven or eight orders of magnitude lower than Cy. Thus,
5-16
-------
Cv will usually make an insignificant contribution to the overall
concentration in plants due to air-to-plant transfer.
5.2.3.1 AIR CONCENTRATION DUE TO EMISSIONS (Cy) — Refer
to Chapter 3 to find methods for determining ground level ambient
air concentrations averaged over the area surrounding the
combustor.
5.2.3.2. FRACTION OF POLLUTANT IN VAPOR PHASE (Fv) — Using
Fv accounts for the fact that organic compounds can exist in the
atmosphere in both the vapor and particulate phase, and only the
vapor concentration is considered for this pathway. The ratio of
vapor to particle depends on the chemical properties of the
pollutant and the total suspended particulate concentration
(Bidleman, 1988).
Junge (as reported in Bidleman, 1988) developed a model for
estimating the adsorbed fraction of contaminants; this model has
been modified to determine Fv as shown below:
Fv
where:
Fv
c
ST
P'L
= 1 - [cST/(p\ + cST)]
(Equation 5-14)
= fraction of pollutant in vapor phase (unitless)
= assumed constant of 1.7xlO"4 (atm-cm)
= Whitby's average total surface area of particles
(cm2/ cm3)
= liquid-phase vapor pressure (atm)
A detailed discussion of these parameters is presented in
Bidleman (1988).
5-17
-------
Bidleman (1988) calculated the percentage of particulate for
several semivolatile organic compounds as a function of liquid-
phase vapor pressure, assuming an ST of 3.5xlO"6 cm2/cm3 for a
background plus local sources particulate situation. Based on
Bidleman's calculation, this document assumes that the Fv for
benzo(a)pyrene is 0.18. Note that this value is highly uncertain
because it is derived from a single calculation, and correlation
of the model with field studies is still in progress.
5.2.3.3. AIR-TO-PLANT BIOTRANSFER FACTORS (Bv) — The air-
to-plant biotransfer factor is defined as the ratio of contaminant
concentration in aerial plant parts (fJ.g/g DW) to the concentration
of pollutant in air (ftg/g) (Travis and Hattemer-Frey, 1988) .
Little is known about the mechanisms of vapor uptake by leaves, and
there are few experimentally derived air-to-leaf biotransfer
factors available in the literature. Therefore, models to predict
Bv based on empirical correlations with chemical properties have
been developed. Research on azalea leaf uptake of ten organic
vapors has shown that Bv is related to both the Henry's Law
constant (H) and the octanol-water partition coefficient (Kow) in
the following manner (Bacci et al., 1989):
Log(Bv • H) = -0.93 + 1.14 LogK
OH
(Equation 5-15)
where:
Bv
H
K,
ow
air-to-leaf biotransfer ([ug/g DW]/[jug/g])
Henry's Law constant (Pa m/mol)
octanol-water partition coefficient (unitless)
5-18
-------
Thus, lipophilic compounds with a high Henry's Law constant will
tend not to bioaccumulate, but compounds with a high octanol-water
coefficient will tend to bioaccumulate (Bacci et al., 1989).
If appropriate measured values for Bv are not available, risk
assessors can use Equation 5-15 to predict values for Bv. (Note:
Equation 5-15 was derived using H values with the units Pa
and H values with the same units must be used to predict Bv.)
5.2.3.4. AIR CONCENTRATION DUE TO VOLATILIZATION PROM SOIL
(Cv) — Two assumptions were made concerning volatilization of
contaminants from soil: the area over which the pollutant
volatilizes is essentially circular and the soil emits a uniform
amount of pollutant per unit area. Given these assumptions, the
air concentration of pollutant is estimated by a gaussian plume
formulation for an area source:
6 R
[Cv] - X F /
i-1 * 0
2q
bu
exp
where:
Ht
dx
(Equation 5-16)
Cv = concentration of pollutant in air due to
volatilization from soil (/ig pollutant/ m3 air)
Ff = fraction of time of the i* atmospheric stability
(unitless)
R = radius of the contaminated area (m)
q = emission rate of pollutant from soil (jug/m2/s)
Ht = height of plant above soil (m)
[u] = average wind speed above contaminated area (m/s)
CTZI- = gaussian plume dispersion parameter for the
vertical direction for the ith Pasquill atmospheric
stability category (m)
5-19
-------
5.2.3.4.1. Fraction of Time of Atmospheric Stability
(2?j) __ rphis parameter represents the percent of time that each of
the six Pasquill Stability categories is assumed to occur at a
particular site. The Pasquill Stability categories relate the
types of turbulence to weather conditions at a site (Slade, 1968) .
The six categories include A, extremely unstable; B, moderately
unstable; C, slightly unstable; D, neutral; E, slightly stable; and
F, moderately stable. F5 is determined by collecting meteorologic
data at a site for several years. These data are available from
the National Weather Service for some locations.
5.2.3.4.2. Radius of Contaminated Area (R) — This parameter
defines the size of the source being considered to emit pollutants.
In general, risk assessors may want to choose a value of R that
represents the approximate size of the field in which plants being
exposed to volatiles from soil are growing. In the example
calculations/ R is assumed to be 100 m.
5.2.3.4.3. Emission Rate (q) — The emission rate of
pollutant from soil (q) is calculated by:
q = Sc
ksv
BD
3.171X10"4 (Equation 5-17)
where:
q = emission rate of pollutant from soil
pollutant/m2/s)
Sc = soil concentration of pollutant (/ag pollutant/g
soil)
ksv = soil loss constant due to volatilization (yr )
Z = soil depth (cm)
BD — soil bulk density (1.5 g/cm3)
3.171X10"4 = conversion factor, 104 cm2/m2 and 3.171xlO"8 yr/s
5-20
-------
The input values required for calculating emission rate are all
discussed in Chapter 4, where default values are presented. A soil
concentration based on 20 cm depth is assumed for all plants raised
in tilled soil. A soil concentration based on 1 cm is assumed for
forage, which is raised in untilled soil.
5.2.3*4.4° Height of Plant (Ht) — This parameter represents
the average height from the ground for the plants under
consideration. This methodology assumes a generic plant height of
0.168 meters. Risk assessors may obtain site-specific data from
the local extension service.
5.2.3.4.5. Average Wind Speed ([u]) •— Wind speed data are
collected by the National Weather Service at several stations
around the country. Average U.S. wind speeds range from 2.8 to
6.3 m/s. Wind speed data for selected U.S. stations are presented
in Table 5-2. When selecting average wind speed data, assessors
should choose data from the closest city that has similar climate
and topographical characteristics to the site under consideration.
5.2.3.4.6. Gaussian Plume Dispersion Parameter (crz{) —
The gaussian plume dispersion parameter (crzj) is defined as the
standard deviation of the distribution of material in a plume in
the vertical direction (Slade, 1968) and reflects the ability of
5-21
-------
0)
•&
t>
*>
o
4-f
tn
8
o
s
*4
19
1
g
4^
0
fc
o<
c.2
5-22
-------
contaminants to diffuse in the vertical direction. az for each
Pasquill stability category can be estimated by the following
equation:
a.n- = a,-(x)bi (Equation 5-18)
where:
crzi = gaussian plume dispersion parameter for the ith
Pasquill atmospheric stability category (m)
a,- and b,- = empirically determined parameters for each
stability category
x = downwind distance from source (m)
The interval 0 < x < R is divided into a specified number of
smaller segments. This number determines the size of each segment.
For example, if 50 segments were specified for a source with an R
of 100 meters, the interval would be broken down into 50 segments
with a length of 2 meters. The endpoints of each of these segments
correspond to an estimate of x. The values of a and b vary
according to distance and stability category and are presented in
Table 5-3.
5.2.4. Phytotoxicity. Pollutants emitted from combustors may be
toxic to plants at certain plant concentrations. If the
contaminant is phytotoxic, a maximum tissue concentration for a
given crop can be determined based on available phytotoxicity data
and assumed as an upper limit to plant uptake. Maximum
concentrations are those associated with severe yield reduction
(>75%) or plant death, which would preclude pollutant passage up
the foodchain. Phytotoxicity may be altered by soil pK (see
5-23
-------
TABLE 5-3
Parameters Used to Calculate az
Pasquill Category
A
B
C
D
E
F
x (km)
<0.1
0.10-0.15
0.16-0.20
0.21-0.25
0.26-0.30
0.31-0.40
0.41-0.50
0.51-3.11
>3.11
<0.20
0.21-0.40
>0.40
All
<0.30
0.31-1.00
1.01-3.00
3.01-10.0
10.01-30.0
>30.0
<0.10
0.10-0.30
0.31-1.00
1.01-2.00
2.01-4.00
4.01-10.0
10.01-20.0
20.01-40.0
>40.0
<0.20
0.21-0.70
0.71-1.00
1.01-2.00
2.01-3.00
3.01-7.00
7.01-15.0
15.01-30.0
30.01-60.0
>60.0
a
122.8
158.1
170.2
179.5
217.4
258.9
346.8
453.9
**
90.7
98.5
109.3
61.1
34.5
32.1
32.1
33.5
36.7
44.1
24.3
23.3
21.6
21.6
22.5
24.7
27.0
35.4
47.6
15.2
14.5
14.0
14.0
14.8
16.2
17.8
22.7
27.1
34.2
b
0.945
1.05
1.09
1.12
1.26
1.41
1.73
2.12
**
0.932
0.983
1.10
0.915
0.870
0.811
0.644
0.605
0.566
0.512
0.837
0.820
0.757
0.631
0.572
0.505
0.467
0.376
0.296
0.816
0.784
0.685
0.632
0.545
0.465
0.415
0.327
0.274
0.217
Source: U.S. EPA, 1986e
5-24
-------
Section 5.2.1.2.1.); the phytotoxicity data chosen should be
appropriate for the soil pH of site of the assessment.
5.3. CALCULATING CONCENTRATION OF POLLUTANT IN ANIMAL TISSUES
The animals that humans usually consume as food take up
pollutants from the food chain by ingesting plants and ingesting
soil while grazing. The food chain model calculates the
concentration of pollutant in animal tissues by considering the
concentration of pollutant in plants and soil, the quantity of
plants and soil that animals consume, and the biotransfer factor
of each type of animal tissue:
Aj =
(QS * Sc)]
Bai
(Equation 5-19)
where:
Ai
P,-J
. Sc
Bai
= concentration of pollutant in the jth animal tissue
group (/ig pollutant/g animal tissue DW)
= quantity of ith plant group eaten by the j animal
each day (kg plant tissue DW/day)
= total concentration of pollutant in the i* plant
group eaten by the jth animal (/xg pollutant/g plant
tissue DW)
= quantity of soil eaten by the j animal each day
(kg soil/day)
= soil concentration (jitg pollutant/g soil)
= biotransfer factor for the j animal tissue group
(d/kg)
Section 5.5. calculates Aj for each of seven animal tissue
groups that humans are most likely to consume: beef, beef liver,
dairy, pork, poultry, eggs, and lamb. Note that a separate Aj
should be calculated for each pollutant to be considered.
5-25
-------
5.3.1. Quantity of Plants Consumed by Animals (Qp{j) . Animals
raised for food are fed a variety of plants including grain (corn,
oats, wheat, etc.), forage (pasture grass, hay) and silage. In
addition, grazing animals ingest soil that clings to plants. Plant
ingestion rates are generally presented as the amount of plant
tissue (kg dry weight) consumed in one day.
Most of the available data on plant ingestion rates are
limited to cows. Beef cattle have overall ingestion rates ranging
from 6-18 kg DW/day (Hoffman and Baes, 1979; Travis and Arms, 1988;
McKone and Ryan, 1989) with an average of 12 kg DW/day (McKone and
Ryan, 1989). Dairy cattle have overall ingestion rates ranging
from 12-25 kg DW/day (Simmonds and Linsley, 1981; Hoffman and Baes,
1979; McKone and Ryan, 1989) with an average of 17 kg DW/day
(McKone and Ryan, 1989). Calves consume 8 kg DW/day (NAS, 1987).
Boone et al. (1981) have estimated the ingestion rates of seven
types of grain, two types of forage, and two types of silage for
both beef and dairy cattle. For beef cattle, the average grain,
forage and silage consumption is 0.47 kg DW/day, 8.8 kg DW/day and
2.5 kg DW/day, respectively. For dairy cattle, the average grain,
forage and silage consumption is 2.6 kg DW/day, 11 kg DW/day and
3.3 kg DW/day, respectively. These average values are used as
input variables in the example calculations.
Less information about feed consumption rates is available
for other food producing animals. Reported ingestion rates for
hogs range from 3.4 (Boone et al., 1981) to 5.2 kg DW/day for
lactating sows (NAS, 1987). The average of this range, 4.3 kg
DW/day, is used in the example calculations. Based on data in U.S.
5-26
-------
EPA (1982b) , 30% (1.3 kg DW/clay) of the average daily intake for
hogs could be silage and the remaining 70% (3.0 kg DW/day) would
be grain. Because hogs are not grazing animals, they are assumed
not to eat forage.
Reported ingestion rates for chickens range from 0.074 (NAS,
1987) to 0.087 kg DW/day (Boorie et al., 1981). The average of this
range, 0.08 kg DW/day, is used in the example calculations. This
average daily intake is assumed to be all grain because poultry are
not grazing animals. Boone et al. (1981) have presented the
ingestion rates of seven types* of grain for both hogs and chickens,
if this information is required for a site-specific assessment.
Ingestion rates for sheep range from 1.2 (NAS, 1987) to 1.6
kg DW/day (Belcher and Travis, 1989). The average of this range,
1.4 kg DW/day, is used in the example calculations. No data
concerning the types of feed sheep consume are available.
Therefore, sheep are assumed to eat only forage because forage has
the highest pollutant concentration of plants eaten by animals.
5.3.2. Quantity of Soil Consumed by Animals (QSj). Soil ingestion
is of concern for grazing animals, such as cows and sheep, that may
ingest plant roots and adhering soil while grazing. Animals could
be exposed to pollutants in combustor emissions soon after
deposition on soil. In addition, pastures where animals graze are
irregularly tilled or untilled, so contaminants are not likely to
become incorporated. Therefore, for this pathway, all of the
5-27
-------
deposited contaminant is assumed to be within the uppermost 1-cm
soil layer, and ingested soil is assumed to originate from the same
1-cm layer.
Soil ingestion rates for animals are usually presented as a
percent of the daily food intake. Reported soil ingestion rates
for cattle are 3-6% (Simmonds and Linsley, 1981) and 1-18%
(Thornton and Abrams, 1983) of dry matter intake. Resulting daily
soil intake for cattle could range from 0.17 to 3.1 kg/day based
on an average dry matter intake of 17 kg/day in dairy cattle.
McKone and Ryan (1989) report daily soil intake for cattle ranging
from 0.1 to 0.7 kg/day with an average of 0.4 kg/day. Thornton and
Abrams (1983) note seasonal variation in soil ingestion rates; soil
ingestion is highest in early spring when there is limited forage.
In this document, approximately 3% of the forage intake for beef
and dairy cattle will be used in the example calculations for the
quantity of soil consumed.
Reported soil ingestion rates for sheep are 20% (Simmonds and
Linsley, 1981) and up to 30% for sheep during winter months when
forage is reduced (Thornton and Abrams, 1983). Resulting daily
soil intake for sheep is 0.28-0.42 kg/day based on an average dry
matter intake of 1.4 kg/day. In this document, approximately 20%
of the forage intake for sheep will be used in the example
calculations for the quantity of soil consumed.
Although hogs and poultry are not grazing animals, these
animals are usually raised outdoors and will undoubtedly ingest
5-28
-------
some soil when they are raised for home consumption. However, this
document will assume no soil ingestion for hogs or poultry because
appropriate data are not available.
5.3.3. Biotransfer Factors (Bap.
Biotransfer factors are
defined as the ratio of pollutant concentration in animal tissue
to the daily intake of pollutant by the animal. These factors are
a measure of the degree to which contaminants are transferred from
the environment (food and water) to specific animal tissues; as
such, they are chemical and tissue specific. Transfer factors for
several inorganic compounds in beef, pork, lamb, poultry, and eggs
are presented in Ng et al. (1982) and summarized in Table 5-4.
Often uptake data in animal tissue are presented in the
literature as the ratio of contaminant concentration in animal
tissue to contaminant concentration in feed ([/xg contaminant/kg
tissue DW]/[/zg contaminant/kg food DW]) . The biotransfer factor
can be determined by dividing the concentration ratio by the animal
feed ingestion rate (kg food DW/day) (Ng et al., 1982; Belcher and
Travis, 1989). If literature data are presented as wet weight,they
can be adjusted using water content value of 0.6 for beef muscle
and 0.9 for cows milk (Baes et al., 1984). Water content values
are also presented in U.S. Department of Agriculture (1975).
The biotransfer factor of an organic compound is directly
proportional to the compound's octanol-water partition coefficient,
Kow (Travis and Arms, 1988) . Therefore, the KOH can be used to
estimate biotransfer factors for organic compounds if experimental
5-29
-------
TABLE 5-4
Biotransfer Factors for Selected Metals
Metal
Animal Tissue
Biotransfer Factor (d/kg)
Cadmium8
Copper"3
Ironb
Leadb
Magnesium6
Manganese6
Selenium6
Zincb
* Source: Belcher
Source: Ng et al
Beef
Beef liver
Lamb
Pork
Poultry
Dairy
Eggs
Beef
Pork
Lamb
Poultry
Eggs0
Beef
Pork
Lamb
Poultry
Eggs0
Beef
Beef
Beef
Pork
Lamb
Poultry
Eggsc
Pork
Poultry
Eggs0
Beef
Pork
Lamb
Poultry
Eggs0
and Travis, 1989
., 1982
0 to 0.12
0.045 to 0.379
0.003 to 0.06
0 to 0.12
1.36 to 36.57
0.0036 to 0.0076
0.489 to 0.85
0.005 to 0.012
0.014 to 0.029
0 .039
0.51
0.72
0.012 to 0.024
0.026
0.073
1.5
2.0
0.0007
0.023
0.0006
0.002 to 0.005
0.002 to 0.012
0.01 to 0.068
0.031 to 0.13
0.32
1.5 to 6.8
2.0 to 7.5
0.035 to 0.2
0.14 to 0.24
4.1
6.5
4.6
c Units for eggs are d/kg egg contents
5-30
-------
data are not available. The relationship between KOH and
biotransfer factor for beef and milk can be expressed as:
and
log Ba(beef) = -7.6 + log KOH
log Ba(milk) = -8.1 + log Kow
(Equation 5-20)
(Equation 5-21)
Similar equations for other animal tissues are not available and
the applicability of these equations to other types of animal
tissues is not known.
5.4. CALCULATING HUMAN DAILY INTAKE
Human daily pollutant intake from consumption of contaminated
plants is calculated by multiplying the concentration of pollutant
in each plant group by the amount of contaminated plant group
consumed daily. Similarly, daily pollutant intake from the
consumption of meat, dairy products, or eggs from animals that have
fed upon contaminated forage, grain or soil is determined from the
concentration of pollutant in the animal tissue and the amount of
each contaminated animal tisstae that is consumed daily.
5.4.1. Calculating Daily Intake from Contaminated Plants. The
human daily intake of pollutants due to consumption of a specific
plant group is calculated by:
Ip, = I?,- • Fp,- • Cp,-
(Equation 5-22)
5-31
-------
where:
Fp.
CPi
= human daily intake of pollutant due to consumption
of the i plant group (/zg pollutant/kg BW/day)
= total concentration of pollutant in the ith plant
group (/zg pollutant/g plant tissue DW)
= fraction of i plant group assumed to originate
from contaminated soil (unitless)
= human daily dietary consumption of the ith plant
group for the appropriate age group (g plant tissue
DW/kg BW/day)
Total daily intake from consumption of all plant groups is
calculated by summing the contributions of each individual plant
group:
where:
IP
IP =
(Equation 5-23)
= total daily intake from consumption of plant
tissue (/zg pollutant/kg BW/day)
= daily intake from consumption of ith plant group
(/zg pollutant/kg BW/day)
5.4.1.1. PLANT GROUP CONSUMPTION RATES (Op,-) — An up-to-
date and detailed source of information regarding food consumption
habits of the U.S. population may be found in the U.S. EPA Office
of Pesticide Programs' "Tolerance Assessment System (TAS)". A
specific analysis conducted by Technical Assessment Systems, Inc.
(U.S. EPA, 1988d) for the U.S. EPA using the TAS is of particular
relevance to this methodology because the results are converted to
a dry rather than wet weight basis. Also, the data are presented
as frequency distributions of consumption instead of as population
averages. This analysis determined the distribution of mean,
individual daily consumption, in g DW food/kg BW, of seven plant
food groups for the total U.S. population and five subgroups: males
5-32
-------
> 13 years, females > 13 years, infants < 1 year, children 1-6
years, and children 7-12. The results are presented as a frequency
distribution in equal intervals based on population percentiles.
The seven food groups include grains and cereals; potatoes,
including sweet potatoes; leafy vegetables, excluding brassica;
brassica vegetables; legumes; fruiting vegetables (e.g., tomatoes,
cucumbers, squash, melons, etc.) and root vegetables. A complete
description of each food group is presented in Table 5-5. These
seven groups can be placed into the three broad plant categories
(leafy, exposed, protected) when calculating plant pollutant
concentration in the following manner. Foods in the leafy
vegetables and brassica vegetables food groups are considered to
be leafy produce. Foods in the grains, potatoes, legumes and root
vegetables food groups are considered to be protected produce. In
addition, melons, which are in the fruiting vegetables food group,
are also considered to be protected produce. The remaining foods
in the fruiting vegetables group as well as fruits are considered
to be exposed produce.
The consumption data used in the U.S. EPA (1988d) analysis
were collected for the U.S. Department of Agriculture Nationwide
Food Consumption Survey (NFCS) of 1977-78 (U.S. Department of
Agriculture, 1983). Three-day food consumption records were
collected during the period of April 1977 through March 1978.
About 15,000 households representing 30,770 individuals were
surveyed. The NFGS has the advantage of a large
5-33
-------
TABLE 5-5
Description of Seven Plant Food Groups
Food Group
Food Name
Food Group Food Name
Grains
Leafy
Vegetables
barley
corn -
bran
sugar
endosperm
popcorn
sweet corn
oil
oats
rice -
milled
rough
rye -
germ
flour
rough
sorghum
wheat -
germ
bran
flour
rough
millet
buckwheat
beet tops
celery
chicory
lettuce -
leafy
NS*
head
cress -
garden
upland
field
dandelion
parsley
rhubarb
spinach
swiss chard
turnip tops
Potatoes
Brassica
white -
NS*
peeled
dry
peel only
whole
sweet potatoes
broccoli
brussel sprouts
cabbage
cauliflower
collards
kale
kohlrabi
mustard greens
Root
Vegetables
carrots
taro root
garlic
artichokes
leeks
onions
radishes
rutabagas
salsify
shallots
turnips
parsnips
cassava
beets
5-34
-------
TABLE 5-5 (cont.)
Food Group
Food Name
Food Group Food Name
Leafy, cont.
Legumes
Fruiting
Vegetables
taro greens
endive
radish tops
rutabaga tops
dry beans -
great northern
kidney
lima
navy
pinto
broadbeans
pigeon
hyacinth
blackeye peas
garbanzo
succulent beans -
lima
green
yellow/wax
broadbeans
peas, dry and green
lentils, whole and split
sprouts, mung and soy bean
carob
soybeans -
oil
mature
flour
NS*
paprika
cantaloupe
casaba
crenshaw
honey dew
persian melon
watermelon
cucumber
pumpkin
summer squash
winter squash
eggplant
peppers -
sweet
chili
other
tomatoes -
whole
juice
puree
paste
catsup
* NS = not specified
Source: U.S. EPA (1988d)
5-35
-------
sample size, and it has been weighted to be statistically
representative of the U.S. population including race, age, sex,
regions and seasons (U.S. EPA, I988d).
The major limitation of the NFCS for exposure assessment is
that consumption is expressed in terms of foods "as eaten" (pizza,
apple pie, french toast) rather than in terms of the commodities
for which residue data are most often collected (wheat, beef,
apples) (U.S. EPA, 1988d). The U.S. EPA (1989b) has adapted the
NFCS so that it can be used to estimate the consumption of raw
agricultural commodities. A total of 3734 foods were examined and
the following information was entered into a master file:
components that made up the food; the percentage, by weight, of
each component; the reference source for the formula used to
estimate food composition in terms of components; and the form of
each component as it was eaten (raw, cooked, fried, etc.) . The TAS
analysis is based on the data entered into this master file, and
the mean daily consumption rates of the seven food groups predicted
by TAS are presented in Table 5-6.
The U.S. EPA (1988d) analysis included fruiting vegetables,
but not fruits per se (e.g., apples, strawberries, cherries, etc.).
The consumption rates for fruits were developed instead from
information presented in Appendix A of U.S. EPA (1989b) . The fruit
data were grouped, and the total average dry-weight consumption for
children (2 years, 13.5 kg BW) and adults (males and females >14,
70 kg BW) was calculated. Coefficients of variation for
consumption of fruits and vegetables (child=0.0955, adult=0.04)
were calculated from data in McKone and Ryan (1989). Consumption
5-36
-------
TABLE 5-6
Consumption Rate of Seven Plant Groups
(g DW/kg BW/day)
Plant
Group
Grains
Potatoes
Leafy
Brassica
Legumes
Fruiting
Root
Fruits8
Grains
Potatoes
Leafy
Brassica
Legumes
Fruiting
Root
Fruits0
Grains
Potatoes
Leafy
Brassica
Legumes
Fruiting
Root
Fruits8
U.S.
Population
4.4
0.72
0.06
0.07
1.4
0.30
0.11
2.4
0.34
0.02
0
0.67
0.13
0.04
1.6
0.19
0.008
0
0.44
0.07
0.02
Males
>13 yrs
90-95
3.0
0.58
0.05
0.06
1.0
0.24
0.08
0.
70-75
2.0
0.31
0.02
0
0.57
0.11
0.03
0.
50-55
1.5
0.19
0.008
0
0.40
0.06
0.02
0.
a Calculated from data presented
(1989),
Average
c Data for
see text.
of males and
2 -year-old
Females Children
>13 yrs 7
th Percentile
2.3
0.50
0.06
0.07
0.89
0.24
0.09
34b
th Percentile
1.5
0.26
0.02
0.001
0.49
0.10
0.03
33b
th Percentile
1.2
0.15
0.009
0
0.34
0.05
0.02
32b
in U.S. EPA
-12 yrs
4.9
0.91
0.06
0.08
1.7
0.38
0.13
2
3.6
0.49
0.02
0
0.98
0.19
0.05
2
2.9
0.30
0.009
0
0.69
0.11
0.03
1
(1989b)
Children
1-6 yrs
6.9
1.3
0.06
0.09
2.3
0.20
0.20
.19C
5.0
0.65
0.02
0
1.3
0.24
0.07
.08C
4.1
0.39
0.004
0
0.87
0.13
0.03
.95C
Infants
<1 yr
7.7
1.2
0.008
0
4.6
0.32
0.63
4.6
0.25
0
0
2.4
0.07
0.26
3.0
0.07
0
0
1.1
0.01
0.08
and McKone and Ryan
females >14 years
children only
Source (except fruits): U.S. EPA, 1988d
5-37
-------
was assumed to be normally distributed over the U.S. population,
and the above coefficients of variation were used with "Z-scores"
of 0.68 and 1.28 (corresponding to areas under the unit normal
distribution of 75% and 90% of the total area, respectively) to
estimate 70-75th and 90-95th percentiles of fruit consumption.
These values are presented in Table 5-6.
5.4.1.2. FRACTION ASSUMED TO ORIGINATE FROM CONTAMINATED SOIL
(Fp{) — The fraction of plant foods produced on contaminated soil
(at home) is primarily influenced by the number of households
gardening and the size of home garden. A total of 38% of U.S.
households raised a home garden in 1986, and the average garden was
325 square feet in 1986 (U.S. EPA, 1989a) . It will be assumed that
home gardeners may produce and consume leafy, legume and root
vegetables; potatoes; fruiting vegetables; fruits and some grains.
The most commonly raised garden vegetables are tomatoes, peppers,
onions, cucumbers, beans and lettuce (U.S. EPA, 1989a) . Fp,- was
calculated from the 1977-78 NFCS and values are presented in Table
5-7. For foods in the "General" category on Table 5-7, Fp,- was
calculated by dividing the reported weekly quantities of foods
produced at home by the quantities of foods from all sources. For
the specific food categories on Table 5-7, the quantities produced
at home were not reported. Therefore, the reported quantities of
bought foods were subtracted from the reported quantities of food
from all sources, and this difference was divided by the quantities
of food from all sources to determine the percentage "produced at
5-38
-------
TABLE 5-7
Percentage of Some Plant Foods Produced At Home (Fp,-)*
Food All
Groups Urbanizations
General
Flour, cereal
Sugar, sweets
Potatoes
Vegetables (fresh,
frozen, canned)
Fruit (fresh,
frozen, canned)
Juice
(vegetable, fruit)
Dried fruit,
vegetables
Potatoes
potatoes, white
sweet potatoes
Leafy Vegetables
celery
lettuce, unspec.
spinach
rhubarb
Brass ica Vegetables
broccoli
cabbage
cauliflower
collards
kale
mustard greens
Legumes
beans, lima
beans, succulent
peas, garden green
Fruiting Vegetables
cantaloupe
other melons
cucumbers
pumpkin , winter
squash
peppers
tomatoes
0.1
1.5
8.6
15.9
5.7
1.8
3.0
11.8
17.6
4.3
4.2
20.0
75.0
6.7
16.2
11.1
28.6
0.0
28.6
75.0
65.9
72.7
14.1
18.1
39.5
52.9
27.8
48.7
Central
City
0.1
0.5
1.5
5.8
2.0
0.3
1.3
3.01
9.52
4.3
2.59
0.0
50.0
5.88
6.49
0.0
15.4
0.0
15.4
50.0
34.6
50.0
5.08
7.89
21.2
38.5
15.0
31.7
Suburban
0.2
1.6
4.6
14.0
5.5
1.3
4.0
6.54
15.4
0.0
4.44
16.7
50.0
5.26
14.3
9.09
33.3
0.0
33.3
66.7
57.6
62.5
10.8
17.0
39.0
45.0
26.3
46.6
Non-
Metropolitan
0.2
2.6
17.2
27.0
9.8
3.8
7.7
22.0
33.3
5.3
5.83
33.3
83.3
20.0
27.1
16.7
66.7
50.0
60.0
85.7
85.7
80.0
25.0
30.5
59.0
70.6
42.9
62.3
5-39
-------
TABLE 5-7 (cont.)
Food
Groups
All
Urbanizations
Central
City
Suburban
Non-
Metropolitan
Root Vegetables
beets 85.7
carrots 15.9
onions, dry bulb 10.0
turnips 50.0
Grains
corn, sweet 44.8
50.0
7.32
1.75
0.0
19.5
83.3
14.3
8.00
33.3
34.4
90.0
26.8
19.5
60.0
66.7
Other Fruit
oranges
strawberries
apples
peaches
pears
grapes
plums
9.3
33.3
16.5
29.7
28.6
11.8
20.0
4.9
12.5
8.0
16.1
14.3
5.3
18.2
23.6
26.7
15.8
30.0
21.7
5.3
25.0
12.0
41.7
23.9
40.0
42.1
15.4
37.5
*Source: Calculated from U.S. Department of Agriculture, 1983
5-40
-------
home". However, Fp,- for the specific food categories may also
include foods received as gifts or eaten while the responder was
a guest. Thus, the Fp,- values in these categories may be higher
than those for foods in the "General" category. For the example
calculations, Fp{ for each food category was determined by
averaging the values of each individual food within the category.
On Table 5-7 the classes of urbanization are defined as
follows. Central city refers to those people who live within the
city limits of a Standard Metropolitan Statistical Area (SMSA).
Suburban refers to those people who live in a SMSA but do not live
within the city limits. Non-metropolitan refers to those people
who live outside a SMSA (U.S. Department of Agriculture, 1983).
5.4.2. Calculating Daily Intake from Animal Tissue. Human daily
intake due to ingestion of a single contaminated animal tissue is
calculated by:
IaJ
(Equation 5-24)
where:
IaJ
AJ
Faj
CaJ
total daily intake from ingestion of jth animal
tissue group (jig pollutant/kg BW/day)
th
= concentration of pollutant in the j n animal tissue
group (p.g pollutant/g animal tissue DW)
fraction of the jth animal tissue group assumed to
originate from contaminated soil (unitless)
= daily human consumption of the j animal tissue
group (g animal tissue DW/kg BW/day)
Total human daily intake from consumption of all animal groups
is calculated by summing the contributions of each specific animal
group:
5-41
-------
la = ? la-
'
(Equation 5-25)
where:
la
la
total daily intake from consumption of animal
tissue (/iig pollutant/kg BW/day)
daily intake from consumption of j animal group
pollutant/kg BW/day)
j
5.4.2.1. ANIMAL TISSUE CONSUMPTION RATES (Caj) — The
consumption rates for meat and dairy products were developed from
data presented in U.S. EPA (1989b) in the same manner as described
for fruits in Section 5.4.1.1. For each meat group and for dairy
products, the average dry-weight consumption of muscle tissue (or
milk solids) and fat were added. The data for children 2 years old
were divided by a body weight of 13.5 kg (Nelson et al., 1969).
The data for males and females >14 years old were averaged and
divided by a body weight of 70 kg.
A coefficient of variation for meat (child=0.5583,
adult=0.0326) and a weighted-average coefficient of variation for
dairy (child=0.0669, adult=0.0244) were calculated from data in
McKone and Ryan (1989). These values were used with z-scores of
0.68 and 1.28 to approximate the 70-75th and 90-95th percentile
meat and dairy consumption. These values are presented in Table
5*~8.
5.4.2.2. FRACTION OF ANIMAL TISSUE ORIGINATING FROM
CONTAMINATED SOIL (Faj) — The model presented in this methodology
allows assessors to consider both animals raised at home for home
consumption and animals raised commercially in the vicinity of a
5-42
-------
TABLE 5-8
Consumption Rates of Meat and Dairy Products (Ca.)
(g DW/kg BW/day)a
Food
Adultc
Child0
Beef
Beef liver
Lamb
Pork
Poultry
Dairy
Eggs
Beef
Beef liver
Lamb
Pork
Poultry
Dairy
Eggs
Beef
Beef liver
Lamb
Pork
Poultry
Dairy
Eggs
90-95th Percentile
0.5652
0.0166
0.00566
0.3439
0.1176
0.7432
0.1250
7 0-7 5th Percentile
0.5546
0.0163
0.00556
0.3375
0.1154
0.7327
0.1227
50th Percentile
0.5425
0.0160
0.00544
0.3301
0.1129
0.7207
0.1200
2.1290
0.0403
0.0206
1.646
0.6053
4.1270
0.9119
1.7130
0.0324
0.0166
1.3245
0.4870
3.9745
0.7337
1.2417
0.0235
0.0120
0.9600
0.3530
3.8015
0.5318
a Calculated from data presented in U.S. EPA (1989b) and
McKone and Ryan (1989), see text.
b Average of males and females >14 years
c Data for 2-year-old children only
5-43
-------
combustor that are distributed locally. Information on the
fraction of animal consumption that is home produced is presented
in the 1977-78 NFCS (U.S. Department of Agriculture, 1983). Values
of Faj for home produced animal foods are presented in Table 5-9
and are calculated as discussed in Section 5.4.1.2. The specific
values for beef, pork, and poultry were used to calculate human
daily intake of benzo(a)pyrene and cadmium. For the remaining
meats (beef liver and lamb), the "general meat" value was used
because specific meat values were not available. The "general"
values for dairy and eggs were also used in the example
calculations. However, information on the fraction of commercial
animal products that are distributed locally is not available
nationally. If this information is important for a site-specific
scenario, assessors should use local data.
5.5. EXAMPLE CALCULATIONS
This section illustrates a food-chain exposure to
benzo(a)pyrene and cadmium emitted from a combustor for Scenario
B. As discussed in Chapter 2, Scenario B is represented by a child
who grows up within 5 km of the combustor and remains in the area
for 30 years. A portion of all foods are assumed to be produced
at home; the rates for suburban households (Tables 5-7 and 5-9) are
used to estimate the amount of home-grown food. The amounts of
vegetables and meats not produced at home are assumed to be
imported into the area and do not contribute to exposure because
they are not contaminated. The amount of milk not produced at home
5-44
-------
TABLE 5-9
Percentage of Animal Foods Produced At Home (Fa-)*
Food All
Groups Urbanizations
General
Central
City
Suburban
Non-
Metropolitan
Dairy (milk, cream,
cheese , etc . )
meat
poultry, fish
eggs
Beef
Pork
fresh
cured
Poultry
chicken
turkey , other
1.2
4.0
6.8
3.3
7.0
4.8
4.7
5.1
4.3
2.6
14.0
0.0
0.2
1.9
0.3
1.4
1.4
0.6
1.9
1.6
0.7
8.3
0.3
2.0
5.9
' 2.2
3.9
3.4
2.8
4.2
3.6
1.4
13.0
3 „ 1
9. 7
10.8
7.9
15.3
9.0
10.7
6.9
8 O
\J • \J
6.0
20.0
*Source: Calculated from U.S. Department of Agriculture, 1983
5-45
-------
is assumed to be commercially produced from an area within 50 km
of the combustor and, therefore, does contribute to exposure.
The estimataed 70-75th percentile of plant consumption rates
(Table 5-6) is used to estimate exposure through plant foods. The
rates for children age 1-6 were used to estimate the individual's
exposure as a child, and the average rates of males and females >14
were used to estimate the individual's exposure as an adult. The
70-75th percentile of meat consumption rates is used to estimate
exposure through meats and animal products. These values were
calculated as described in Section 5.4.2.1.
5.5.1. Benzo(a)pyrene. This section presents examples of
calculating benzo(a)pyrene concentration in plants and animals and
calculating total daily intake of benzo(a)pyrene by humans. The
input variables used for calculating exposure to benzo(a)pyrene are
presented in Table 5-10.
5.5.1.1. CALCULATING BENZO(a)PYRENE CONCENTRATION IN
PLANTS — Exposure through the following eight plant groups is
illustrated: forage, grain, potatoes, leafy vegetables (includes
brassica), root vegetables, fruit, fruiting vegetables and legumes.
Pollutant concentration in these plants is the sum of concentration
due to root uptake, direct deposition of particles onto exposed
surfaces, and air-to-plant transfer of volatile compounds.
5-46
-------
TABLE 5-10
Input Variables For Scenario B Exposure Calculations: B(a)P
Input Variable
Sc (/ig/g)
5 km: 20 cm
1 cm
50 km: 20 cm
1 cm
Br,- [(jug/g plant DW)/(/xg/g
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
Dyd (g/m2/yr)
5 km
50 km
Dyw (g/m2/yr)
5 km
50 km
Fw (unitless)
Rp,- (unitless)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
kp (yr'1)
Tp,. (yr)
forage
other plants
Value Document Section
4.4.2.
3.81X10"6
7.66X10"5
2.70X10"7
5.43X10"6
soil)] 5. 2. 1.2. 2. a
0.003
0.51
0.51
0.51
0.067
0.067
0.38
0.38
3.6.
2.45X10'7
3.77X10"8
3.6.
6.74X10"7
2.74X10"8
0.02 5.2.2.2.b
5.2.2.3.°
0.0
0.008
0.0
0.0
0.05
0.05
0.16
0.47
126.5 5.2.2.4.
O • A • £t • O •
0.123
0.164
5-47
-------
TABLE 5-10 (cont.)
Input Variable
Value
Document Section
Ypj (kg DW/m2)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
BVj [(Mg/g plant DW)/(/itg/g air)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
cy (^g/m3)
5 km
50 km
cv (Mg/ro3)
5 km: 20 cm soil depth
1 cm soil depth
50 km: 20 cm soil depth
1 cm soil depth
5.2.2.6.
0.22
0.43
0.43
0.43
0.26
0.26
0.14
0.31
] 5.2.3.3.e
0.0
6.1X106
0.0
0.0
6.1X106
6.1X106
6.1X106
6.1X106
3.6.
3.32X10"6
6.10X10"7
5.2.3.4.
1.45X10"14
2.92X10"13
1.03X10"15
2.07X10"14
Fv (unitless)
pa (g/m3)
0.18
119 o
5.2.3.2.
QSj (kg DW/day)
beef
beef liver
dairy
pork
poultry
eggs
lamb
5.3.2.
0.3
0.3
0.4
0
0
0
0.4
5-48
-------
TABLE 5-10 (cont.)
Input Variable
Value
Document Section
QP
1J
(kg DW/day)
beef-grain
beef-forage
beef-silage
beef liver-grain
beef liver-forage
beef liver-silage
dairy-grain
dairy-forage
dairy-silage
pork-grain
pork-silage
poultry
eggs
lamb
(d/kg)
beef
beef liver
dairy
pork
poultry
eggs
lamb
0.47
8.8
2.5
0.47
8.8
2.5
2.6
11.0
3.3
3.0
1.3
0.08
0.08
1.4
0.027
0.036
0.004
0.019
0.031
0.049
0.020
5.3.1.
5.3.3,
Fpj (unitless)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
Fa- (unitless)
beef
beef liver
dairy
pork
poultry
eggs
lamb
5. 4.1. 2.
0.344
0.623
0.110
,347
,212
0.308
0.178
0,
0,
0.039
0.020
0.003
0.034
0.036
0.022
0.020
5.4.2.2.
5-49
-------
TABLE 5-10 (cont.)
Input Variable
Value
Document Section
(g DW/kg BW/day)
Children Adults1
grains 5.0 1.75
legumes 1.3 0.530
potatoes 0.65 0.285
root vegetables 0.07 0.03
fruits 2.08 0.33
fruiting vegetables 0.24 0.105
leafy vegetables 0.02 0.02
Ca, (g DW/kg BW/day)
Children Adults
beef 1.713 0.555
beef liver 0.032 0.016
dairy 3.975 0.733
pork 1.325 0.338
poultry 0.487 0.115
eggs 0.734 0.123
lamb 0.017 0.006
5.4.1.1,
5.4.2.1,
Source: Risk Assessment of the Dickerson Site (1988) as cited in Belcher
and Travis (1989)
Source: Risk Assessment of the Dickerson Site (1988) as cited in Belcher
(1989)
Sources: Baes et al. (1984), Hoffman and Baes (1979) and Shor et al.
(1982) as cited in Belcher and Travis (1989)
Source: For forage only, Hoffman and Baes (1979) as cited in Belcher and
Travis (1989). All other plant groups, Shor et al. (1982) data for
Hillsboro County, Florida. Adjusted to DW using water content data in
Baes et al. (1984)
Calculated using Equation 5-15; H = 0.157 Pa-m/mol, logKow = 6.06 (U.S.
EPA, 1986b)
Source: Belcher and Travis (1989)
Table 5-7, average of individual foods within each plant group for
suburban households
Table 5-9
-th
Table 5-6, average of 70-75 percentile of males and females >13 years old
5-50
-------
Benzo(a)pyrene concentration in plants due to root uptake (Pr)
is calculated according to Equation 5-2, assuming a soil depth of
20 cm.
For grains:
"6
Pr = 3.81X10" Mg/g * 0.003 (Mg/g plant DW)/(Mg/g soil)
= 1.14X10"8 Mg B(a)P/g plant tissue
For legumes, potatoes, and root vegetables:
Pr
= 3.81X10"
= 1.94xlO
"6
* 0.51
B(a)P/g plant tissue
plant DW)/(Mg/g soil)
For fruits and fruiting vegetables:
"6
Pr = 3.8ixio" Mg/g • o.oev (Mg/g plant ow)/(Mg/g soil)
= 2.55x10 Mg B(a)P/g plant tissue
For leafy vegetables and forage:
Pr
= 3.81xlO
= 1.45xlO
"6
"6
* 0.38 (Mg/g plant DW)/(Mg/g soil)
B(a)P/g plant tissue
Benzo(a)pyrene concentration in plants due to direct
deposition (Pd) is calculated according to Equation 5-4. Pd is not
calculated for grains, potatoes, and root vegetables because they
are "protected produce" and are not exposed to direct deposition.
Pd =
For legumes:
1000 • [2.45X10"7 + (0.02 • 6.74X10"7)]
0.008 • [l-exp(-126.5 • 0.164)]
= 3.80x10
-8
(0.43) (126.5)
B(a)P/g plant tissue
5-51
-------
For fruits and fruiting vegetables:
Pd m 1000 • [2.45X10"7 + (0.02 • 6.74X10"7)] • 0.05 • [1-exp (-126.5 • 0.164)]
3.93x10
(0.26) (126.5)
B(a)P/g plant tissue
Pd
For leafy vegetables:
1000 • [2.45X10"7 + (0.02 • 6.74xlO"7)] • 0.16
[l-exp(-126 . 5 • 0.164)]
(0.14) (126.5)
» 2.34x10"° /ig B(a)P/g plant tissue
For forage:
Pd = 1000 • [2.45X10"7 + (0.02 • 6.74X10"7)] • 0.47 • [1-exp (-126.5 • 0.123)]
(0.31) (126.5)
- 3.10x10'° ng B(a)P/g plant tissue
Benzo(a)pyrene concentration due to air-to-plant transfer (Pv)
is calculated according to Equation 5-13. Pv is not calculated for
grains, potatoes, and root vegetables because they are "protected
produce" and are not directly exposed to airborne volatile
compounds. Theoretically, one would use Cv calculated from a soil
concentration at 1 cm depth (2.92xlO~13) to calculate Pv for forage.
However, since Cv is so small compared to Cy, using a Cv for 1 cm
will not cause the Pv for forage to be any different than Pv for
other plant groups. Because the inputs for this pathway are
chemical specific and not species specific, Pv is the same for each
of the remaining plant groups.
5-52
-------
Pv = [(0.18 • 3.32X10
"6
+' 1.45X10
"14
• 6.1xl0
1190 g/nr
.-3
= 3.06x10 /zg B(a)P/g plant tissue
The total benzo(a)pyrene concentration in plants (P,-) is
calculated by adding the pollutant concentrations due to root
uptake, direct deposition, and air-to-plant transfer (Equation 5-
1).
For grains:
= 1.14X10
"8
B(a)P/g tissue
For legumes:
P, = 1.94X10"6 + 3.80X10"8 + 3.06X10"3
= 3.06xlO~3/ig B(a)P/g tissue
For potatoes:
P,- = 1.94X10"
B(a)P/g tissue
For root vegetables:
P, = 1.94x10"
B(a)P/g tissue
For fruits and fruiting vegetables:
Pf = 2.55X10"7 + 3.93X10"7 + 3 .06xlp"3
= 3.06xlO"3/zg B(a)P/g tissue
For leafy vegetables:
P,. = 1.45X10"6 + 2.34X10"6 + 3.06X10"3
= 3.06xlO"3/zg B(a)P/g tissue
For forage:
Pi = 1.45X10"6 + 3.10X10"6 + 3.06X10"3
= 3.06xlO"3/Ltg B(a)P/g tissue
5-53
-------
5.5.1.2. CALCULATING POLLUTANT CONCENTRATION IN ANIMAL
TISSUES — Exposure through the following seven animal food groups
is illustrated: beef, beef liver, dairy, pork, lamb, poultry, and
eggs. Animals ingest pollutants from contaminated plants and soil.
For cattle, separate ingestion rates for grain, forage and silage
are available; thus, intake from each of these groups is
calculated. Silage is assumed to be derived mainly from vegetative
plant parts from a crop such as corn or sorghum. Thus, the plant
tissue concentrations are assumed to be similar to those of forage.
Unlike forage, however, silage is not consumed by grazing and,
therefore, is not assumed to contribute to soil ingestion.
For the remaining animals, a single ingestion rate is
available. Sheep are assumed to consume only forage. Hogs are
assumed to consume grain and silage. Poultry is assumed to consume
only grain. Since hogs and poultry are not grazing animals and no
data on soil ingestion is available, hogs and poultry are assumed
not to consume soil. Benzo(a)pyrene concentration in animal
tissues is calculated according to Equation 5-19.
For beef:
[(0.47 kg grain • 1.14xlO"8 /xg/g) + (8.8 kg forage
3.06X10"3 jug/g) + (2.5 kg silage • 3.06xlO"3
(0.3 kg soil • 7.66X10"5
] • 0.027
9.34xlO"4 /ig benzo(a)pyrene/g animal tissue
5-54
-------
For beef liver:
A,
[(0.47 kg grain • 1.14x10"° Mg/g) + (8.8 kg forage •
3.06xlO"3 Mg/g) + (2.5 kg silage • 3.06x10°
"
For
(0.3 kg soil • 7.66X10"5 Mg/g) ] • 0.036
= 1.25xlO"3 Mg benzo(a)pyrene/g animal tissue
dairy:
= [(2.6 kg grain •> 1.14xlO"8 Mg/g) + (11.0 kg forage •
3.06xlO"y Mg/g) + (3.3 kg silage • 3.06x10° Mg/g) +
"5
(0.4 kg soil • 7.66X10" Mg/g)] * 0.004
1.75xlO
"4
benzo(a)pyrene/g animal tissue
For pork:
A-
For
AJ
For
AJ
[(3.0 kg grain - 1.14xlO
3 . 06xlO"y /ng/g) ] • 0.019
"8
+ (1-3 kg silage
"5
B(a)P/g animal tissue
1-8
7.56x10
poultry:
(0.08 kg grain • 1.14x10"° Mg/g) • 0.031
= 2.83xlO"11 Mg B(a)'P/g animal tissue
eggs:
(0.08 kg grain • 1.14x10
-8
0.049
4.47xlO
"11
B(a)P/g animal tissue
For lamb:
AJ -
[(1.4 kg forage
7.66x10
-3
3.06x10
-3
(0.4 kg soil
8.63x10
-5
. 0.020
B(a)P/g animal tissue
5.5.1.3. CALCULATING HUMAN DAILY INTAKE — Human daily intake to
plants and animals is based on the pollutant concentration in the
plant or animal tissue, the consumption rate of the tissue, and the
fraction of the food assumed to be produced at home. Because
Scenario B is represented by a child who grows up and remains in the
5-55
-------
area as an adult, this section presents daily intake of both a child
and an adult. The adjustments to exposure that are necessary for
calculating overall risk are presented in Chapter 15.
The daily intake of benzo(a)pyrene from plants is calculated
according Equations 5-22 and 5-23.
Child's intake from grains:
Ipj - 1.14X10"8 /ig B(a)P/g grain • 0.344 • 5.0 g DW/kg BW/day
= 1.96X10"8 jug B(a)P/kg BW/day
Adult's intake from grain:
Ip{ - 1.14x10'° jug B(a)P/g grain • 0.344 • 1.75 g DW/kg BW/day
» 6.86X10'9 /ig B(a)P/kg BW/day
Child's intake from legumes:
Ipf - 3.06X10"3 /ig B(a)P/g legume • 0.623 • 1.3 g DW/kg BW/day
- 2.48X10"3 /ig B(a)P/kg BW/day
Adult's intake from legumes:
Ip,- = 3.06X10'3 /tg B(a)P/g legume • 0.623 • 0.53 g DW/kg BW/day
= l.OlxlO"3 jug B(a)P/kg BW/day
Child's intake from potatoes:
Ipf = 1.94X10"6 /tg B(a)P/g potato • 0.11 • 0.65 g DW/kg BW/day
- 1.39xlO"7 jug B(a)P/kg BW/day
Adult's intake from potatoes:
Ip, = 1.94X10"6 /ig B(a)P/g potato • 0.11 • 0.285 g DW/kg BW/day
- 6.OSxlO"8 /ig B(a)P/kg BW/day
5-56
-------
Child's intake from root vegetables:
Ip,. = 1.94x10"° ng B(a)P/g root • 0.347 « 0.07 g DW/kg BW/day
= 4.71X10"8 fj.g B(a)P/kg BW/day
Adult's intake from root vegetables:
Ip1 = 1.94x10"° /^g B(a)P/g root • 0.347 • 0.03 g DW/kg BW/day
= 2.02xlO
"8
B(a)P/kg BW/day
Child's intake from fruits:
Ip,- = 3.06X10"3 jig B(a)P/g fruit • 0.212 • 2.08 g DW/kg BW/day
= 1.35X10"3 fig B(a)P/kg BW/day
Adult's intake from fruits:
Ipj = 3.06X10"3 jig B(a)P/g fruit • 0.212 • 0.33 g DW/kg BW/day
= 2.14xlO
"4
B(a)P/kg BW/day
Child's intake from fruiting vegetables:
Ip,. = 3.06X10"3 jug B(a)P/g fruit • 0.308 • 0.24 g DW/kg BW/day
= 2.26X10"4 ng B(a)P/kg BW/day
Adult's intake from fruiting vegetables:
Ip,. = 3.06X10"3 jug B(a)P/g fruit • 0.308 • 0.105 g DW/kg BW/day
= 9.90X10"5 ng B(a)P/kg BW/day
0.02 g DW/kg BW/day
Child's intake from leafy vegetables:
Ip,. = 3.06xlO"3 fj.g B(a)P/g leafy • 0.178
= 1.09X10"5 /Ltg B(a)P/kg BW/day
Adult's intake from leafy vegetables:
Ip, = 3.06X10"3 /ig B(a)P/g leafy • 0.178
0.02 g DW/kg BW/day
= 1.09x10"
B(a)P/kg BW/day
5-57
-------
The total daily intake from plants is the sum of the
benzo(a)pyrene concentration of each plant group:
Total Child's Intake from Plants:
Ip ^ (1.96X10'8 + 2.48X1Q'3 + 1.39X10'7 + 4.71X10*8 + 1.35xlO'3 +
2.26x10" + 1.09X10'5 ng B(a)P/kg BW/day) • 1 mg/103 /zg
« 4.07X10'6 Kg/kg BW/day
Total Adult's Intake from Plants:
Ip - (6.86X10"9 + l.OlxlO"3 + 6.08X10'8 + 2.02xlO'8 + 2.14xlO'4 +
° '
9.90x10° + 1.09X10
= 1.3 3x10 "6 mg/kg BW/day
B(a)P/kg BW/day) • l mg/10
3
The daily intake from animal foods is calculated according to
Equabions 5-23 and 5-24:
Child's intake from beef:
9.34X10'4 tig B(a)P/g beef • 0.039 • 1.713 g beef/kg
BW/day
6.24X10"5 /xg B(a)P/kg BW/day
Adult's intake from beef:
la.
9.34X10"4 /Ltg B(a)P/g beef • 0.039 • 0.555 g beef/kg
BW/day
2.02xlO"
B(a)P/kg BW/day
Child's intake from beef liver:
"3
la, = 1.25X10"3 fig B(a)P/g liver • 0.020 • 0.032 g liver/ kg
BW/day
S.OOxlO
"7
B(a)P/kg BW/day
5-58
-------
Adult's intake from beef liver:
la, = 1.25X-10"3 ftg B(a)P/g liver • 0.020 • 0.016 g liver/kg
BW/day
.-7
= 4.00x10' jug B(a)P/kg BW/day
Child's intake from pork:
la, = 7.56x10
B(a)P/g pork • 0.034 * 1.325 g pork/kg
BW/day
= 3.41x10"° Atg B(a)P/kg BW/day
Adult's intake from pork:
la, = 7.56xlO"5 jug B(a)P/g pork • 0.034
BW/day
= 8.69xlO"7 jug B(a)P/kg BW/day
Child's intake from poultry:
0.338 g pork/kg
2.83X10"11
kg BW/day
B(a)P/g poultry • 0.036 • 0.487 g poultry/
= 4.96x10"" Mg B(a)P/kg BW/day
Adult's intake from poultry:
la, = 2.83X10"11 jug B(a)P/g poultry • 0.036
kg BW/day
= 1.17X10"13 jug B(a)P/kg BW/day
0.115 g poultry/
0.734 g eggs/kg
Child's intake from eggs:
la- = 4.47xlO"11 jug B(a)P/g eggs • 0.022
BW/day
= 7.23X10"13 jug B(a)P/kg BW/day
Adult's intake from eggs:
la, = 4.47xlO"11 pg B(a)P/g eggs • 0.022 • 0.123 g eggs/kg
BW/day
= 1.21x10"
B(a)P/kg BW/day
5-59
-------
Child's intake from lamb:
la
,
"5
8.63X10" ng B(a)P/g lamb • 0.020 • 0.017 g lamb/kg
BW/day
= 2.93X10"8 ^g B(a)P/kg BW/day
Adult's intake from lamb:
la
,
"5
8.63X10" ng B(a)P/g lamb • 0.020 -0.006 g lamb/kg
BW/day
1.04X10
"8
B(a)P/kg BW/day
Dairy products from two sources are assumed to contribute to
exposure: a portion produced at home, within a 5 km area of the
combustor, and the remainder produced commercially within a 50 km
area of the combustor. Thus, determining exposure from dairy
products reguires two calculations. The pollutant concentration in
dairy products (Aj dairy) originating from 50 km was calculated as
shown on page 5-45 using the values of soil concentration and
deposition appropriate for the 50 km area.
Child's intake from dairy:
la, (5km) = 1.75xlO"4 /zg B(a)P/g dairy • 0.003 • 3.975 g dairy/
kg BW/day
= 2.09X10"6 p.g B( a) P/kg BW/day
la, (50km) = 8.05X10"3 jig B(a)P/g dairy • 0.997 • 3.975 g dairy/
kg BW/day
Total
3.19X10"
B (a) P/kg BW/day
2.09X10"6 + 3.19xlO"2 /zg B(a)P/kg BW/day
"2
3.19xlO"2 p,g B(a)P/kg BW/day
5-60
-------
Adult's intake from dairy:
»'4 ••- B(a)P/g dairy • 0.003 • 0.733 g dairy/
la, (5km) = 1.75x10
kg BW/day
= 3.85X10"7
la, (50km) = 8.05X10"3
kg BW/day
= 5.88X10"3
B(a)P/kg BW/day
B(a)P/g dairy • 0.997 • 0.733 g dairy/
B(a)P/kg BW/day
Total laj = 3.85X10"7 + 5.88X10"3 M9 B(a)P/kg BW/day
= 5.88X10
"3
B(a)P/kg BW/day
The total daily intake from animals is the sum of the
benzo(a)pyrene concentration of each meat or animal products group:
Total Child's Intake from Animal Foods:
la = (6.24X10"5 + 8.00X10"7 + 3.41xlO"6 + 4.96X10"13 + 7.23xlO~13 +
2.93xlO"8 + 3.19xlO"2 ng B(a)P/kg BW/day) • 1 mg/103 ng
= 3.20X10"5 mg B(a)P/Jcg BW/day
Total Adult's Intake:
la = (2.02X10"5 /ig + 4.00X10"7 + 8.69xlO"7 H- 1.17X10"13 + l.2lxlO"13 +
1.04X10"8 + 5.88xlO"3 fj.g B(a)P/kg BW/day) • 1 mg/103 jug
= 5.90X10"6 mg B(a)P/kg BW/day
5.5.2. Cadmium. This section presents examples of calculating cadmium
concentration in plants and animals and calculating total daily intake
of cadmium by humans. The input variables used for calculating exposure
to cadmium are presented in Table 5-11.
5-61
-------
TABLE 5-11
Input Variables For Scenario B Exposure Calculations: Cadmium
Input Variable
Sc (/zg/g)
5 km: 20 cm
l cm
50 km: 20 cm
1 cm
Br( [(/xg/g plant DW)/(/ig/g
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
Dyd (g/m2/yr)
5 km
50 km
Dyw (g/m2/yr)
5 km
50 km
Fw (unitless)
Rpj (unitless)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
kp (yr"1)
Tpf (yr)
forage
other plants
Value Document Section
4.4.1.
6.88X10"2
4.06X10"1
2.87X10"3
2.78X10"2
soil)] 5.2.1.2.1.a
0.05
0.24
0.09
1.98
1.16
1.16
1.18
0.39
3.6.
1.59X10"4
2.34X10"5
3.6.
2.17X10"4
6.60X10"6
0.02 5.2.2.2.b
5.2.2.3.c
0.0
0.008
0.0
0.0
0.05
0.05
0.16
0.47
31.88 5.2.2.4.d
O*^«^£*«3«
0.123
0.164
5-62
-------
TABLE 5-11 (cont.)
Input Variable
Value
Document Section
Bvf
Cy
Cv
(kg DW/nT)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
0.22
0.43
0.43
0.43
0.26
0.26
0.14
0.31
plant DW)/(/ug/g air)]
grans
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
forage
0.0
0 . 0
0 . 0
0.0
0 . 0
0 . 0
0.0
0 . 0
0.0
0.0
5.2.2.6.
5.2.3.3
3.6.
5.2.3.4
QPij (kg DW/day)
beef-grain
beef-forage
beef-silage
beef liver-grain
beef liver- forage
beef liver-silage
dairy-grain
dairy-forage
dairy-silage
pork-grain
pork-silage
poultry
eggs
lamb
5.3.1.
0.47
8.8
2.5
0.47
8.8
2.5
2.6
11.0
3.3
3.0
1.3
0.08
0.08
1.4
5-63
-------
TABLE 5-11 (cont.)
Input Variable
Value
Document Section
QSj (kg DW/day)
beef
beef liver
dairy
pork
poultry
eggs
lamb
B&J (d/kg)
beef
beef liver
dairy
pork
poultry
eggs
lamb
Fp{ (unitless)
grains
legumes
potatoes
root vegetables
fruits
fruiting vegetables
leafy vegetables
Faj (unitless)
beef
beef liver
dairy
pork
poultry
eggs
lamb
5.3.2.
0.3
0.3
0.4
0
0
0
0.4
5.3.3.f
0.003
0.135
0.0055
0.028
12.45
0.6496
0.027
5. 4.1. 2. a
0.344
0.623
0.110
0.347
0.212
0.308
0.178
5.4.2.2.h
0.039
0.020
0.003
0.034
0.036
0.022
0.020
5-64
-------
TABLE 5-11 (cont.)
Input Variable
Value
Document: Section
(g DW/kg BW/day)
Children
grains 5.0
legumes 1.3
potatoes 0.65
root vegetables 0.07
fruits 2.08
fruiting vegetables 0.24
leafy vegetables 0.02
(g DW/kg BW/day)
Adults1
1.75
0.530
0.285
0.03
0.33
0.105
0.02
5.4.1.1,
5.4.2.1,
J
beef
beef liver
dairy
pork
poultry
eggs
lamb
Children
1.713
0.032
3.975
1.325 ,
0.487
0.734
0.016
Adults
0.555
0.016
0.733
0.338
0.115
0.123
0.006
Sources: Risk Assessment of the Dickerson Site (1988) and U.S. EPA (1985b)
as cited in Belcher and Travis (1989)
Source: Risk Assessment of the Dickerson Site (1988) as cited in Belcher
(1989)
Sources: Baes et al. (1984), Hoffman and Baes (1979) and Shor et al.
(1982) as cited in Belcher and Travis (1989)
Sources: Baes et al. (1984) and U.S. EPA (1985b) as cited in Belcher and
Travis (1989)
Source: For forage only, Hoffman and Baes (1979) as cited in Belcher and
Travis (1989). All other plant groups, Shor et al. (1982) data for
Hillsboro County, Florida. Adjusted to DW using water content data in
Baes et al. (1984)
Source: Risk Assessment of the Dickerson Site (1988) as cited in Belcher
and Travis (1989)
Table 5-7, average of individual foods within each plant group for
suburban households
Table 5-9
Table 5-6, average of 70-75th percentile of males and females >13 years old
5-65
-------
5.5.2.1. CALCULATING CADMIUM CONCENTRATION IN PLANTS —
Exposure through the following eight plant groups is illustrated:
forage, grain, potatoes, leafy vegetables (includes brassica) , root
vegetables, fruit, fruiting vegetables and legumes. Cadmium
concentration in these plants is the sum of concentration due to
root uptake and direct deposition of particles onto exposed
surfaces. Because cadmium is not volatile, plant concentration due
to air-to-plant transfer is not considered. Cadmium concentration
in plants due to root uptake (Pr) is calculated according to
Equation 5-2.
For grains:
Pr = 6.88xio"2 Aig/g -o.os
= 3.44x10 p.g Cd/g plant tissue
For legumes:
Pr - 6.88xlO"2 p.g/g • 0.24
- 1.65X10"2 ng Cd/g plant tissue
For potatoes:
Pr « 6.88xlo"2 ng/g • 0.09
- 6.19X10"3 /itg Cd/g plant tissue
For root vegetables:
Pr = 6.88X10"2 M9/g ' 1.98
» 1.36X10"1 jitg Cd/g plant tissue
For fruits and fruiting vegetables:
Pr =» 6.88X10"2 /xg/g • 1.16.
= 7.98x10 /zg Cd/g plant tissue
For leafy vegetables:
Pr = 6.88X10"2 p.g/g • 1.18
= 8.12xlO"2 fj,g Cd/g plant tissue
5-66
-------
For forage:
-2
Pr = 6.88x10" p.g/g • 0.39
= 2.68x10
Cd/g plant tissue
Cadmium concentration in plants due to direct deposition (Pd)
is calculated according to Equation 5-4. Pd is not calculated for
grains, potatoes, and root vegetables because they are "protected
produce" and are not exposed to direct deposition.
For legumes:
Pd = 1000 • [1.59X10"4 + (0.02 • 2.17X10"4)] • 0.008 • [l-exp(-31.88 • 0.164)]
(0.43) (31.88)
= 9.51X10"5 jug Cd/g plant tissue
Pd
For fruits and fruiting vegetables:
1000 • [1.59X10"4 + (0.02 • 2.17X10"4)] • 0.05
[l-exp(-31.88 • 0.164)]
= 9.83xlO"
(0.26) (31.88)
Cd/g plant tissue
For leafy vegetables:
"4
"4
Pd = 1000 • [1.59x10"* + (0.02 • 2.17x10"*)] • 0.16 • [1-exp(-31.88 • 0.164)]
(0.14) (31.88)
.-3
= 5.84x10 p,g Cd/g plant tissue
Pd =
For forage:
1000 • [1.59X10"4 + (0.02 • 2.17X10"4)]
• 0.47 • [l-exp(-31.88 • 0.123)]
= 7.75x10"
(0.31) (31.88)
Cd/g plant tissue
5-67
-------
Cadmium concentration due to air-to-plant transfer (Pv) is not
calculated because cadmium is not present in the Vapor phase.
Therefore, the total cadmium concentration in plants (P^ is
calculated by adding the pollutant concentrations due to root
uptake and direct deposition (Equation 5-1).
For grains:
PJ = 3.44X10"3 /zg Cd/g tissue
For legumes:
-5
Pf = 1. 65x10"" + 9.51x10
= 1.66x10 /ig Cd/g tissue
For potatoes:
r
PI = 6.19X10"3 /ig Cd/g tissue
For root vegetables:
P{ = 1.36X10"1 fj.g Cd/g tissue
For fruits and fruiting vegetables:
Pt = 7.98X10"2 + 9.83X10"4
= 8.08xlO"2 ng Cd/g tissue
For leafy vegetables:
P,. - 8.12X10'2 + 5.84X10"3
= 8.70x10 p.g Cd/g tissue
For forage:
P,- = 2.68X10"2 + 7.75X10"3
= 3.46xlO"2 p,g Cd/g tissue
5-68
-------
5.5.2.2. CALCULATING CADMIUM CONCENTRATION IN ANIMAL
TISSUES -- ,Exposure through the following seven animal food groups
is illustrated: beef, beef liver, dairy, pork, lamb, poultry, and
eggs. Animals .ingest pollutants from contaminated plants and soil.
For cattle, separate ingestion rates of grain, forage and silage
are available; thus, intake from each of these groups is
calculated. Silage is assumed to be derived mainly from vegetative
plant parts from a crop such as corn or sorghum. Thus, the plant
tissue concentrations are assumed to be similar to those of forage.
Unlike forage, however, sileige is not consumed by grazing and,
therefore, is not assumed to contribute to soil ingestion.
For the remaining animals, a single ingestion rate is
available. Sheep are assumed to consume only forage. Hogs are
assumed to consume grain and silage. Poultry is assumed to consume
only grain. Since hogs and poultry are not grazing animals and no
data on possible soil ingestion is available, hogs and poultry are
assumed not to consume soil. Cadmium concentration in animal
tissues is calculated according to Equation 5-19.
For beef:
[(0.47 kg grain • 3.44x10 Mg/g) + (8.8 kg forage •
3.46X10"2 jug/g) + (2.5 kg silage • 3.46xlO"2 ' '
(0.3 kg soil • 4.06X10"1 /xg/g) ] • 0.003
v-3
= 1.54x10 jug Cd/g animal tissue
For beef liver:
Ai =
[(0.47 kg grain • 3.44x10 M9/9) + (8.8 kg forage •
3.46X10"2 /ig/g) + (2.5 kg silage • 3.46xlO"2 ••-'-* •
(0.3 kg soil • 4.06X10"1 /Ltg/g) ] ' 0.135
6.94x10
-2
Cd/g animal tissue
5-69
-------
For dairy:
[(2.6 kg grain • 3.44x10
3.46xlO"2 p.g/g) +
+ (11.0 kg forage
(0.4 kg soil • 4
3.66xlO"3 ng Cd/g animal tissue
(3.3 kg silage • 3.46X10"* jug/g) +
06xlO"1 Mg/g) ] ' 0.0055
For pork:
A
[(3.0 kg grain •
3.46X10""2 /ng/g) ]
3.44X10"3 ng/g) + (1.3 kg silage •
• 0.028
- 1.55x10 ° /
For poultry:
A, = (0.08 kg grain •
Cd/g animal tissue
3.43X10
"3
Cd/g
3.44xlO"3 /xg/g) • 12.45
animal tissue
For eggs:
(0.08 kg grain • 3.44xlO"3 p.g/g) • 0.6496
1.79x10"* p.g Cd/g animal tissue
For lamb:
[(1.4 kg forage
4.06X10"1
"2
• 3.46x10
• 0.027
5.69x10"° fig Cd/g animal tissue
+ (0.4 kg soil
5.5.2.3. CALCULATING HUMAN DAILY INTAKE — Human daily
intake to plants and animals is based on the pollutant
concentration in the plant or animal tissue, the consumption rate
of the tissue, and the fraction of the food assumed to be produced
at home. Because the B scenario is represented by a child who
growg up and remains in the area as an adult, this section presents
daily intake of both a child and an adult. The adjustments to
exposure that are necessary for calculating overall risk are
5-70
-------
presented in Chapter 15. The daily intake of cadmium from plants
is calculated according Equations 5-22 and 5-23.
Child's intake from grains:
Ip, = 3.44x10"
= 5.92X10"
Cd/g grain • 0.344 • 5.0 g DW/kg BW/day
Cd/kg BW/day
Adult's intake from grain:
"3
Ip,. = 3.44X10" /ig Cd/g grain • 0.344 • 1.75 g DW/kg BW/day
"
= 2.07X10"
Cd/kg BW/day
Child's intake from legumes:
Ip,. = 1.66x10"* Mg Cd/g legume • 0.623 • 1.3 g DW/kg BW/day
- - - x~2
= 1.34X10" /ug Cd/kg BW/day
Adult's intake from legumes:
= 1.66x10
= 5.48X10
-2
-3
Cd/g legume • 0.623
Cd/kg BW/day
0.530 g DW/kg BW/day
Child's intake from potatoes:
Ip,- = 6.19X10"3 /Lig Cd/g potato • 0.110 • 0.65 g DW/kg BW/day
= 4.43xlO"4 ng Cd/kg BW/day
Adult's intake from potatoes:
Ip,- = 6.19X10"3 /Ltg Cd/g potato • 0.110 • 0.285 g DW/kg BW/day
= 1.94X10"4 yug Cd/kg BW/day
Child's intake from root vegetables:
,-1
Ip,. = 1.36x10 ' p.g Cd/g root • 0.347 • 0.07 g DW/kg BW/day
0.03 g DW/kg BW/day
= 3.30xlOJ fig Cd/kg BW/day
Adult's intake from root vegetables:
Ip,- = 1.36X10"1 /ig Cd/g root • 0.347
= 1.42X10"3 jig Cd/kg BW/day
5-71
-------
Child's intake from fruits:
"2
IP,
8.08X10"
"2
3.56X10
Adult's intake from fruits:
Cd/g fruit • 0.212 • 2.08 g DW/kg BW/day
Cd/kg BW/day
"2
Ip{ = 8.08xlO"
= 5.65xlO~3 /zg Cd/kg BW/day
Cd/g fruit • 0.212 • 0.33 g DW/kg BW/day
Child's intake from fruiting vegetables:
IP, = 8.08xlO"2 jug Cd/g fruit • 0.308
= 5.97xlO"3 p.g Cd/kg BW/day
0.24 g DW/kg BW/day
Adult's intake from fruiting vegetables:
IP
, = 8.08X10"2 /zg Cd/g fruit • 0.308 • 0.105 g DW/kg BW/day
= 2.61xlO"3
Cd/kg BW/day
Child's intake from leafy vegetables:
Ip
{ = 8.70X10"2 p.g Cd/g leafy • 0.178 • 0.02 g DW/kg BW/day
= 3.10X10"4
Cd/kg BW/day
Adult's intake from leafy vegetables:
'2
Cd/kg BW/day
Ip, = 8.70X10'2 fig Cd/g leafy • 0.178 • 0.02 g DW/kg BW/day
- 3.10x10"*
The total daily intake from plants is the sum of the cadmium
concentration of each plant group:
"4
"3
Total Child's Intake from Plants:
Ip = (5.92X10"3 + 1.34X10"2 + 4.43x10"* + 3.30xlO"J + 3.56x10
+ 5.97X10"3 + 3.10X10"4 jug Cd/kg BW/day) • 1 mg/103 M9
= 6.49X10"5 mg/kg BW/day
"2
5-72
-------
Total Adult's Intake from Plants:
Ip = (2.07X10"3 + 5.48X10"3 + 1.94X10"4 + 1.42xlO~3 + 5.65X10"3
+ 2.61X10"3 + S.lOxlO"4 /ig Cd/kg BW/day) • 1 mg/103 fj.g
= 1.77X10"5 mg/kg BW/day
The daily intake from animal foods is calculated according to
Equations 5-23 and 5-24:
Child's intake from beef:
= 1.54xlO"3 jug Cd/g beef -0.039
= 1.03X10"4 jitg Cd/kg BW/day
1.713 g beef/kg BW/day
Adult's intake from beef:
"3
= 1.54X10" /xg Cd/g beef • 0.039 • 0.555 g beef/kg BW/day
"
= 3.33X10"5
Cd/kg BW/day
Child's intake from beef liver:
laj = 6. 94x10"^ fj,g Cd/g liver • 0.02 • 0.032 g liver/kg BW/day
= 4.44xlO"5 ng Cd/kg BW/day
Adult's intake from beef liver:
laj = 6.94x10"* jug Cd/g liver -0.02
= 2.22xlO"5 /Ltg Cd/kg BW/day
0.016 g liver/kg BW/day
Child's intake from pork:
laj = 1.55xlO"3 ng Cd/g pork • 0.034 • 1.325 g pork/kg BW/day
= 6.98xlO"5 /jg Cd/kg BW/day
Adult's intake from pork:
= 1.55X10"3 /xg Cd/g pork • 0.034 • 0.338 g pork/kg BW/day
= 1.78xlO
Cd/kg BW/day
5-73
-------
Child's intake from poultry:
la
j
3.43x10
BW/day
-5
Cd/g poultry • 0.036
6.01X10"3 ng Cd/kg BW/day
Adult's intake from poultry:
la, = 3.43X10"3 fj,g Cd/g poultry • 0.036
BW/day
0.487 g poultry/kg
0.115 g poultry/kg
= 1.42xlO
"5
Cd/kg BW/day
Child's intake from eggs:
la, ~ 1.79x10"* ng Cd/g eggs • 0.022 • 0.734 g eggs/kg BW/day
= 2.89xlO"6 Mg Cd/kg BW/day
Adult's intake from eggs:
laj » 1.79xlO"_* ng Cd/g eggs • 0.022 • 0.123 g eggs/kg BW/day
= 4.84X10
Cd/kg BW/day
Child's intake from lamb:
la, * 5.96x10"^ fj.g Cd/g lamb • 0.020 • 0.016 g lamb/kg BW/day
= 1.82X10"6 ng Cd/kg BW/day
Adult's intake from lamb:
la, = 5.96xlO"3 fj.g Cd/g lamb • 0.020 • 0.006 g lamb/kg BW/day
* 6.83xlO"7 Mg Cd/kg BW/day
Dairy products from two sources are assumed to contribute to
exposure: a portion produced at home, within a 5 km area of the
combustor, and the remainder produced commercially within a 50 km
area of the combustor. Thus, determining exposure from dairy
products requires two calculations. The pollutant concentration
in dairy products (A{ dairy) originating from 50 km was calculated
as shown on page 5-45 using the values of soil concentration and
deposition appropriate for the 50 km area.
5-74
-------
Child's intake from dairy:
la- (5km)
Total la.
= 3.66x10 p,g Cd/g dairy • 0.003 • 3.975 g dairy/kg
BW/day
= 4.36x10
-5
Cd/kg BW/day
,-4
(50km) = 2.39x10"* ng Cd/g dairy • 0.997 • 3.975 g dairy/kg
BW/day
-4
= 9.47xlO~* jitg Cd/kg BW/day
= 4.36X10"5 + 9.47X10"4 jitg Cd/kg BW/day
= 9.91X10"4 Aig Cd/kg BW/day
Adult's intake from dairy:
la, (5km) = 3.66x10"
BW/day
= 8.05x10"
la, (50km) = 2.39X10
BW/day
"4
= 1.75X10
"4
"6
Total la, = 8. 05xlO" + l.75xlO
= 1.83X10"4 ng Cd/kg BW/day
Cd/g dairy • 0.003 • 0.733 g dairy/kg
Cd/kg BW/day
Cd/g dairy • 0.997 • 0.733 g dairy/kg
Cd/kg BW/day
Cd/kg BW/day
"4
The total daily intake from animals is the sum of the cadmium
concentration of each meat and animal-product group:
Total Child's Intake from Animal Foods:
la = (1.03X10"4 + 4.44X10"5 + 6.98xlO"5 + 6.01X10"5 + 2.89xlO"6
+ 1.82xlO"6+ 9.91X10"4 /ng Cd/kg BW/day) • 1 mg/103 fj.g
= l.27xlO"6 mg Cd/kg BW/day
Total Adult's Intake from Animal Foods:
la = (3.33X10"5 + 2.22X10"5 + 1.78xlO"5 + 1.42xlO"5 + 4.84X10"7
+ 6.83xlO"7 + 1.83x10"* p.g Cd/kg BW/day) • 1 mg/103 ng
= 2.72X10"7 mg Cd/kg BW/day
5-75
-------
5.5.3. Summary. An individual's total food-chain exposure to
benzo(a)pyrene and cadmium for Scenario B has been calculated.
Exposure as a child and as an adult has been estimated. This
exposure is summarized in Table 5-12. The overall risk to the
individual will be discussed in Chapter 15.
5-76
-------
TABLE 5-12
Summary of Terrestrial Food Chain Exposure for Scenario B
Age/Source of Exposure
Daily Intake (rog/ka BW/day)
B(a)P Cadmium
Child
plants
animals
total
Adult
plants
animals
total
4.07x10
3.20x10
-6
-5
3.61x10
-5
1.33x10
5.90x10
-6
-6
7.23x10
-6
6.49X10
1.27X10
-5
-6
6.62x10
-5
1.77x10
2.72x10
-5
-7
1.80x10
-5
5-77
-------
-------
6. DETERMINING EXPOSURE FROM SOIL INGESTION
6.1. INTRODUCTION
Contaminants from combustor emissions deposited on soil can
be directly ingested by humans who eat soil intentionally or
incidentally by hand-to-mouth transfer. Some children with pica,
the intentional ingestion or mouthing of nonfood items, may consume
larger quantities of soil. Both adults and children may
incidentally ingest small amounts of soil? young children (i.e.,
preschool, 1-6 years of age) will incidentally consume more soil
than adults and older children.
The rate of soil ingestion (i.e., amount/day) and the soil
concentration of the contaminant affect exposure by this pathway.
Soil ingestion rates for children are based on studies that measure
quantities of nonabsorbable tracer minerals in the fecal material
of the children. For adults, the soil ingestion rate is determined
by making assumptions of the thickness of soil on skin, the number
of times hands are placed in the mouth and the exposed surface area
of the hands that contacts the mouth.
6.2. HUMAN DAILY INTAKE (DI)
Since this methodology assesses only the risk associated with
the increase in exposure due to combustor emissions (incremental
risk), the background concentration of contaminants in soil is not
included in the quantitation of daily intake.
6-1
-------
Human daily intake (DI, mg/kg/day) resulting from ingestion
of contaminated soil is a function of soil concentration and soil
ingestion rate:
where:
DI
Sc
Cs
BW
DI = Sc
CS
BW
.-1
(Equation 6-1)
= total human daily intake (mg/kg/day)
— soil concentration of pollutant after the total
time of deposition (mg pollutant/g soil)
= soil ingestion rate (g soil/day)
= body weight (kg)
The DI for this pathway is compared with the RfD0 (reference
dose for chronic oral exposure, mg/kg/day) for systemic toxicants
to determine if the contaminant adversely affects human health.
If the chemical is a carcinogen, the DI is used with the human
cancer potency to determine excess risk (ER). This is further
discussed in Chapter 15.
i
Assumptions and their ramifications of the soil ingestion
pathway can be found in Table 6-1.
6.2.1. Soil Concentration (So). Humans could possibly be exposed
to combustor-emitted contaminants soon after deposition on soil and
before contaminants are incorporated into soil layers. Therefore
for this pathway, 100% of the deposited contaminant is assumed to
be incorporated in the uppermost 1 cm of soil and ingested soil is
assumed to originate from the same 1 cm layer.
6-2
-------
TABLE 6-1
Assumptions for Soil Ingestion Exposure Pathway
Functional
Area
Assumptions
Ramifications/
Limitations
Exposure
Assessment
Deposited emissions
are not necessarily
soil incorporated and
may be concentrated
in the uppermost soil.
layer. Deposited
contaminant is
assumed to be distri-
buted within the upper-
most 1 cm of soil, and
ingested soil is eissumed
to originate from the
same 1 cm layer.
Overestimates exposure in
situations where soil
incorporation occurs to
any depth >1 cm. If
incorporation is to a
lesser depth, exposure is
underestimated. For
example, if exposure
is to fallout dust
directly, ingested soil
could approach 100%
deposited particulate.
Soil
Ingestion
Rate
Deposition-contaminated
soil may be ingested
intentionally by
children at the rates
observed in studies.
Outdoor soil:
Children = 0.1-5 g/day
Adults = 0.025-0.05 g/day
Overestimates exposure to
the extent the child
frequents some areas
where deposition has not
occurred.
May over- or under-
estimate exposure depending
on whether soil and/or
dust are considered.
6-3
-------
Another depth of soil incorporation may be chosen if it is more
appropriate for a site-specific assessment. The soil concentration
as determined in Chapter 4.0 should be used as input into Equation
6-1.
6.2.2. Soil Ingestion Rate (Cs) . Soil ingestion has been
recognized as an important source of exposure to several
pollutants. Soil ingestion rates vary depending upon the age of
the individual, amount of outdoor/indoor activity, frequency of
hand-to-mouth contact and seasonal climate. Few studies have been
conducted on soil ingestion rates in adults. LaGoy (1987)
estimated an ingestion rate of 0.025 g/day for an average adult
and 0.05 g/day if the adult exhibits frequent hand-to-mouth contact
(e.g., smokers). Children tend to have higher ingestion rates due
to more frequent hand-to-mouth contact; also, some children
intentionally eat soil. Soil ingestion rates for children range
from 0.1 to 5 g/day, with the high extreme of 5 g/day representing
children who intentionally eat soil (pica) (U.S. EPA, 1984a).
Studies aimed at more accurately determining the range of
ingestion rates (incidental not intentional) have yielded some
data, but are as yet inconclusive. Binder et al. (1985, 1986) used
tracers (aluminum, silicon and titanium) and determined an average
ingestion rate of soil for children to be 108 mg/day with a range
of 4-708 mg/day. Clausing et al. (1987) also used tracer elements
to determine an ingestion rate for children of 100 mg/day with a
range of 21-362 rag/day based only on soil that was taken from the
school playground. From these two studies, the average soil
6-4
-------
ingestion for children incidentally ingesting soil is estimated to
be 0.1-0.2 g/day with 0.2 as a reasonably protective value and 1
g/day as an upper-range (U.S. EPA, 1988a).
Seasonal changes also have a role in determining the amount
of ingested soil by affecting the amount of time an individual
spends outside. In a tropical or subtropical climate, people tend
to spend more time outdoors and the opportunity for contact with
the contaminated soil would be greater. In a temperate climate,
people spend less time outdoors and would most likely ingest
contaminated soils at a lower rate.
Hawley (1985) incorporated indoor dust as an additional
variable in determining total ingestion of contaminants in soil.
Values of 50 mg/day during the six warm months and 100 mg/day in
the remaining six months for young children and 0.56 mg/day for
adults were assumed for ingestion of dust. LaGoy (1987) suggests
that while children under 1 year usually do not come in direct
contact with soil, they can be exposed to contaminated house dust
with an average ingestion rate of 50 mg/day (250 mg/day maximum).
This methodology uses soil ingestion rates for outdoor
exposure in the example calculations. If indoor dust exposure is
important in a site-specific assessment, soil ingestion of both
outdoor soil and indoor dust could each be determined and added
together (Cs [outdoor] + Cs [indoor]).
6-5
-------
6.2.3. Body Weight (BW). For this pathway, a body weight of 70
kg is used for adults. A body weight of 17 kg is used for children
ages 1-7 years (Nelson et al., 1969) since soil ingestion rates
have been derived from data for children in this age range.
6.3. EXAMPLE CALCULATIONS
This section illustrates exposure to benzo(a)pyrene and
cadmium via soil ingestion represented by Scenario B. As described
in Chapter 2, Scenario B is represented by a child who grows up in
an area of moderate deposition and remains in the area for 30
years. Within this time the individual participates in gardening.
The values of the input variables chosen for Scenario B are
presented in Table 6-2, and are generally in the middle of the
range of values discussed earlier.
6.3.1. Benzo(a)Pyrene. The daily intake of benzo(a)pyrene due to
ingestion of soil for children can be calculated according to
Equation 6-1:
DI = 7.66x10"° mg/g • 1 g/day • (1/17 kg)
s 4.5lxlO"9 mg/kg/day
The daily intake of benzo(a)pyrene due to ingestion of soil for
adults can be calculated also using Equation 6-1:
DI - 7.66xlO"a mg/g • 0.04 g/day • (1/70 kg)
=* 4.38xlO"11 mg/kg/day
6-6
-------
TABLE 6-2
Input Variables for Soil Ingestion Pathway
Input Variable
Value
Children
Adults
Sc (mg/g)
B(a)P
Cadmium
Cs (g/day)
BW (kg)
7.66x10
4.06x10
1
17
-8
-4
(Z = 1 cm and Tc =60 yr)
7.66x10"°
4.06x10
0.04
70
-4
6-7
-------
These values are used in Chapter 15 to determine excess risk
for exposure to benzo(a)pyrene in combustor emissions.
6.3.2. Cadmium. The daily intake of cadmium due to the ingestion
of soil for children can be calculated using Equation 6-1:
DI - 4.06X10"4 mg/g • l g/day • (1/17 kg)
» 2.39X10"5 mg/kg/day
The daily intake of cadmium due to the ingestion of soil for adults
can be calculated using Equation 6-1:
DI = 4.06xlO"4 mg/g • 0.04 g/day • (1/70 kg)
= 2.32xlO"7 mg/kg/day
These values are compared with the RfD0 according to the
approach described in Chapter 15.
6-8
-------
7. DETERMINING EXPOSURE FROM
DERMAL ABSORPTION VIA SOIL
7.1. INTRODUCTION
The Dermal Exposure Model determines exposure from human skin
contact with contaminants in the soil. The issue of dermal
absorption of deposited contaminants is very complex. There is a
fundamental lack of data regcirding absorption of chemicals in soil
by human skin. Other factors important for estimating human
exposure to contaminants by the dermal route (i.e., contact time,
contact amount and surface area) also have many uncertainties.
The model described here is offered as a possible approach for
estimating human exposure and risk assbciated with dermal contact
of deposited emissions. However, in most if not all cases, the
available data will not provide a satisfactory basis for risk
calculations.
7.2. DAILY DERMAL INTAKE (DDI)
Since this methodology eissesses only the risk associated with
the increase in exposure due to combustor emissions (incremental
risk), the background concentration of contaminants in soil is not
included in estimation of dermal daily intake.
The daily dermal intake (DDI, mg/kg/day) represents the
increase above background in daily human dermal intake due to the
contaminant from combustor emissions. The DDI is determined as
follows:
7-1
-------
DDI - CT
where:
DDI
CT
SA
CA
AF
Sc
BW
10"3/24
SA • CA
AF
Sc
BW
"1
(10"3/24)
(Equation 7-1)
human daily dermal intake (mg/kg/day)
contact time (hrs/day)
exposed skin surface area (cm2)
contact amount (mg/cm2)
absorption fraction (%/day)
soil concentration (mg/g)
body weight (kg)
units conversion factor (10~6 g • 1 day • 103g)/
(g • 24 hrs • mg)
The assumptions and uncertainties associated with each of these
input parameters are discussed in the following sections and
summarized in Table 7-1.
To assess whether the contaminant poses a risk to human health
by this pathway, the DDI may be compared with a dermal RfD (if
available) or may be transformed to an equivalent oral DI and
compared with the oral RfD for chronic exposure (RfD0) for systemic
toxicants. If the chemical is a carcinogen, the DDI may be used
with the dermal human cancer potency (if available) or may be
transformed to an equivalent oral DI and used with the human oral
cancer potency to determine excess risk (ER). Assessment of risk
to humans is discussed in Chapter 15.
7.2.1. Contact Time (CT). Contact time is defined as the duration
(hrs/day) of daily contact with contaminants in soil. The duration
of daily contact is measured from the time of first contact to the
time skin is cleaned. Duration is affected by both age and time
spent outdoors vs. indoors in the vicinity of a combustor.
Age affects duration of contact because children generally
spend more of their time playing outside and retain soil on their
7-2
-------
TABLE 7-1
Assumptions and Uncertainties for the Dermal Exposure Model
Functional
Area
Assumptions/
Uncertainty
Ramifications/
Limitations
Exposure
Assessment
Dermal intake
Fraction of
contaminant
absorbed
Deposited emissions are
not necessarily soil
incorporated and may
concentrate in the upper-
most soil layer. Depos-
ited contaminant is
assumed to be distributed
within the uppermost 1 cm
of soil, and the soil
that contacts the skin
is assumed to originate
from the same 1 cm layer.
Linear with respect to
soil concentration
Chemical- and matrix-
specific
Data needed for dermal
absorption in humans
may not be available
Dermal absorption
varies with age
May be concentration
dependent
Overestimates exposure
in situations where
soil incorporation
occurs to any depth
>1 cm
May overestimate
exposure
Uncertainty for many
contaminants and
matrices
Uncertainty between
species extrapolation
for dermal exposure;
uncertainty in route-to-
route extrapolation of
absorption
Absorption data for a
contaminant at one age
may over- or under-
estimate absorption at
another age
Study data at one
concentration may
over- or underestimate
actual absorption at
another concentration
7-3
-------
TABLE 7-1 (cont.)
Functional
Area
Assumption/
Uncertainty
Ramifications/
Limitations
Contact time
Exposed skin
Contact Amount
Outdoor soil:
Children = 4-12 hrs/day
Adults •* 1-2 hrs/day
Total intake is propor-
tional to exposed
surface area:
Adults - 910-2940 cm2
Children = 300-1400 cm2
Outdoor soil:
Children = 0.5-1.5 mg/cm2
Adults = 0.5-3.5 mg/cm2
May over- or under-
estimate exposure
depending on whether soil
and/or dust are
considered
May over- or under-
estimate exposure
May over-or under
estimate exposure
depending on whether
soil and/or dust are
considered
7-4
-------
skin a for a longer time after coming indoors (U.S. EPA, 1988a).
This methodology assumes that children spend from 4-12 hours each
day in contact with soil, playing, depending on how much time they
spend in school. Adults would spend less time outdoors. This
methodology assumes adults spend 1-2 hours each day in contact with
soil, working in the yard and garden, after work and on weekends.
Time spent outdoors vs. indoors affects duration of contact
with the contaminant because the composition of outdoor soil and
indoor dust may be different. Because soil carried into the house
contributes to the household dust, about 80% of indoor dust has the
same contaminant concentration as outdoor soil. The remaining 20%
of indoor dust is particulate matter from cooking, smoking, wearing
fabrics, or using sprays and powders (Hawley, 1985).
Hawley (1985) estimated that children would be in contact with
outdoor soil for 4-12 hours/day and indoor dust/soil for 16-24
hours/day, depending on the age of the child and the time of year.
Adults were estimated to be in contact with outdoor soil for 8
hours/day and indoor dust/soil for 16 hours/day, depending on the
season.
The regional climate could increase or decrease the frequency
of daily contact with contaminants in soil. In a tropical
environment, individuals are likely to spend more time outdoors,
increasing the frequency of contact, than in a temperate or arctic
environment. Hawley (1985) estimated the contact frequency for
children was approximately 5 days/week for 6 months/year, or about
130 days/year, at a duration of 12 hours/day. Adults were
7-5
-------
estimated to have a contact frequency of approximately 2 days/week
for 5 months/year, or about 43 days/year, at a contact duration of
5 hours/day (U.S. EPA, 1988a).
This methodology uses the contact times reported for outdoor
soils in the example calculations. If indoor dust is important to
a site-specific assessment, contact time to outdoor,soil and indoor
dust could each be determined and added together. If regional
climate .is likely to affect the total contact time, a relevant
site-specific frequency of contact can be used in conjunction with
duration to determine total contact time (CT = hrs/day • days/yr).
7.2.2. Surface Area (SA). Total dermal intake of a contaminant
is approximately proportional to the exposed surface area.
Depending on the type of clothing worn, exposed surface area may
range from 910 - 2940 cm2 for adults, and 300 - 1400 cm2 for
children (U.S. EPA, 1984b). For example, the exposed surface area
of adults wearing short-sleeved, open-necked shirts, pants and
shoes, with no gloves or hat, is «2940 cm2, whereas that of
children wearing the same clothing is «980 cm2. U.S. EPA (1989a)
reports the total mean body surface area for adult males as 19,400
cm2. The mean surface area for adult males for the hands (1140 cm2)
and forearms (840 cm2) is 1980 cm2 or «11% of the total body surface
area (U.S. EPA, 1989a). The surface area of the forearms comprise
approximately one-half the total surface area of the arms. For
children aged 2-5 years the 50th percentile for the total body
surface area is 5790-7310 cm2 (males and females; midpoint «6550
7-6
-------
cm )(U.S. EPA, 1989a). The surface area of the hands and forearms
for children is, therefore, approximately 750 cm2 (»11% of 6550
cm2) .
Regional climate will affect the amount and/or type of
clothing worn. Obviously, people in cold climates may have less
exposed surface area than people in tropical climates. Risk
assessors should consider site-specific conditions when conducting
a risk assessment and should choose the high or low end of the
range accordingly.
7.2.3. Contact Amount (CA). Contact amount is defined as the
amount of soil accumulating on skin and is assumed to be in the
range of 0.5 to 1.5 ing/cm2 for both children and adults. This
range also represents an average for the entire exposed area (U.S.
EPA, 1984b; U.S. EPA, 1988a based on the reports of Snyder, 1975;
Lepow et al., 1975; and Roels et al., 1980). Soil on the skin
originates from both outdoor soil or indoor dust, although outdoor
soil contributes more significantly to overall dermal exposure
(Hawley, 1985). Based on the weight of soil on exposed hands,
Hawley (1985) assumed a contact amount of 3.5 mg/cm2 for adults
doing yard work and 0.51 mg/cm2 for children playing outdoors.
Hawley (1985) also estimated the average dust covering
indoors, 560 mg/m2, based on a range of 110-590 mg/m2 reported in
Solomon and Hartford (1976) and Schaefer et al. (1972). From this
estimate, Hawley (1985) calculated contact amount of indoor dust
to be 0.56 mg/cm2 for young children with hands, feet, and forearms
7-7
-------
exposed and older children with only hands exposed. Contact amount
of indoor dust for adults with hands and/or arms exposed was 0.56 -
1.8 mg/cm2.
This methodology will use the outdoor soil contact amount in
the example calculations. In a site-specific assessment, this
parameter could be modified to include indoor dust contact (CA=
CA[outdoor] + CA[indoor]).
7.2.4. Absorption Fraction (AF) . The absorption fraction is the
fraction of the applied dose of the compound absorbed by human skin
in one day (i.e., 6%/24 hours) . If no specific data are available,
a conservative assumption may be made. The assumption of complete
absorption (i.e., AF=1) would be the most conservative approach,
but may be unrealistic. For example, Kimbrough et al. (1984)
suggested human dermal absorption of TCDD from soil was 1%. This
value is most likely an upper estimate of what is likely to be
absorbed by humans (Shu et al., 1988).
Dermal absorption of a contaminant depends on the
physicochemical properties of the contaminant and the matrix (U.S.
EPA, 1982c; Hawley, 1985) as well as the characteristics of the
skin. Properties of the contaminant such as lipid solubility, pH
and molecular size will affect dermal absorption. The fraction of
the contaminant absorbed may vary with concentration (Feldman and
Maibach, 1974). Shu et al. (1988), however, concluded that, in the
rat, dermal bioavailability of TCDD in soil was not altered by
7-8
-------
concentration of the chemical. This methodology conservatively
assumes that absorption of the contaminant increases continuously
with the contaminant concentration in soil.
Absorption of a contaminant contained in a matrix or vehicle
is affected by the physicochemical properties of the matrix, but
there is uncertainty as to how these soil or particle types
influence absorption of deposited contaminants. Poiger and
Schlatter (1980) demonstrated that a soil matrix reduced the
absorption of TCDD when TCDD was applied to the skin of rats in a
soil-water paste.
Absorption of contaminants may be altered by skin damage
(disease, laceration or abrasion), age, anatomical location (U.S.
EPA, 1982b; Maibach et al., 1971), temperature and humidity;
however, the specific way in which these factors affect the
absorption is unclear due to a scarcity of information.
Estimating the absorbed fraction of a contaminant is very
complex and depends on many factors as discussed above. A paucity
of data regarding the dermal absorption of chemicals in humans,
particularly from particulate or soil matrices, will make
estimating the fraction of contaminant absorbed from soil difficult
for this methodology. In many cases, DDI will not be able to be
calculated due to lack of data.
Values of absorption fractions may not be available for use
in exposure assessment. One approach might be to extrapolate
absorption fractions from other types of studies. Although studies
of absorption by the dermal route of exposure may not be available
in the literature, studies of absorption by other routes are often
7-9
-------
available. However, route-to-route extrapolation poses
uncertainties, and guidelines for this type of extrapolation have
not been clearly outlined. In addition, extrapolation of dermal
absorption data among animal species introduces uncertainty to the
assessment.
7.2.5. Soil Concentration (So). Humans could possibly be exposed
to contaminants soon after deposition on soil and before
contaminants are incorporated into soil layers. Individuals would
then be exposed to the greatest concentrations of the contaminants.
Therefore for this pathway, 100% of the deposited contaminant is
assumed to be incorporated in the uppermost 1 cm of soil and soil
available for contact is assumed to originate from the same 1 cm
layer. Another depth of soil incorporation may be chosen if it is
more appropriate for a site-specific assessment. The soil
concentration as determined in Chapter 4 should be used as input
into Equation 7-1.
7.2.6. Body Weight (BW). For this pathway, a body weight of 70
kg is used for adults. A body weight of 16.7 kg is used for
children ages 2-5 years (Nelson et al., 1969) since surface area
measures are approximated for children in this age range.
7.3. EXAMPLE CALCULATIONS
This section illustrates a dermal absorption to benzo(a)pyrene
and cadmium represented by Scenario B. As presented in Chapter 2,
Scenario B is represented by a child who grows up in an area where
7-10
-------
pollutants have been emitted for the 60 years of combustor
operation. The child remains in the area for 30 years. The
individual gardens and is exposed by dermal contact to pollutants
in soil. The values of the input variables chosen for Scenario B
are generally in the middle of the range of values discussed
earlier and are presented in Table 7-2. Surface area for exposed
hands and forearms was used.
7.3.1. Benzo(a)Pyrene. Exposure to benzo(a)pyrene from dermal
absorption of soil for children can be calculated according to
Equation 7-1:
DDT = 8 hr/day • 750 cm2 • 1.0 mg/cm2 • AF • 7.66xlO~8
mg/g • (1/16.7 kg) • (l day • 10"3g)/(24 hr • mg)
= Cannot be quantitated because input data for AF are not
available
Exposure to benzo(a)pyrene from dermal absorption for adults can
be calculated using Equation 4-1:
DDI = 1.5 hr/day • 1980 cm2 • 1.0 mg/cm2 • AF • 7.66xlO"8
rag/g • (1/70 kg) • (1 day • 10"3g)/(24 hr • mg)
= Cannot be quantitated because input data for AF are not
available -
If data were available, the DDI could be used in Chapter 15
to determine excess risk for exposure to benzo(a)pyrene in
combustor emissions.
7-11
-------
TABLE 7-2
Input Variables for Dermal Exposure Model
Input Variable
CT
SA
CA
AF
SC
BW
(hrs/day)
(cm2)
(mg/cm2)
(%/day)
(mg/g) (Z =
B(a)P
Cadmium
(kg)
Values
Children
8
750
1.0
NA
1 cm and Tc = 60 yr)
7.66X10"8
4.06X10"4
16.7
Adults
1.5
1980
1.0
NA
7.66X10"8
4.06X10"4
70
NA: Data not available
7-12
-------
7.3.2. Cadmium. Exposure to cadmium from dermal absorption for
children can be calculated using Equation 4-1:
DDI = 8 hr/day • 750 cm2 * 1.0 mg/cm2 • AF • 4.06xlO~4
mg/g • (1/16.7 kg) • (1 day • 10"3g)/(24 hr • mg)
= Cannot be quantitated because input data for AF are not
available
Exposure to cadmium from dermal absorption for adults can be
calculated using Equation 4-1:
DDI = 1.5 hr/day • 1980 cm2 • 1.0 mg/cm2 • AF • 4.06xlO"4
mg/g • (1/70) • (1 day • 10"3g)/(24 hr • mg)
= Cannot be quantitated because input data for AF are not
available
If data were available, the DI could be used for comparison
to the RfD according to the approach described in Chapter 15.
7-13
-------
-------
8. DUST RESUSPENSION
8.1. INTRODUCTION
Pollutants in the soil can be resuspended as dust by wind
erosion. A particular fraction of these particles, that is those
< 10 /im in diameter, may be inhaled by persons (U.S. EPA, 1988b) .
Methodologies have been developed to assess human exposure to
pollutants resuspended by wind from surface contamination sites.
This section describes the process of soil erosion and the factors
that influence dust resusperision, and provides two examples to
illustrate its use. The calculations were performed using the
methodology described in Rapid Assessment of Exposure to
Particulate Emissions from Surface Contamination Sites (U.S. EPA,
1985a) to estimate short-term (24-hour) and long-term (annual
average) exposure. For both examples, the annual emission rates
for dust resuspension are less than the stack emission rate.
8.2. SOIL EROSION PROCESS
The process of soil erosion is a function of continual soil
motion: surface creep, saltation and suspension. Surface creep
describes the rolling or sliding movement of particles 500-1000 ^m
in diameter (Sehmel, 1980). These surface creep particles are too
heavy to be lifted up by the wind, but roll and exchange momentum
during collision with smaller particles, which undergo saltation.
Saltation is leaping and bouncing particles lifted by the wind that
are too heavy to remain suspended. Saltation particles range from
100 to 500 /zm in diameter (Sehmel, 1980) . Suspension of particles
8-1
-------
occurs when saltation particles collide with smaller particles,
that is particles <100 /wn. If saltation was prevented, the
suspension particles would not be readily eroded (Sehmel, 1980).
8.3. FACTORS AFFECTING WIND EROSION
Soil erosion is affected by soil characteristics and wind.
The factors affecting dust resuspension include: moisture content,
presence of nonerodible surfaces (i.e., rocks and vegetation),
surface crust stability, wind velocity, particle size, and soil
roughness height.
Moisture hinders the resuspension of dust by forming a
cohesive matrix among the dust or soil particles. When the soil
is saturated, erosion will not occur (Sehmel, 1980). As the
surface layer moisture is removed by evaporation, resuspension
occurs. Evaporation is influenced by wind speed, temperature and
soil characteristics.
The presence of nonerodible surfaces reduces the amount of
dust resuspension by decreasing the soil surface area exposed to
the wind. Nonerodible surfaces include clumps of grass, and stones
>1 cm in diameter. Soil surfaces containing nonerodible surfaces
are considered to be "limited reservoirs" because of finite
availability and threshold friction velocities >75 cm/s (U.S. EPA,
1985a). Areas that are highly erodible have threshold friction
velocities <75 cm/s and are considered to have unlimited erosion
potential (U.S. EPA, 1985a). For example, an area covered with a
8-2
-------
continuous layer of grass has minimal soil erosion by wind as
compared with an agricultural field consisting of erodible
particles.
The surface crust, which is composed of small particles,
hinders wind erosion and evaporation. Thus, crusted surfaces have
greater threshold friction velocities. Soils without crusts such
as sandy or disturbed soils are more erodible (U.S. EPA, 1985a).
Particle size influences the wind speed required to initiate
dust resuspension by wind erosion. Particles ranging in diameter
from 100 to 150 /Ltm have the lowest threshold wind speeds. As the
particle size decreases below 100 /zm the threshold speed increases.
For particles >150 jum, the threshold speed also increases. If
particle sizes are mixed, the threshold wind speed is less than the
speed required to erode the largest particle (Sehmel, 1980).
Soil roughness height is related to height, width and spacing
of clods and furrows. This parameter is required to convert the
threshold friction velocity to a wind speed comparable with the
wind speeds at 7 m above the surface (U.S. EPA, 1985a) . Cowerd and
Guenther (1976) developed the roughness height scale for various
conditions of ground cover, such as natural snow and plowed field.
8.4. EXAMPLE CALCULATIONS
Some methodologies have been developed by the U.S. EPA to
determine the total wind erosion soil loss resulting from surface
creep, saltation, and suspension (Skidmore and Woodruff, 1968;
Sehmel, 1980; and Smith et al., 1982) while one other methodology
8-3
-------
determines the soil erosion loss for only the portion that is both
suspendible and respirable (U.S. EPA, 1985a). To assess human
exposure to pollutants resuspended by wind erosion (U.S. EPA,
1985a) , this latter methodology has been applied. In applying this
method to deposited incinerator emissions, dust resuspension yields
an emission rate that can be compared with the incinerator stack
emission rate. Both of these emission rates are input for air
dispersion modeling, which predicts ambient air concentrations to
which persons are exposed. Below are two examples that illustrate
the emission rate of the resuspended dust is less than or equal to
the stack emission rate. The equations used to calculate both the
annual average emission rate and the worst-case 24-hr emission rate
were taken from Rapid Assessment of Exposure to Particulate
Emissions from Surface Contamination Site (U.S. EPA, 1985a).
Both examples compare the cadmium emission rate from the stack
of an incinerator with the cadmium emission rate from dust
resuspended by wind erosion. The stack emission rate of cadmium
was based on an incinerator in western Florida and was previously
estimated to be 1.9xl04 Atg/s (PEI Associates, Inc. and H.E. Cramer-
Company, Inc., 1986). This section presents a soil concentration
modeled for the Florida incinerator. It also presents an estimate
of average annual emission rate and worst-case 24-hr emission rate
for two hypothetical sources: a 10,000 m2 field in western Florida
and a 400,000 m2 field in North Platte, Nebraska.
8-4
-------
8.4.1. Estimation of Soil Concentration. The soil
concentration of cadmium for each example was estimated by using
the deposition rate modeled for the Florida incinerator and
assuming the soil type is silt/loam. For Florida, the soil
concentration of cadmium is 4.06xlO~4 mg/g as calculated for
Scenario B in Section 4.1.1. In order to calculate the soil
concentration for Nebraska, the soil loss constant of cadmium for
silt/loam soils is needed. The soil loss constant for losses due
to leaching is estimated for each example by using geographic-
specific parameters in Equation 4-3.
For North Platte, Nebraska, the soil loss constant due to
leaching is calculated by:
ksl =
P + I - Ev
e
Z « [1.0 + (BD •
where:
P
I
Ev
e
z
BD
= 50 cm/yr (Baes et al., 1984)
= 40 cm/yr (Baes et al., 1984)
=62.5 cm/yr (Baes et al., 1984)
= 0.22 mL/cm
= 1 cm
=1.5 g/cm3 (NRG, 1984)
= 148 mL/g
ksl =•
50 cm/yr + 40 cm/yr - 62.5 cm/yr
0.22 mL/cnr* • 1.0 cm • [1.0 + (1.5 g/cm3 • 148 mL/g/0.22 mg/cm3) ]
= 0.18 yr"1
8-5
-------
The soil loss constant due to leaching (ksl) is assumed to equal
the soil loss constant for all processes (ks) since cadmium does
not undergo degradation (abiotic and biotic) or volatilization.
The cadmium soil concentration is estimated by using Equation 4-
1 and by assuming soil depth is 1 cm, total time period for
emissions is 60 years, and soil loss constant due to leaching is
equal to soil loss constant for all processes. For Nebraska,
cadmium soil concentration is calculated by:
Sc =
(Dyd + Dyw) • [1.0-exp(-ks • Tc)] • 0.1
BD
ks
where:
Dyd
Dyw
ks
Tc
Z
BD
= 1.59xlO~4 g/m2/yr
= 2.17xlO~4 g/m2/yr
=0.18 yr
= 60 yr
= 1 cm
=1.5 g/cm3
(1.59X10"4 + 2.17X10"4) • [1.0-exp(-0.18 • 60)] • 0.1
Sc =
1 • 1.5 • 0.18
= 1.39X10"4 mg Cd/g soil
The soil concentrations of cadmium due to the deposition of
incinerator emissions are 4.06xlO"4 mg/g for Florida and 1.39xlO"4
mg/g for Nebraska.
8-6
-------
8.4.2. Annual Average Emission Rate. The annual emission rate
of contaminant as respirable particles (R10) is the product of the
area of the field, the pollutant concentration in the soil and the
emission factor of respirable particulate matter, those < 10
diameter (PM10) (U.S. EPA, 1985a) .
where:
R
A"
a
E
RIO =
a
E
10
(Equation 8-1)
10 = annual emission rate of the contaminant (mg/s)
= area of the source (m )
= pollutant concentration in the PM10 fraction of soil (mg
pollutant/mg soil)
10 = annual average PM10 emission factor (g/m2/hr)
For both examples, the contaminated site is assumed to be
unvegetated with no nonerodible elements and no evidence of
crusting (U.S. EPA, 1985a). Since both examples are hypothetical,
their wind erosion threshold friction velocities are assumed to be
50 cm/s (U.S. EPA, 1985a), and their roughness heights are assumed
to be 2.0 cm (for grasslands) (U.S. EPA, 1985a). Threshold
friction velocity is usually determined by hand sieving during a
site investigation; field procedures are outlined in U.S. EPA
(1985a). For the first example, the hypothetical source is a field
in Tampa, Florida with an area, A, of 10,000 m2. For the second
example, a field in North Platte, Nebraska, the area, A, is assumed
to be 400,000 m2. (Both examples are modifications of an example
Rapid Assessment of Exposure to Particulate Emissions from
8-7
-------
Surface Contamination Sites (U.S. EPA, 1985a) with various
parameter values changed to represent the conditions of fields
contaminated by combustor emissions.)
Calculation of E10 and R10 for the two examples are described
below. Choosing the appropriate calculation for E10 depends on
whether the field in question is a "limited" or "unlimited"
reservoir. Fields with wind erosion threshold velocity of <75 cm/s
are considered "unlimited reservoirs". For both examples the
threshold erosion friction velocity is <75 cm/s, so both fields are
categorized as "unlimited reservoirs". The annual average emission
rate of contaminated surface with unlimited erosion potential (E10)
is calculated by:
E10 = 0.036
(1-V) • ([u]/ut)3 • F(x) (Equation 8-2)
where:
E
v10
[u]
ut
X
P(x)
emission factor of respirable (PM10) particles (g/m /hr)
fraction of contaminated surface vegetation cover
mean annual wind speed (m/s)
threshold wind speed (m/s)
0.886 ut/[u]
function of x (derived from Figure 8-1)
For the first example, the field is assumed to be bare, so V
is assumed to be 0. The mean annual wind speed for Tampa is 3.9
m/s (U.S. EPA, 1985a) and is used to calculate threshold wind
speed. Threshold wind speed is calculated by:
8-8
-------
1.5
as
a-
o
c
£ 0.5
—
Pt
^\
— ^
*•«»
•^
s
>
\
F(x) Tends to 1.91
as x tends to zero.
x = 0.886 ut/[u]
\
\
1
\
S
V
\
\
>
V
\
V
~\
See Appendix B for
larger values of x.
\
\
•
U-
\
•
k
I*
I
0.5
1.5
Figure 8-1,
Source: U.S. EPA, 1985a
Graph of Function F(x) Needed to
Estimate Unlimited Erosion
8-9
-------
ut =
(Equation 8-3)
where:
= threshold wind speed (m/s)
R^ = ratio of wind speed at 7m to friction velocity as a
function of roughness height
Fv » wind erosion threshold friction velocity (m/s)
For the Florida example, the wind erosion friction velocity
is assumed to be 0.50 m/s and the roughness height is assumed to
be 2.0 cm for grassland. Rw for a roughness height of 2.0, derived
from Figure 8-2, is 15.0.
ut =
where:
RH =15.0
Fv =0.50 m/s
ut= 15.0 • 0.50 m/s
=7.5 m/s
Therefore, x can be calculated by the following equation.
x = 0.886 • (ut/[u]) (Equation 8-4)
where:
ut =7.5 m/s
[U] =3.9 m/s (U.S. EPA, 1985a)
x = 0.886 • (7.5 m/s/3.9 m/s)
= 1.7
Using Figure 8-1, F(x) = 0.62.
E10 for the field in western Florida is calculated by using
equation 8-2.
E10 = 0.036 (l-V) ([u]/ut)3 F(x)
8-10
-------
X4ioo|a/\ UOI40IJ.J 04 uj/ 40 paad$ pui^ jo 014031
Figure 8-2. Ratio of wind speed at 7m to friction velocity
as a function of roughness height
Source: U.S. EPA, 1985a
8-11
-------
where:
V
[u]
ut
x
F(X)
= 0
3.9 m/s (U.S. EPA, 1985a)
7.5 m/s
1.7
0.62
E10 = 0.036
(1-0) • (3.9 m/s/7.5 m/s)
(0.62)
= 0.0031 g/hr/HT
For the second field example, the field is also assumed to be
bare, so V is 0. The mean annual wind speed [u] of Nebraska is 4.6
m/s (U.S. EPA, 1985a). The wind erosion friction velocity is
assumed to be 50 cm/s and the roughness height is assumed to be
2.0 cm. Rw for a roughness height of 2.0 is 15 as derived from
Figure 8-2. Thus, the threshold wind speed for the field in
Nebraska is 7.5 m/s.
ut = RH • Fv
where:
P^ = 15
Fv = 50 cm/s
ut= 15 • 50 cm/s
=7.5 m/s
Therefore x is calculated to be 1.4.
x = 0.886 • (ut/[u])
where:
t =7.5 m/s
=4.6 m/s
X = 0.886 • (7.5 m/s/4.6 m/s)
= 1.4
8-12
-------
Thus, F(x) is 1.0 as estimated from the graph in Figure 8-1. E10
for the field in Nebraska is calculated by:
E10 = 0.036 • (1 - V) • ([u]/ut)'
F(x)
where:
V = 0
[u] =4.6 m/s
ut =7.5 m/s
x = 1.4
F(x) = 1.0
E10 = 0.036 • (1-0) • (4.6 m/s/7.5 m/s)3 • (1.0)
= 0.0083 g/hr/m2
Since the area of the field, the pollutant concentration in
the soil and the PM10 emission factor are known for each example,
the annual emission rates can be calculated using equation 8-1.
For the field in western Florida, the emission rate is
calculated by:
where:
A
a
E
R10 = A
=•10
10
10,000 XT
4.06xlO"A mg Cd/g soil
0.0031 g/hr/m2
R10 = 10,000 m2 • 4.06X10"4 mgr/g • 0.0031 g/hr/m2 • 1 hr/3600 s
= 3.50X10"6 mg/s
The annual emission rate of the contaminant is 3.50xlO"6 mg/s.,
For the Nebraska example, the annual average emission rate of
cadmium is calculated by:
R10
a
8-13
-------
where:
A
a
E
400,000
-4
10
— 1.39x10 mg Cd/g soil
- 0.0083 g/hr/m2
R
10
400,000 m2 • 1.39X10"4 mg/g • 0.0083 g/hr/nr • 1 hr/3600 s
-4
1.28x10 mg/s
8.4.3. Worst-Case 24-hr Emission Rate. The worst-case 24-hr
emission rate of contaminant on PM10 particles (R10) is the product
of the area of the field, the pollutant soil concentration, and the
emission factor of PM10 particles.
R10 = A • a - E10
(Equation 8-5)
10
worst-case 24-hr emission rate (mg/s)
area of the source (m2)
soil concentration of the contaminant (mg/g)
where:
R
A
a -
E10 = worst-case 24-hr emission factor (g/hr/m£)
The difference between the worst-case and mean annual emission rate
is the emission factor (E10) . The worst-case emission factor is
derived using the maximum 6-hr wind speed whereas the mean annual
emission factor is based on the average wind speed.
For the first example, the field in western Florida is assumed
to have an area of 10,000 m2. The soil concentration of cadmium is
assumed to be 4.06xlO"4 mg/g as previously calculated. For the
second example, the field in Nebraska, the area is assumed to be
400,000 m2. The soil concentration is assumed to equal 1.39xlO~4
mg/g.
8-14
-------
The 24-hr emission factor is calculated by:
E10 = 0.036 • (1 - V) • [u, hr]
(Equation 8-6)
where:
V
= fraction of contaminated surface vegetation cover
= expected maximum 6-hr mean wind speed during the
year (m/s)
The worst-case 24-hr emission factor is based on the expected
maximum 6-hr wind speed during the year [u6.hr] • The expected
maximum 6-hr mean wind speed during the year is calculated by:
[U6.hp] = [u+] - 2 m/s (Equation 8-7)
where:
[U*] = annual fastest mile (m/s)
The annual fastest mile reflects the magnitude of wind gusts; it
is speed of the fastest whole mile of wind movement that passed by
the 1-mile contact anemometer (U.S. EPA, 1985a).
For the first example, the field is assumed to be bare, so V
equals 0. The fastest mile in Tampa, Florida is 22.2 m/s (U.S.
EPA, 1985a) , so the expected maximum 6-hr mean wind speed is
calculated using equation 8-7.
Cu6-hr] = tu*3 - 2 m/s
where:
[u+] =22.2 m/s
[u6_hr] =22.2 m/s - 2 m/s
= 20.2 m/s
Thus, the 24-hr emission factor for the field in western Florida
is calculated using equation 8-6.
E10 = 0.036 » (1 - V) • [u6.hr]3
8-15
-------
where:
V =o
Cu6-hrl " 20.2 m/s
E10 = 0.036 • (1-0) • [20.2 m/s]3
= 297 g/hr/m2
For the second example, the field was assumed to be bare, so
V equals 0. The fastest mile for North Platte, Nebraska is 27.7
m/s (U.S. EPA, 1985a) so the expected maximum 6-hr mean wind speed
during the year is calculated by:
where:
[u+] =27.7 m/s
Cu6-hr] =27.7 m/S - 2 m/S
=25.7 m/s
Thus, the 24-hr emission factor for the field in Nebraska is
calculated by:
E10 = 0.036 . (1 - V) • (U6.hr)
where:
V =o
Cu6-hr3 =25.7 m/S
E10 = 0.036 • (1-0) • (25.7 m/s)3
= 611 g/hr/m2
When the area, soil concentration, and E10 are known for a
particular source, the 24-hr emission rate may be estimated using
Equation 8-5. For the field in western Florida, the 24-hr emission
rate is calculated by:
8-16
-------
R10 - A
=•10
where:
A
a
E
= 10,000 m'
-4
10
R10 =
= 4.06x10 mg Cd/g soil
297 g/hr/mr
10,000 m2 • 4.06xlO"4 mg/g
297 g/hr/m^ • 1 hr/3600 s
"1
= 3.35xlO" mg/s
For the field in Nebraska, the 24-hr emission rate is
calculated by:
R10 = A • a • E10
where:
A
a
E
"4
10
= 400,000
= 1.39xlO" mg Cd/g soil
= 611 g/hr/mr
R
"4
10
= 400,000 m • 1.39xlO" mg/g • 611 g/hr/m • 1 hr/3600 s
= 9.44 mg/s
8.4.4. Results. The results of these calculations indicate that
the emission rate of cadmium in resuspended dust may be similar to
the stack emission rate of the incinerator (1.9xl04 jig/s) depending
on the scenario. For the long-term exposure assessment, the mean
annual emission rate for each example is at least 3 orders of
magnitude less than the stack emission rate even though "worst-
case" situations are assumed. Each field is assumed to have no
vegetative cover to provide a conservative estimate of dust
release. If vegetation is present, the resulting emission factor
8-17
-------
and rate will decrease. The mean annual emission rates based on
this conservative assumption are 3.50xlO"6 mg/s (western Florida)
and 1.28xlO~4 mg/s (Nebraska) .
For the short-term assessment, the 24-hr emission rate for
each example is greater than the stack emission rate. By assuming
no vegetative cover is present at either site, the worst case 24-
hr emission rates are found to be 0.335 mg/s (western Florida) and
9.44 mg/s (Nebraska).
To determine the extent of human exposure to the resuspended
dust, the emission rates, either mean annual emission rate or 24-
hour emission rate, must be incorporated into an air dispersion
model. The model uses this input along with site-specific
meteorology data (i.e., wind speed and cloud cover) to predict
ambient air concentrations.
For cases in which the emission rates are less than the stack
emission rate and the human health risk posed by inhalation of the
pollutants emitted from the stack is small, the analyst may chose
not to run the air dispersion model since the ambient air
concentrations would be less than or equal to the ambient air
concentrations emitted from the combustor. The cancer risk caused
by the inhalation of the resuspended pollutants by wind erosion
would be less than that caused by the inhalation of the stack
emissions.
8-18
-------
8.5. CONCLUSION
When determining exposure to pollutants resuspended in the
soil, the amount of vegetation, soil concentration, size of the
source, wind velocity, moisture content and soil roughness height
should be examined. Various methodologies have been developed to
determine the soil loss resulting from wind erosion. However in
applying these methods to deposited combustor emissions, the
analyst should review the approaches discussed in U.S. EPA (1985a),
Farino et al. (1983), Sehmel (1980), and Smith et al. (1982) to
select the most appropriate methodology for the exposure scenario.
The different types of exposure scenarios, whether long-term or
short-term, may yield emission rates differing by orders of
magnitude.
8-19
-------
-------
9. CALCULATING WATER CONCENTRATIONS
9.1. INTRODUCTION
Contaminants associated with particles emitted by combustors
are deposited on land and water downwind from the combustor at
rates determined by meteorology, terrain and particle physics.
Following deposition, these contaminants may be dissolved,
repartitioned or degraded. They may be transported over the ground
in runoff to a surface water body or infiltrate the ground and
recharge the groundwater. This chapter focuses on the contaminant
concentration found in surface water bodies, collected
precipitation and groundwater.
9.2. SURFACE WATER
The aim of the Surface Water Model is to estimate the
concentration of a given contaminant in a water body following
deposition onto a watershed. This model was developed for non-
volatile compounds. Estimated contaminant levels will be the
result of runoff and soil erosion, as well as direct deposition of
particulate matter onto the water body. The model is designed to
accommodate both long-term and short-term loading scenarios. The
surface water model follows a three-tiered approach beginning with
a simple, extremely conservative calculation (Tier 1), and
proceeding to more detailed calculations when necessary (Tier 2
and Tier 3). Tier 2 and 3 calculations differ only in the use of
literature input values (Tier 2) vs. site-specific (Tier 3) input
variables.
9-1
-------
The methodology derived to calculate risks from the surface
runoff pathway originally was developed to evaluate impacts from
the land application of municipal wastewater sludge. A detailed
discussion of the model is available in the document entitled
Development of Risk Assessment Methodology for Land Application
and Distribution and Marketing of Municipal Sludge (U.S. EPA,
1989b).
9.2.1. Overview of the Model. The Surface Runoff Model is used
to calculate the concentration of contaminant, both dissolved and
adsorbed to suspended particles, that accumulates in a surface
water body both from deposition directly onto the water body and
deposition onto the watershed followed by surface runoff. These
predicted concentrations are used to estimate human risk that may
result from drinking water, swimming and bathing in, or eating fish
from a contaminated surface water body.
Tier 1 of the Surface Runoff Model is a conservative screening
method that assumes that all contaminant deposited onto a
watershed/water body is transported to the receiving water body
during the loading period (i.e., 1 year for long-term loading and
24 hours for short-term loading). The contaminant concentration
in the receiving water is calculated from the annual mass of
contaminant in the fallout per unit area, the total area of the
watershed, and dilution volume of the water body under
consideration (see discussion in Section 9.2.2). The Tier 1 water
concentration is used to calculate daily intake due to water
ingestion, dermal absorption, and fish ingestion. These daily
9-2
-------
intakes are either compared with a reference dose for chronic oral
exposure (RfD0) or used to calculate a risk level as described in
Chapter 15. If these analyses show unacceptable excess risk, a
Tier 2 or Tier 3 analysis is required to determine a more reliable
estimate of the receiving water concentration.
Tier 2 and 3 calculations provide more detailed analysis of
the fate of contaminant in the watershed. Tier 2 and 3
calculations are identical; they differ only in origin of input
values. Tier 2 relies on input values found in the literature.
Tier 3 is based on site-specific data for input parameters. In
Tier 2 or 3, contaminant concentration in water is calculated for
long-term loading (i.e., 1 year) or short-term loading (i.e., 24
hours).
For long-term loading, the model assumes all contaminants
entering the receiving water are absorbed to eroded particles and
do not partition between soil particles and water. The contaminant
concentration in water is calculated from the mass of contaminant
per area of soil, the distribution coefficient (Kd) , the area of
soil eroded to the water body, and the appropriate dilution volume
of the water body in question.
For short-term loading, the model assumes that the contaminant
enters the receiving water body by direct deposition, absorbed to
eroded soil particles, and dissolved in the runoff water. Once the
contaminants enter the receiving water, they repartition between
the absorbed and dissolved state. The contaminant concentration
in water, dissolved and adsorbed, is calculated from the total mass
9-3
-------
of chemical in the water body, the fraction of chemical dissolved
or adsorbed, and the appropriate dilution volume of the water body
in question.
A number of assumptions about runoff generation and subsequent
mixing in the receiving water were required to formulate the risk-
based methodology. The key assumptions are listed in Table 9-1,
with a discussion of their impact on the methodology. Because the
science is not exact, the assumptions generally are conservative
(i.e., overpredict contaminant concentrations) . In some instances,
however, the nature of the effect of a given assumption may vary
with site-specific considerations.
9.2.2. Calculating Contaminant Concentrations for Tier 1. In Tier
1, all contaminant deposited onto a watershed, is assumed to be
transported to the receiving water in the loading period. The
total mass flux of contaminant is distributed among the watersheds
in which the fallout is deposited. The appropriate dilution volume
then is used to calculate the resulting concentration in the
receiving water:
[Fa • (WAL + WAH)]/Vfx • 106mg/kg • m3/103 L
(Equation 9-1)
9-4
-------
TABLE 9-1
Surface Water Methodology Assumptions
Functional Area
Assumption
Ramifications
Long-Term Concentrations:
Tier 1
Tier 2 and
3 General
Source
Area
All contaminant deposited
on an annual basis is
transported to the re-
ceiving water in a dis-
solved form.
Loadings to the receiving
water can be described as
a function of solids
loading.
Facility operates over a
period sufficient for sur-
face soil levels to reach
equilibrium where annual
losses equal annual inputs.
No settling of particles
in the deposition zone,
gross erosion reaches the
edge of field.
Short-Term Concentrations:
Tier 1
Tier 2 and
3 Source
Area
All contaminant emitted
in a year is lost in a
single runoff event.
No kinetically limited
release of contaminant
from residual/soil mix-
ture ; i.e., total con-
taminant concentration
is fully equilibrated
into adsorbed and dis-
solved phases.
Provides an extremely con-
servative estimate since no
losses are considered.
Mechanistically inappro-
priate for contaminants
with low partition coef-
ficients.
Overpredicts contaminant
loading by ignoring loss
mechanisms other than those
used in formulation, namely,
runoff and infiltration.
Maximizes contaminant loss.
Provides extremely conser-
vative estimate with no
provision for losses or
incomplete mobilization.
Maximizes contaminant con-
centration available for
runoff in dissolved phase,
thereby maximizing
loadings.
9-5
-------
TABLE 9-1 (cont.)
Functional Area
Assumption
Ramifications
Stream
Stream flow is unchanged
by the storm unless
arterial velocity data
are available from the
hydrograph.
Overestimates stream
concentration, since the
storm will increase stream
flow.
9-6
-------
where:
X
Ci
CiL
Fa
WAL
WAy
Vf
Sh1
= long- or short-term scenario
= concentration of contaminant in the receiving
water for the short-term loading scenario (mg/L)
= concentration of contaminant in the receiving
water for the long-term loading scenario (mg/L)
= annual mass of contaminant in fallout per unit
area (kg/km2-yr)
= land area of watershed receiving fallout (km2)
= area of water receiving fallout (km2)
= long-term dilution volume for water body
Vfx (river)
Vfx(lake)
under consideration (m /yr)
= mean annual flow (m /yr)
= mean lake volume (m3/v:r) + mean annual
outflow from lake (m3/yr)
Vf (estuary) = mean annual estuary flow (m/yr)
Equation 9-1 is appropriate for the long-term loading scenario.
For a short-term loading scenario, Vfx is replaced with VSX, the
short-term dilution volume. VSX for the river and estuary can be
set to Vfx times the short-term time frame (in years); however
ideally, VSX for a river or estuary would be obtained from the
hydrograph for the storm, if it is known. VSX for the lake equals
the mean lake volume.
This very conservative calculation accounts for no losses
during transport and, therefore, overpredicts contaminant levels.
If these overpredictions do not exceed health-based criteria (such
as the reference dose, RfD), the risks are deemed acceptable and
no further analysis is required. If the predictions exceed
criteria, a more detailed Tier 2 or 3 analysis is required to
determine probable receiving water concentrations. The predicted
water concentrations are also used for determination of excess risk
(ER) for carcinogens. The risk characterization for this exposure
is discussed in Chapter 15.
9-7
-------
9.2.3. Calculating Contaminant Concentrations for Tier 2 or 3
9.2.3.1. LONG-TERM EXPOSURE — For the long-term loading
scenario, receiving water contaminant concentration is calculated
from estimates of the contaminant mass per unit area of watershed,
the rate of sediment transport to the receiving water, and the
appropriate dilution for the water body in question (defined in
Equation 9-1) :
Ci
Lg23 ~ Cxe * WAt ' Mm/(Vfx • BD • Z) ] • 105 cm/km • km3/109m3
£ •» f
(Equation 9-2)
+ [Mw • 106 mg/kg • m3/103 L]
where:
Ci
L923
WAL
Mm
BD
Z
Mw
concentration of contaminant in the receiving
water for the long-term loading scenario (mg/L)
sediment loss rate per unit area watershed over
time (kg/km2-yr, see Equation 9-3)
land area of watershed receiving fallout (km2)
maximum contaminant mass per area of soil
(kg/km2, see Equation 9-4)
long-term dilution volume for water body under
consideration (m3/yr)
bulk density of soil (kg/m3)
depth of mixing (cm)
maximum contaminant mass per volume of water
(kg/km , see Equation 9-5)
The rate of sediment loss to the receiving water is computed
using the Universal Soil Loss Equation (USLE) (Wischmeier and
Smith, 1978):
-3
Xc = R • K • LS • C • Ps • 907.18 kg/ton • 1/4.07x10° acre/knT
(Equation 9-3)
9-8
-------
where:
R
K
LS
C
sediment loss rate per unit area watershed over time
(kg/km2-yr)
"erosivity" factor (yr~1)
"erodability" factor (tons/acre)
"topographic or slope length" factor (unitless)
"cover management" factor (unitless)
"supporting practice" factor (unitless)
Guidelines for selection of input factor values for the USLE
in each watershed are provided in Wischmeier and Smith (1978) and
in the runoff discussion of the U.S. EPA (1989b) . The input
parameters used are site-specific values. Xe is an average annual
value and is obtained by multiplying estimates of soil loss from
a rainfall and runoff for a geographic area by a series of ratios.
R is the erosion potential for average annual rainfall at a given
location. K is an experimentally determined value using the
predominant soil type as a guide in obtaining the value for K. LS
takes into account the effect of slope length and steepness, while
C looks at the vegetative cover, crop sequence, crop rotation and
tilling practices. Ps looks at the agricultural techniques such
as contouring and terracing.
In order to estimate contaminant mass per area of soil (Mm),
the combustors are assumed to be in operation long enough for a
steady state concentration to be reached. By definition, these
conditions will prevail when soil levels are high enough for the
sum of zeroth and first-order losses to equal the annual additional
contaminants in fallout. Maximum soil contaminant mass per area
of soil is calculated as:
9-9
-------
where :
Mm
Fa
Mm = Fa • [1 - expfk, • timef)/k,] (Equation 9-4)
maximum contaminant mass per area of soil
(kg/km2)
annual fallout rate for contaminant losses
(kg/km2 -yr)
first-order loss rate (see Equation 9-6)
'
timef
time span in which analysis will be carried
out, typically the life of the combustor and
its replacements (yr)
Since contamination from the combustion facility enters the
water body from both runoff and direct fallout, the contribution
from direct deposition must be accounted for in the model. An
equation similar to Equation 9-4 is used to estimate the
accumulation of the contaminant in the water body from direct
deposition:
Mw = Fa • [WAyVfJ (Equation 9-5)
where:
Mw
Fa
WAU s
Vf" =
maximum contaminant mass per volume of water (kg/m3)
annual fallout rate for contaminant losses (kg/km2-
area of water receiving fallout (km )
long-term dilution volume for water body under
consideration (m /yr)
The first-order loss rate, k,, can be calculated by adding
the loss rates due to infiltration (kn) , erosion (k1E) , and
degradation (k10) ; the equations for each of these loss rates is
presented below:
k. = k1r + k1E + k1D (Equation 9-6)
9-10
-------
6 • Z • [1 + (BD • Kd / 6) ]
(Equation 9-7)
where:
kn
IR
6
Z
BD
First-order loss rate for infiltration (yr"1)
Infiltration rate (cm/yr) ; set default value equal
to 1 cm/yr
Volumetric water content of the soil (mL/cm3)
Depth of incorporation (cm)
Soil bulk density (kg/m )
Distribution coefficient (m3/kg)
k1E =
where:
k
1E
where:
t'
«-ii
Kd • BD
BD • Z 9 + (Kd • BD)
] • 105 cm/km • km3/109 m
(Equation 9-8)
First-order loss rate for erosion (yr"1)
sediment loss rate per unit area watershed over time
(kg/km2-yr)
Depth of incorporation (cm)
k1D = ln2/t,A
(Equation 9-9)
first-order loss rate for degradation (yr"1)
contaminant half-life due to degradation in soil
(yr)
In general, volatilization would also represent a first-order
loss mechanism. For MWC particulate fallout, however, it is
assumed that volatile species will not be present in the solids
settling from the atmosphere. Also, when field measurements are
available for deriving degradation rates, they may reflect all
first-order losses (k,) and not just degradation (k1D) . Therefore,
care must be taken in selecting input values.
9-11
-------
9.2.3.2. SHORT-TERM EXPOSURE — For short-term loading
scenarios, the model calculates the dissolved and adsorbed
contaminant concentration in receiving water after a single storm
event.
To calculate the concentration of chemical present in either
dissolved or adsorbed form in the water body, the mass of chemical
per unit volume of water is divided by the appropriate dilution
volume of the water body in question:
and
CHEMW = CHEMyvS,,
CHEMs1 = CHEMa/X,
a' s
L (Equation 9-10)
(Equation 9-11)
where:
CHEMH *=
CHEMd -
vsx
CHEMS1 =
CHEMfl =
3
Xs
CHEM., =
CHEMS2 = CHEMa/VSx • m3/103 L (Equation 9-12)
mass of dissolved chemical per unit
volume of water (mg/L)
mass'of chemical present in the water body in
the dissolved form (mg)
appropriate dilution volume for the watershed
under consideration over the duration of the
storm event (HI )
mass of adsorbed chemical per mass of sediment in
the water body (mg/kg)
mass of chemical present in the water body in
the adsorbed form (mg)
mass of sediment in the water body (washed, off
from the watershed) (kg)
mass of adsorbed chemical per unit volume of
water (mg/L)
appropriate dilution volume for the watershed
under consideration - use Vfx when calculating
CHEMH for long-term exposure (m3)
9-12
-------
To determine the mass of chemical in the water body in either
the dissolved or adsorbed form, the following equations must be
solved:
and
CHEMd = TOTCHEM • ad • 106 mg/kg (Equation 9-13)
CHEMa = TOTCHEM • aa • 106 mg/kg (Equation 9-14)
where:
CHEMd = mass of chemical present in the water body in the
dissolved form (mg)
TOTCHEM = total mass of the chemical in the water body (kg,
see Equation 9-19)
<*d = fraction of chemical dissolved (uriitless, see
Equation 9-15)
CHEMa = mass of chemical present in the water body in the
adsorbed form (mg)
aa = fraction of chemical adsorbed (unitless, see
Equation 9-16)
In the remainder of this section, the calculations needed to
determine the fractions of dissolved and adsorbed chemical will be
presented first followed by the calculations needed to determine
the total mass of the chemical.
The fraction of chemical present in either the dissolved or
adsorbed state can be calculsited by the following equations:
1
and
+ (Kd • CJ
Kd • Cs
1 + (Kd • CJ
(Equation 9-15)
(Equation 9-16)
9-13
-------
where:
<*.
fraction of chemical dissolved (unitless)
distribution coefficient for the contaminant in the
soil water system (m3/kg)
mean sediment concentration in the water body
(kg/m3, see Equation 9-17)
fraction of chemical adsorbed (unitless)
The concentration of sediment in the water body during a storm
event can then be calculated using appropriate dilution volume of
the water body because the volume of soil that washes off the
watershed is negligible compared with the dilution volume. The
mean sediment concentration is calculated by the following
equation:
where:
s
vs.
(Equation 9-17)
mean sediment concentration in the water body
(kg/m3)
mass of sediment in the water body (washed off from
the watershed) (kg, see Equation 9-27)
the appropriate dilution volume for water body under
consideration over the duration of the storm event
(m3)
The total mass of the chemical in the water body is equal to
the amount of chemical present in the dissolved form in the
watershed plus the mass of chemical in the adsorbed form in the
watershed, or:
TOTCHEM = P
xt
qt
ft
(Equation 9-19)
9-14
-------
where:
TOTCHEM
pxt
P4t
pft
total mass of the chemical in the water body
(kg)
mass of chemical entering water body in
adsorbed form (kg, see Equation 9-21)
mass of chemical entering water body in
dissolved form (kg, see Equation 9-22)
mass of chemical deposited directly onto the
water body (kg, see Equation 9-20)
To determine the amount of chemical that is deposited directly
onto the watershed the following equation must be solved:
Pft = Fa
WA.,
(Equation 9-20)
where:
P
ft
Fa
WAH =
mass of chemical deposited directly onto the water
body (kg)
annual mass of contaminant in fallout per unit area
(kg/km2-yr)
time frame of the model, i.e., 0.002738 years for
the short-term loading scenario and 1 year for the
long-term loading scenario (yr)
area of water receiving fallout (km2)
The mass of contaminant entering the water body in adsorbed or
dissolved form can be calculated using the following equations:
pxt = [XS/(BD • Z) ] • Aa • 105 cm/km • 109 m3/km2
(Equation 9-21)
and
pqt =
Mt)3 ' Da • WAL
(Equation 9-22)
9-15
-------
where:
xt
BD
Z
Aa
Mt
Da
WAL =
mass of contaminant entering water in adsorbed form
(kg)
mass of sediment in the water body (washed off from
the watershed) (kg)
bulk density of the soil (kg/m )
depth to which tilling is practiced (cm), the
default value should be equal to 1
adsorbed contaminant mass per unit of land in
watershed (kg/km2, see Equation 9-23)
mass of contaminant entering water in dissolved form
(kg)
depth of runoff in the watershed (cm, see Equation
9-25)
depth of total rainfall for the storm event (cm)
depth of snow melt for the storm event (cm), set
default value equal to zero
dissolved contaminant mass per unit of land in
watershed (kg/km2, see Equation 9-24)
land area of watershed receiving fallout (km )
If the area under consideration is an agricultural area that
is likely to be tilled, soil depth is assumed to be 20 cm. In all
other cases, soil depth is assumed to be 1 cm.
To determine the mass of dissolved and adsorbed contaminant
entering the water body, the total mass of contaminant, Mm,
deposited on land surface must be partitioned between the adsorbed
portion, Aa, and the dissolved portion, Da. These are derived as:
Aa = [l/(l+(0/Kd • BD))]
Mm
(Equation 9-23)
and
Da = [l/(l+(Kd • BD/6)) ] • Mm (Equation 9-24)
9-16
-------
where :
Aa
6
BD
Mm
Da
adsorbed contaminant mass per unit of land in
watershed (kg/Jan2)
volumetric water content of the top cm of soil
(mL/ciir)
distribution coefficient for contaminant in soil
water system (m/kg)
bulk density of the soil (kg/m3)
maximum contaminant mass per unit area of soil
(kg/km2)
dissolved contaminant mass per unit area of land in
watershed (kg/km )
The runoff depth of carrying dissolved contaminant to the water
body is estimated by the following equation:
(Rt -f Mt - 0.2S)'
(Rt
M
0.8S)
(Equation 9-25)
where:
M..
depth of runoff in the watershed (cm)
depth of total rainfall for the storm event (cm)
depth of snow melt during the storm event (cm)
the watershed retention parameter (cm)
The watershed retention factor (S) is calculated from the Soil
Conservation Service (SCS) runoff curve number (CN) according to
the following equation:
S = 2.54 • [(1000/CN)-10]
(Equation 9-26)
The mass of sediment eroded from the watershed to receiving
water during a storm event is estimated with the Modified Universal
Soil Loss Equation (MUSLE):
9-17
-------
where:
LS
C
P
Xs = 2.04X106 • (Q • qp)°'56
K • LS • C • P
(Equation 9-27)
mass of sediment in the water body (washed off from
the watershed) (kg)
volume of runoff (km -cm, see Equation 9-28)
peak runoff (m3/s, see Equation 9-29)
"erodability" factor (tons/acre)
"topographic or slope length" factor (unitless)
"cover management" factor (unitless)
"supporting practice" factor (unitless)
Again, selection of K, LS, C and P is discussed elsewhere
(Wischmeier and Smith, 1978; U.S. EPA, 1989b). This equation is
an empirical relationship and the units must be consistent with
those shown above.
The storm runoff volume is calculated from the runoff depth
according to:
Q = WAL • DR (Equation 9-28)
where:
Q
WAL
DD
volume of runoff (km -cm)
land area of the watershed receiving fallout (km )
depth of runoff from the storm event (cm)
A trapezoidal hydrograph is assumed so that the peak runoff rate
can be calculated as:
qp
WAL - DR .
(Rt - 0.2S)
hr/3.6x!03s • km/105 cm • 109 m3/km3
(Equation 9-29)
9-18
-------
where:
P
WAL
DR
Rt
T..
peak runoff rate (m /s)
land area of the watershed (km )
depth of runoff in the watershed (cm)
depth of total rainfall for the storm event (cm)
duration of the storm event (hrs)
water retention parameter (cm)
Example calculations for this model are presented in Section
9.5.
9.3. PRECIPITATION
In some remote areas, rain and/or snowfall are collected off
surfaces and stored in cisterns to be used as a source of drinking
water. Individuals in these areas may be exposed to pollutants
deposited on surfaces downwind from the combustor from which the
water is collected.
9.3.1. Water Concentration (We). The water concentration of
pollutants (mg/L) found in collected precipitation can be
determined as follows:
where:
We
Dy
P
We =
Dy
water concentration (mg/L)
total deposition (g/m2/yr)
annual precipitation (mm/yr)
(Equation 9-30)
9-19
-------
The value for water concentration obtained from this equation is
used as an input variable in calculations to determine the human
health risk associated with dermal absorption and ingestion of
water.
9.3.1.1. Total Deposition (Dy). The total deposition is from
both wet and dry site-specific deposition rates, which are
determined by air-dispersion and deposition modeling. Methods for
determining areal-averaged deposition rates are described in
Chapter 3.
9.3.1.2. Annual Precipitation (P). Site-specific values for
the annual precipitation are available in Baes et al. (1984).
9.4. GROUNDWATER
Once contaminated particles emitted from combustors are
deposited on surfaces downwind from the facility, the contaminants
may become dissolved in precipitation such as mist, rain, hail,
sleet and snow. The dissolved portion can either form surface
runoff as described in Section 9.2 or infiltrate the ground as
leachate and recharge the groundwater. Individuals will be exposed
if the contaminated groundwater is used as a source of drinking
water.
In determining the risk from exposure to groundwater, a number
of factors affect leachate generation and subsequent transport, of
contaminants in the unsaturated zone. Some factors that affect
leachate generation are the physicochemical characteristics of the
9-20
-------
emitted chemical, the magnitude of precipitation and environmental
transport. Infiltration of the leachate can be affected by
thickness of the saturated layer, moisture content of the soil,
degree of compaction, macrostructure of soil, vegetative cover,
temperature, and entrapped air (U.S. EPA, 1988a). For metals, the
contaminant concentration in the groundwater is affected by
geochemical reactions. For organics, the contaminant concentration
is altered by degradation as it travels through the unsaturated
zone.
A Groundwater Infiltration Model (GIM) was developed to model
contaminant movement to groundwater. The model, described in U.S.
EPA (1987c), was used to make preliminary determinations of the
potential for groundwater contamination under worst-case
circumstances by several inorganic and organic contaminants. These
findings indicated very limited potential for contamination.
Therefore, further evaluation of this exposure route is considered
unnecessary. This section summarizes the GIM and the results of
the preliminary calculations.
9.4.1. Overview of the Model. The GIM is formulated in three
successive tiers (see Figure 9-1), which begin with simple but very
conservative estimates and proceed to more detailed analyses if the
first tiers predict unacceptable risks. Groundwater contaminant
concentrations (absorbed and dissolved fractions) predicted by the
model are compared with health-based water criteria to ascertain
if a potential human health risk exists.
9-21
-------
Health
Criteria
Distribution
Coefficient
Loss Rate
Inorganic
Contaminant
Gcochemical
Considerations
FIGURE 9-1
Logic Flow for Groundwater Pathway Evaluation
Calculate Contaminant Concentration
as Fallout Flux/Recharge
Tierl
No
Determine Time of Travel and
Losses in Unsaturated Zone
Determine Concentration in the
Aquifer Based on
MINTEQ Curves
Is
(X:)*> Health'
Criteria?
No
END
* (X.) = Concentration of Contaminant i
END
Tiers
Experimentally Determine
Retardation and Degradation
Values
Tier 2
Yes
9-22
-------
9.4.1.1. TIER 1 — Tier 1 is an extremely conservative
approach in which projected leachate concentrations are compared
with health-based criteria. Leachate concentrations are predicted
on the basis of annual fallout and recharge rates. It is assumed
that soil contaminant levels will increase over time until the
contaminant concentration in the leachate is sufficiently high to
deplete the input from fallout each year. If the contaminant
concentration in the leachate is below the relevant health-based
criterion, the contaminant can be eliminated from further
consideration. If the contaminant concentration in the leachate
exceeds the criterion, the contaminant is carried forward to a Tier
2 or 3 analysis.
9.4.1.2. TIER 2 and 3 — As in Tier 1, the Tier 2 and Tier
3 analysis begins by calculating the contaminant concentration in
the leachate. Tier 2 or 3, however, allows for site-specific
inputs to predict dispersion, degradation, and retardation effects,
which reduce resultant exposure levels. The difference between
Tier 2 and 3 lies in the number of input values that are determined
experimentally. In Tier 3, degradation rates and retardation
coefficients are measured directly.
The initial step in Tier 2 or 3 is to define the concentration
of contaminant in recharge water. As with Tier 1, this is
accomplished by assuming that equilibrium will be reached when the
inputs from fallout each year are transported away from the soil
by leachate. Subsequently, the time required for the leachate to
move downward through the unsaturated zone to the aquifer is
9-23
-------
determined using factors such as field moisture content, depth to
the saturated zone and recharge rate. For multiple layer systems,
a travel time to the aquifer is calculated for each layer and the
total summed across the unsaturated zone. The total time it takes
for the leachate to reach the aquifer and the total depth of the
unsaturated zone are used to derive the equivalent velocity of the
leachate. The velocity of contaminant traveling through the
unsaturated zone can be retarded if the contaminant interacts with
the soil in transit.
9.4.2. Determining Leachate Contaminant Concentration. The U.S.
EPA has used the GIM to evaluate possible health effects from
exposure to groundwater contaminated with several metals and
organics emitted by municipal waste combustors (U.S. EPA, 1987c).
Exposures were modeled for a facility in Hampton, Virginia and a
hypothetical model plant in western Florida, chosen to represent
a reasonable worst-case scenario at an existing facility and a
scenario for planned facilities, respectively. Tables 9-2 and 9-
3 show the deposition rate of emissions for these facilities when
either a fabric filter or electrostatic precipitator were used as
control technology in the U.S. EPA analysis. The resulting
concentration of the leachate in the aquifer and time for the
leachate to reach the aquifer for these scenarios are also shown.
For all the chemicals, the groundwater pathway was an insignificant
source of exposure as further described below.
9-24
-------
og
nics
(0
&>
rH
o
o
H
0) '
T3
0
g
O
•H •
(0
•H
MH
C
H
r-l
Q)
4->
(0
J5
c
o
rH
O
01
C
,_J
*rt
I*
rH ?*
0 i
rH g
X* r«
r\ >-H
^1 Jr
i-H ."3
ftQ
•O
0)
1 j
(tf
c
rH *"
0 >1
rH g1
o 2
>.-§,
ft m
0)
•0
•s
•a
r-H
(0
LJ
0
fe
•O
(1)
4->
(0 U]
C 0)
•H C
rH (1)
O N
rH C
X! a)
O 03
01
(U
a
rH
>
rH
Q)
T)
O
^^
rH
1
(0
JH
"*s^
O»
^
*^-*
*-*
(0
fe
"—
0)
4J
«
o
•H -
4J C
-H O
Ul 4->
B&
0) (0
Q «
tn *»
'o 'o
H rH
X X
r** I***
o in
H H
••o -j-
'o 'o
H H
X X
o co
H H
CO H
m CM
'o 'o
H H
X X
H (N
G\ O
<* r^
-* K>
O O
H H
X X
CO O1
>* in
^1* VO
4->
^•^ (~*
ft >>
rH O rH
0) -H rH Q)
4-> 4-> O T3
rH 10 4J 0
•H 4-> (0 g
fe Ul 4->
O -H rH
U rH ft (0
•H 4-> -H O
rH O O -H
J3 0) Q) 4.)
(0 rH rH (1)
fv. ivi rt . fH
4J
o
&
^
'o
H
X
CO
m
«
<0
'o
i-l
X
CT>
H
CM
0
rH
X
f-
r^
H
tn
'o
H
X
fs^
n
vo*
rH
0)
4J
rH
•H
fe
o
(0 -H
T3 rH
•H XI
IH <0
O fe
pH
fe
•H
•o
'o
H
X O O O O
0
O
in
(n
'o
i-H
X O O O O
CO
M
'S
X o o o o
r-
VO
CM
** M (NJ t-
** e\j t\j INJ cj
'o 'o 'o 'o 'o
H H H H c-H
X XX XX
co r-» r- 01 co
m HO H in
O\ *& vo O1 *J*
^^_ ^^
CM 01
0) )H
4J >H 4J >i
^(00) -. C ^ —
ft X! -H p-i nj p-i
W O 4-> W rH W rH
W CO •"* W OH WO)
^-* fl) ^*^ ^HM' (1 1
O (4—. ^O rH rHO -H
•H }H iJ 0) -H ,H Q) Q) -H M p
4JO4JOT) 4J4JQ D
I04-) OtJi rH ffl4-> O rH (04J i<
4JIO g -H4->(Og -H4->(8
U14JC— ' fe014-> fe014JX!
O -H O O -H rH O -H O
rHft-H4->< O rHft (0 O rHft (0
4-1 -H 4-> C > -H 4-> -H O (0 -H 4-) -H Q)
OO (OcO ^ UO -H-O rH GO «
0)0) rHC* XI -H X> 0)0)
rHM 4J-HC Id • rH M 0)rH (0 (HrH O
fji CL C 6 O PI^ &3 P^i .£^ O pfj fjj p. | i
(U (0 4-> -P rH
O 4-> ft O fe 0)
C C S ft g
O O (0 >i C -H
0 U W K -H EH
O O
o o
0 0
CO VO
O) vo
in n
o o
O 0
O H
VO O
H CO
H
H [^
in <*
CM H
o o
> VD
o n
•* CM
4J
q
(0
i-H
Pk
iH
Q)
•a
o
g
rH
< a
> u to
•H T3
4-> -H
C (DM
4-J 4-J i — 1
ft 0 fe
e ft
(0 >-. C
K K -H
9-25
-------
•K
t-i
1
oj
O
a*
CM «- _ _
o 'o 'o 'o
H rH H H
XX XX
a\ I-H n co
o r*- co m
H H U) CO
'o 'o o 'o
rH H t-H H
XX XX
o o M n
n TJ- m a\
oa VD co 4-> O O 4J+JOC
•P r-j ffl4J B H ifl4J
f8 -H4JrO -H4JIOC
K to HI 4> rl CX.014JO
gg 0 S'g, gcO 0 82 '41
O -r) 4J -H -rl HJ -rl +J -H CO
•H « r) OO 4J-H M OO M
•PC # a Q di^, a Q Q ^
•HO IB rHr) J30 CO rHr) C
OT-P to WrH -PrH pc< P4Pl Q)
O ft O fe O
M R P[ f"
3 W S *H O
MM C\J (M
o o 'o 'o
H H H H
XX XX
vo co H in
CO H ^f 01
CM CO H CO
'o 'o 'o 'o
rH rH H rH
XX XX
^1* CO Ol t-H
rH rH VO Ol
'o 'o 'o 'o
rH H H H
XX XX
•* CO H t"*
O VO *4* Ol
VO rH H CO
Kl Kl to OJ
o o o 'o
rH H H t-H
XX XX
TJ* t-, vo in
•* CO H tt
co 01 in H
r- N. \f> *t
o o 'o 'o
H H rH H
XX XX
•* CM VO H
CM rH CO in
CM VO in H
in
M
4-1 >
^-, nj ,
fit r-H fit M
W Ot co OJ
H W O MCt OCO O M& Q)
10 -rl p -H .rl -O -H 4J -H «
C» M OO 4-> -H M OO
rICX) 0) 01 (DMA 0) CD O
EO 10 rHM .CO l3 rHM 4J
KJ 4-> lit W Ot 4JrH fct M ft
p gt o lit a)
UK K -rl EH
0 0
O VO
H in
CM H
O 0
o o
0 0
o o
o o
t- t--
H H
vo o
CO O
CO CO
CO CO
o o
o o
CO CO
t^ CO
^
10
,— {
Hampton, VA
Hypothetical Mode]
in Florida
*Tier 3
9-26
-------
9.4.2.1. ORGANICS—- For organic chemicals (Table 9-2),
the concentration in the leachate (Tier 2) reaching the aquifer
was negligible. Degradation of organics in the unsaturated zone
was the primary reason that only trace amounts of the organic
contaminants reached the aquifer. Furthermore, except in a few
cases, these contaminants took an extremely long time (thousands
of years) to reach the aquifer. In the instances where the organic
compound was predicted to reach the aquifer in <1000 years, the
concentration was predicted to be so small that it was estimated
as zero.
9.4.2.2. METALS — Metal contaminants (Table 9-3) in the
leachate do not degrade; however, they move more slowly than
organic contaminants because of adsorption and precipitation
reactions with soil (Freeze and Cherry, 1979). Although metal
concentration in the aquifer may be above the health-based criteria
(as with cadmium), a significant amount of the metal is adsorbed
to suspended solids, which can be filtered from water. Therefore,
these metals would not contribute to human exposure. Furthermore,
the model predicted thousands of years before the estimated
concentrations would reach the aquifer.
9.4.3. Conclusion. Contaminants in combustor emissions may
dissolve, infiltrate the ground and become available for recharging
the groundwater. However, as shown by the U.S. EPA analysis
discussed above, these contaminants degrade and move slowly through
the unsaturated zone to the aquifer. Thus, groundwater is not
9-27
-------
likely to be a significant route for human exposure to contaminants
from deposited combustor emissions. Risk assessors may want to
determine contaminant concentration in groundwater using site-
specific values to assess the contribution of groundwater to human
exposure at a particular site.
9.5. EXAMPLE CALCULATIONS
This section illustrates the methods for determining
benzo(a)pyrene and cadmium concentrations in surface water (Surface
Water Model, Section 9.2) for three different exposure scenarios:
A, B and C. In determining the water concentration of a
contaminant, the deposition area used differed in the case of each
exposure scenario. In Scenario A, the area of the watershed from
the incinerator is within a 50 km radius. In Scenario B, the
watershed is within a 5 km radius and in Scenario C the watershed
is at the point of maximum deposition at 0.2 km, as discussed in
Chapter 2. The scenarios also differ in the operating lifetime of
the incinerator. A summary of the calculations for all three
scenarios can be found in Table 9-6 for benzo(a)pyrene, and Table
9-9 for cadmium.
9.5.1. Benzo(a)pyrene. The values chosen for the input
variables of the Surface Runoff Model are given in Table 9-4. The
concentration of a contaminant in the water is a function of its
deposition rate (Fa) and the lifetime of the facility (timef) . The
9-28
-------
TABLE 9-4
Input Variables for Determination of Benzo(a)Pyrene
Water Concentrations
Variable
Value
WAL (watershed area) (km2)
WAy (water body area) (km )
Vfx (m3/yr)
BD (bulk soil density) (mg/m )
Z (soil till depth) (cm)
R (yr'1)
K (tons/acre)
LS (unitless)
C (unitless)
P (unitless)
k (yr )
6 (mL water/ cm3 soil)
t,/z (yr)
Kd (mykg)
Rt (event rainfall) (cm)
Kt (event snow melt) (cm)
CN (SCS Curve Number)
Tr (storm duration) (hrs)
VSX (storm stream flow volume) (m3)
Time (event) (yr)
IR (cm/yr)
15.0
1.0
3.15X107
1500
1.0
400.0 ;
0.21
0.179
0.5
1.0
0.0
0.22 .!
2.49
120
5.0
0.0
78.0
6.0
/
8.63X104
2.74X10"3
1.0
9-29
-------
values used for these parameters in the three different scenarios
(A, B and C) are given in Table 9-5. Sample calculations for each
scenario follow.
9.5.1.1. SCENARIO A — Tier 1 values for the receiving water
concentration for Scenario A can be calculated directly from the
input variables in Tables 9-4 and 9-5. The concentration of
benzo(a)pyrene in the receiving water for long-term loading in Tier
1, Citgi (m9/k) / can be calculated using Equation 9-1 and converting
to the proper units:
CiL91 "
6.51xlO~5 kg/km3-yr • (15 km2 + 1 km2)
3.15xl07 m3/yr
106 mg m3
kg
103 L
CiLg1 = 3.31X10"8 mg/L
The concentration in the receiving water for short-term
loading in Tier 1, Cish1 (mg/L), is obtained by replacing Vfx with
VSX in Equation 9-1:
Ci
6.51X10"5 kg/km3-yr • (15 km2 + 1 km2)
Sh1
8.63xl04 m3/yr
Ci
sh1
1.21X100 mg/L
106 mg m3
kg
103 L
9-30
-------
TABLE 9-5
Annual Fallout and Facility Life for Benzo(a)Pyrene
Scenario Facility Life (timef, yr) Fallout (Fa, kg/km-yr)
A
B
C
30
60
100
.51X10
-5
9.17x10
-4
2.42X10
-2
9-31
-------
For Tier 2 or 3 calculations, variables that are needed in
subsequent equations must first be obtained. These are S (Equation
9-26), DR (Equation 9-25), Q (Equation 9-28) and qp (Equation 9-
29) :
S = 2.54 • [(1000/78)-10] = 7.16
[5.0 cm + 0 cm - (0.2 • 7.16 cm)]'
5.0 cm+ 0 cm+ (0.8 • 7.16 cm)
= 1.19 cm
Q - 15 knT • 1.19 cm =17.8 km-cm
km 109 m3
15 km2 • 1.19 cm • 5 cm hr
• •———- • ———^^^^— •
6.0 hr • [5.0 cm - (0.2 • 7.16 cm)] 3. 6xl06s 105 cm km3
qp = 11.6 m/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
Xs = 2.04X106 • (17.8 • 11.55)0'56 • (0.210) (0.179) (0.5) (1.0)
Xs ^ 7.56X105 kg
Calculations for the long-term loading in Tier 2 or 3 can now
be initiated. The rate of sediment loss to the receiving water is
computed from the Universal Soil Loss Equation (Equation 9-3):
9-32
-------
Xe =
400 0.21 ton 907.18 kg acre
• • (0.179) (0.5) (1.0) •
yr acre
ton 4.047X10"3 km2
Xe = 1.68x10° kg/km-yr
The first order loss rate, k, (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, kn (Equation 9-7),
erosion, k1E (Equation 9-8) and degradation, k1D (Equation 9-9) :
1 cra/yr
11 0.22 mL/cm3 • 1.0 cm • [H-(1500 mg/m3 • 120 m3/kg)/0.22 mL/cm3)]
k1t = 5.56x10"° yr"1
k1E =
1.685xl06 kg/km-yr
120 m3/kg • 1500 mg/m3
1500 mg/m3 • 1.0 cm 0.22
mL/cm3 + (120 m3/kg • 1500 mg/m3)
k1E = 0.112 yr
-1
In2
2.49
= 0.278 yr
-1
k., = 5.56xlO"6 yr"1 + 0.112 yr"1 + 0.278 yr"1 = 0.391 yr"1
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady state concentration to be
reached:
9-33
-------
1 - exp(-0.39lyr~1 • 30.0 yr)
Mm = 6.51X10"5 kg/km2-yr •
Mm = 1.67x10'* kg/km2
The contaminant also enters the water body by direct
deposition (Equation 9-5):
Mw
Mw
6.51X10"5 kg/km2-yr •
2.07X10'12 kg/ra3
1 km2
3.15xl07 m3/yr
For long-term loading, the receiving water concentration can
now be calculated using Equation 9-2:
Ci
Lg23
1.68X106 kg/km-yr • 15 km2 • 1.67xlO"4 kg/km2
3.15X107 m3/yr • 1500 kg/m3 • 1.00 cm
103 cm
km
km3
106 m3
+ 2.07X10"12 kg/m3
106 mg m3
kg
103 L
CiLg23 - l.lOxlO"8 mg/L
For short-term loading, the Tier 2 and 3 model calculates the
receiving water concentration after a storm event. It takes into
account any repartitioning of the contaminant upon dilution. This
is initiated by calculating the mean sediment concentration in the
water body, Cs, Equation 9-17:
7.56xl05 kg
8.63X104 m3/yr
8.77 kg/m3
9-34
-------
The fraction that would be present in the dissolved, ad
(Equation 9-15), and adsorbed, aa (Equation 9-16), state can now
be determined:
1
aa =
1 + (120 m3/kg • 8.77 kg/m3)
120 m3/kg • 8.77 kg/m3
1 4- (120 m3/kg • 8.77 kg/m3)
= 9.50x10
-4
0.999
After a storm event, the amount of contaminant entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23), and dissolved, Da (Equation 9-24), state:
Aa - [1-f- (1 -+
0.220 mL/cia
120 m3/kg • 1500 kg/m3
) ] • 1.67xlO"4 kg/km2
Aa = 4.44xlO"5 kg/km2
Da = [1 -3- (1. +
120 m3/kg • 1500 kg/m3
0.220 mL/cm
)] • 1.67X10"4 kg/km2
Da = 2 . 04xlO'10 kg/km2
The amount of chemical deposited directly onto the water body
is a function of the deposition rate, area of the water body, and
time (Equation 9-20):
-3
Pft = 6.51X10"3 kg/km-yr • 2.74xlO~J yr • 1.0 km'
-7
P,t = 1.78x10"' kg
9-35
-------
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, Pqt (Equation 9-22), is obtained as follows:
7.56xlOs kg
•xt
•xt
•qt
1500 mg/m • i.oo cm
2.24xlO"6 kg
4.44X10"5 kg/km2 •
105 cm km2
km
io9 m3
1.19 cm
(5.0 cm + 0.00 cm)
• 2.04X10"10 kg/km2 • 15 km2
Pqt - 7.24X10"10 kg
The total amount of the chemical in the dissolved state,
CHEMd (Equation 9-13), and the adsorbed form, CHEMa (Equation 9-
14) , can be determined by first calculating the total amount of
the contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the distribution factors:
TOTCHEM = 2.24X10"6 kg + 7.24X10"10 kg + 1.78xlO"7 kg
-6
TOTCHEM - 2.42x10"° kg
CHEMd = 2.42x10"° kg • 9.50X10"4
CHEMd = 2.30X10"3 mg
CHEMa ^ 2.42X10"6 kg • 0.991
CHEMa = 2.42 mg
10° mg
kg
10° mg
kg
9-36
-------
The final concentrations for the short-term Tier 2 or 3
scenario can now be determined. CHEMU, the dissolved concentration
of the contaminant in the water body in question is obtained from
the dissolved mass and the appropriate dilution volume (Equation
9-10):
-3
CHEMW =
2.30x10 J mg
8.63X104 m3
103 L
CHEMH = 2.66X10"11 mg/L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
CHEMS1 =
2.42 mg
7.56xlOs kg
-6
CHEMs1 = 3.20x10"° mg/kg
The mass of adsorbed contaminant per volume of water, CHEMs2
(Equation 9-12) :
2.42 mg
m
CHEMS2 =
8.63X104 m3 103 L
CHEMs2 = 2.80X10"8 mg/L
9.5.1.2. SCENARIO B ~ Tier 1 values for the receiving water
concentration of benzo(a)pyrene for the B scenario can also be
calculated directly from the input variables in Tables 9-4 and 9-
5. The concentration of benzo(a)pyrene in the receiving water for
9-37
-------
long-term loading in Tier 1, CiLg1 (mg/L), can be calculated using
Equation 9-1 and converting to the proper units:
Ci
9.17X10"4 kg/km3-yr • (15 km2 + 1
106 mg m3
LSI
3.15xl07 m3/yr
kg 105 L
-7
CiLgl « 4.66x10"' mg/L
The concentration in the receiving water for short-term
loading in Tier 1, Cish1 (mg/L) , is obtained by replacing Vfx with
VSX in Equation 9-1:
Ci
sh1
9.17X10'4 kg/km3-yr • (15 km2 + 1 km2)
8.63X104 m3/yr
106 mg m3
kg 103 L
Cish1 « 1.70X10"4 mg/L
For Tier 2 or 3 calculations, variables that are needed in
subsequent equations must first be obtained. These are S (Equation
9-26), DR (Equation 9-25), Q (Equation 9-28) and qp (Equation 9-
29):
S - 2.54 • [(1000/78)-10] = 7.16
[5.0cm+0cm- (0.2 • 7.16 cm)]'
5.0 cm+ 0 cm+ (0.8 • 7.16 cm)
= 1.19 cm
Q = 15 km2 • 1.19 cm =17.8 km2-cm
9-38
-------
km 10y m3
15 km2 • 1.19 cm • 5 cm hr
P 6.0hr • [5.0 cm- (0.2 • 7.16 cm)] 3.6xl06 s 10s cm km3
qp = 11.6 m3/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
X, = 2.04X106 • (17.8 • 11.55)0'56 • (0.210) (0.179) (0.5) (1.0)
Xs = 7 . 56X10' kg
Calculations for the long-term scenario can now be:initiated.
The rate of sediment loss to the receiving water is computed from
the Universal Soil Loss Equation (Equation 9-3):
400 0.21 ton 907.18 kg acre
• • (0.179) (0.5) (1.0) •
yr
acre
ton 4.047xlO"3 km2
Xe = 1.68X106 kg/km-yr
The first order loss rate, k., (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, ku (Equation 9-7),
erosion, k1E (Equation 9-8) and degradation, k1D (Equation 9-9) :
1 cm/yr
0.22 mL/cm3 • 1.0 cm -[1+(1500 mg/m3 • 120 m3/kg)/0.22 mL/cm3)]
k,, = 5.56X10"6 yr"1
9-39
-------
1.685xl06 kg/km-yr
1500 mg/m3 • 1.0 cm
120 m3/kg • 1500 mg/m3
0.22 mL/cra3 +(120 m3/kg • 1500 mg/m3)
k1E = 0.112 yr
-1
In2
2.49
= 0.278 yr
-1
k, - 5.56xlO"6 yr"1 + 0.112 yr"1 + 0.278 yr"1 = 0.391 yr"1
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady state concentration to be
reached:
-1
Mm = 9.17X10"4 kg/km2-yr •
Mm - 2.35xlO"3 kg/km2
1 - exp(-0.391 yr ' • 60.0 yr)
0.391 yr
-1
The contaminant also enters the water body by direct
deposition (Equation 9-5):
Mw = 9.17X10"4 kg/km2-yr •
Mw = 2.91X10"11 kg/m3
3.15X107 m3/yr
For long-term loading, the receiving water concentration can
now be calculated using Equation 9-2:
9-40
-------
Ci
Lg23
1.68X106 kg/km-yr • .15 km2 • 2.35xlO~3 kg/km2
3.15X109 m3/yr•• 1500 kg/m3 • 1.00 cm
10J cm
km
km5
106 m3
-7
4- 2.91X10"11 kg/m3
106 mg
kg
m
103 L
CiLg23 = 1.55x10 mg/L
For the short-term loading scenario, the Tier 2 and 3 model
calculates the receiving water concentration after a storm event.
It takes into account any repartitioning of the contaminant upon
dilution. This is initiated by calculating the mean sediment
concentration in the water body, Cs, Equation 9-17:
7.56X105 kg
8.63x10* m3/yr
== 8.77 kg/m5
The fraction that would be present in the dissolved, ad
(Equation 9-15) , and adsorbed, ota (Equation 9-16) , state can now
be determined:
1 + (120 m3/kg • 8.77 kg/m3)
9.50x10
-4
120 m3/kg • 8.77 kg/m3
1 + (120 m3/kg • 8.77 kg/m3)
= 0.999
After a storm event, the amount of contaminant entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23), and dissolved, Da (Equation 9-24), state:
9-41
-------
Aa
0.220 mL/cm
120 m3/kg • 1500 kg/m3
) ] • 2.35X10"3 kg/km2
Aa - 6.26xlO"4 kg/km2
Da
l +
120 m3/kg • 1500 kg/m3
0.220 mL/cm
)] • 2.35xlO"3 kg/km2
Da = 2.87X10"9 kg/km2
The amount of chemical deposited directly onto the water body
\
is a function of the deposition rate, area of the water body, and
time (Equation 9-20):
Pfc « 9.17X10"4 kg/km2-yr • 2.74X10"3 yr • 1.0 km2
Pft « 2.51X10"6 kg
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, Pqt (Equation 9-22), is obtained as follows:
7.56X105 kg
3
1500 mg/m • 1.00 cm
Pxt » 3.16X10"5 kg
1.19 cm
• 6.26x10"4 kg/km2 •
105 cm
km
109 m3
• 2.87x10"' kg/km2 • 15 km2
(5.0 cm + 0.00 cm)
Pqt « 1.02X10"8 kg
The total amount of the chemical in the dissolved state, CHEMd
(Equation 9-13), and the adsorbed form, CHEMa (Equation 9-14), can
be determined by first calculating the total amount of the
9-42
-------
contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the distribution factors:
TOTCHEM = 3.16X10"5 kg + 1.02xlO"8 kg + 2.51xlO"6 kg
-5
TOTCHEM = 3.41X10"3 kg
CHEMd = 3.41X10"5 kg • 9.50X10"4 •
-2
106 mg
kg
CHEMd = 3.24x10 * mg
-5
CHEM = 3.41X103 kg • 0.991 •
CHEMa = 34.0 mg
10° mg
kg
The final concentrations for the short-term Tier 2 and
Tier 3 scenario can now be determined. CHEMW, the dissolved
.concentration of the contaminant in the water body in question is
obtained from the dissolved mass and the appropriate dilution
volume (Equation 9-10):
m
103 L
3.24xlO"2 mg
CHEMH =
8.63X104 m3
CHEMH = 3.75X10"10 mg/L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
9-43
-------
CHEM81
34.0 mg
7.56X105 kg
CHEMs1 = 4.50X10"5 mg/kg
The mass of adsorbed contaminant per volume of water, CHEMs2
(Equation 9-12):
CHEMs2
34.0 mg
m
8.63X10* m3 103 L
-7
CHEMs2 - 3.95x10"' mg/L
9.5.1.3. Scenario C — Tier 1 values for the receiving water
concentration for Scenario C can be calculated directly from the
input variables in Tables 9-4 and 9-5. The concentration of
benzo(a)pyrene in the receiving water for long-term loading in Tier
1, CiLg1 (mg/L) , can be calculated using Equation 9-1 and converting
to the proper units:
Ci
Lg1
2.42X10"2 kg/km3-yr • (15 km2 + 1 km2)
3.15xl07 m3/yr
106 mg m?
kg
103 L
-5
CiLg1 = 1.23x10 mg/L
The concentration in the receiving water for short-term
loading in Tier 1, Cish1 (mg/L), is obtained by replacing Vfx with
VSX in Equation 9-1:
9-44
-------
Pi ss
*—LSh1
2.42xlO~2 kg/km3-yr • (15 km2 + 1 km2)
8.63x10
106 rag m3
kg
103 L
-3
Cish1 = 4.49x10"* mg/L
For Tier 2 or Tier 3 calculations, variables that are needed
in subsequent equations must first be obtained. These are S
(Equation 9-26), DR (Equation 9-25), Q (Equation 9-28) and qp
(Equation 9-29):
S = 2.54 • [(1000/78)-10] =7.16
DR -
[5.0 cm + 0 cm - (0.2 • 7.16 cm)]'
5.0 cm + 0 cm 4- (0.8 • 7.16 cm)
= 1.19 cm
Q = 15 km2 • 1.19 cm =17.8 km2-cm
15 km • 1.19 cm • 5 cm
hr
km 109 m3
6.0 hr • [5.0 cm - (0.2 • 7.16 cm)] 3. 6xl06s 105 cm km3
qp = 11.6 m/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
,0.56
X, = 2.04X10° • (17.8 • 11.55)0'30 • (0.210) (0.179) (0.5) (1.0)
Xs = 7.56xlOs kg
9-45
-------
Calculations for the long-term scenario can now be initiated.
The rate of sediment loss to the receiving water is computed from
the Universal Soil Loss Equation (Equation 9-3):
400 0.21 ton 907.18 kg acre
• • (0.179) (0.5) (1.0) • -
yr acre
ton 4.047xlO"3km2
Xe - 1.68X106 kg/km-yr
The first order loss rate, k, (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, ku (Equation 9-7),
erosion, k1E (Equation 9-8) and degradation, k1D (Equation 9-9) :
1 cm/yr
11 0.22 mL/cm3 • 1.0 cm •[! + (1500 mg/m3 • 120 m3/kg)/0.22 mL/cm3)]
kn « 5.56xlO"6 yr"1
1.685xl06 kg/km-yr
120 m3/kg • 1500 mg/m3
1E 1500 mg/m3 • 1.0 cm 0.22 mL/cm3 + (120 m3/kg • 1500 mg/m3)
0.112 yr
-1
In2
K
ID
2.49
= 0.278 yr
-1
k, = 5.56xlO"6 yr"1 + 0.112 yr"1 + 0.278 yr"1 = 0:391 yr"1
9-46
-------
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady state concentration to be
reached:
Mm = 2.42X10"2 kg/km2-yr •
Mm = 0.0619 kg/km2
1 - exp(-0.391 yr"1 • 100 yr)
0.391 yr"
The contaminant also enters the water body by direct
deposition (Equation 9-5):
Mw = 2.42xlO"2 kg/km2-yr
Mw = 7.68X10'10 kg/m3
1 km2
3.15xlOr m3/yr
For the long-term loading scenario, the receiving water
concentration can now be calculated using Equation 9-2:
Ci
Lg23
1.68X106 kg/km-yr • 15 1cm2 • 0.0619 kg/km2 105 cm
- -""' " — ' '—•-•'•" " . i..I. .1 I,,. ,
3.15X107 m3/yr • 1500 kg/m3 • 1.00 cm
km
106 m3
+ 7.68X10"10 kg/m3
106 mg
kg
m
103 L
Ci. 23 = 4.08X10"6 mg/L
For the short-term loading scenario, the Tier 2 or 3 model
calculates the receiving water concentration after a storm event.
It takes into account any repeirtitioning of the contaminant upon
dilution. This is initiated by calculating the mean sediment
concentration in the water body, Cs, Equation 9-17:
9-47
-------
7.56xl(T kg
8.63xl04 m3/yr
= 8.77 kg/irr
The fraction that would be present in the dissolved, ad
(Equation 9-15), and adsorbed, aa (Equation 9-16), state can now
be determined:
1 + (120 m3/kg • 8.77 kg/m3)
120 m3/kg • 8.77 kg/m3
*. s ; : r~
1 + (120 nr/kg • 8.77 kg/mj)
= 9.50x10
= 0.999
-4
After a storm event, the amount of contaminant .entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23), and dissolved, Da (Equation 9-24), state:
Aa =
0.220 mL/cm
120 m3/kg • 1500 kg/m3
-) ] • 0.0619 kg/kirf
Aa - 0.0165 kg/knr
Da - [
1 +
120 m3/kg • 1500 kg/m3
0.220 mL/cm
) ] • 0.0619 kg/km*2
Da = 7.57X10"8 kg/km2
The amount of chemical deposited directly onto the water body
is a function of the deposition rate, area of the water body, and
time (Equation 9-20):
9-48
-------
-3
Pft = 0.0242 kg/kmc-yr • 2.74xlO"3 yr • 1.0 km'
-5
Pft = 6.63x10 ' kg
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, Pqt (Equation 9-22), is obtained as follows:
7.56X105 kg
1500 mg/m • 1.00 cm
• 0.0165 kg/km2 •
103 cm
km
109 m3
-4
P t = 8.33x0 H kg
1.19 cm
(5.0 cm + 0.00 cm)
• 7.57X10"8 kg/km2 • 15 km2
-7
Pqt = 2.69x10"' kg
The total amount of the chemical in the dissolved state,
CHEMd (Equation 9-13), and the adsorbed form, CHEMa (Equation 9-
14) , can be determined by first calculating the total amount of
the contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the distribution factors:
TOTCHEM = 8.33x10"* kg + 2.69x10"° kg + 6.636X10"5 kg
TOTCHEM = 8.99X10"4 kg
9-49
-------
CHEMd = 8.99X10'4 kg • 9.50xlO"4
CHEMd = 0.854 mg
CHEMa - 8.99X10"4 kg • 0.991 •
CHEM. = 899 mg
10° mg
kg
106 mg
kg
The final concentrations for the short-term Tier 2 or 3
scenario can now be determined. CHEMW, the dissolved concentration
of the contaminant in the water body in question is obtained from
the dissolved mass and the appropriate dilution volume (Equation
9-10):
CHEM.,
0.854 mg
8.63X104 m3
m
103 L
CHEMH = 9.90xlO'9 mg/L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
899 mg
CHEM.
s1
7.56xlOs kg
-3
CHEMs1 = 1.19x10J mg/kg
The mass of adsorbed contaminant per volume of water, CHEMs2
(Equation 9-12):
9-50
-------
CHEMs2 =
899 mg
8. 63xl04 m3
-5
103 L
CHEMS2 = 1.04x103 mg/L
The final results for the concentration of the contaminant in
the water body for the A, B and C scenarios are summarized in Table
9-6.
9.5.2. Cadmium. The values chosen for cadmium for the input
variables of the Surface Runoff Model are given in Table 9-7. The
concentration of a contaminant in the water is a function of its
deposition rate, and the lifetime of the facility. Annual fallout
(Fa) and facility lifetimes (timef) are given in Table 9-8 for
Scenarios A, B and C. Sample calculations for each scenario
follow.
9.5.2.1. SCENARIO A — Tier 1 values for the receiving water
concentration can be calculated directly from the input variables
in Tables 9-7 and 9-8. The concentration of cadmium in the
receiving water for long-term loading in Tier 1, CiLg1 (mg/L) , can
be calculated using Equation 9-1 and converting to the proper
units:
ci =
^x
8.94X10"3 kg/km3-yr • (15 km2 + 1 km2) 106 mg m3
3.15x107 m3/yr
kg
103 L
-6
CiLg1 = 4.54x10"° mg/L
9-51
-------
Table 9-6
Benzo(a)Pyrene Concentration in Receiving Water for the Three
Different Scenarios
B
Fallout (kg/m -yr)
6.51x10
-5
Facility lifetime (yr)30
(mg/L)
shi
ci
CiL823
CHEMj, (mg/L)
CHEMs1 (mg/kg)
CHEMS2 (mg/L)
3.31X10
-8
1.21x10
-5
1.10x10
-8
2.66x10
-11
3.20x10
-6
2.80x10
-8
9.17X10
60
4.66x10
1.70x10
1.55x10
3.75x10
4.50x10
3.95x10
-4
-7
-4
-7
-10
-5
-7
2.42X10
100
1.23X10
4.49x10
4.08X10
9.90x10
1.19X10
1.04x10
-2
-5
-3
-6
-9
-3
-5
9-52
-------
Table 9-7
Input Variables for Determination of Cadmium
Water Concentrations
Variable
Value
WAL (watershed area) (km2)
WAy (water body area) (km )
Vfx (m3/yr)
BD (bulk soil density) (mg/m )
Z (soil till depth) (cm)
R (yr'1)
K (tons/acre)
LS (unitless)
C (unitless)
P (unitless)
k0 (yr'1)
6 (mL water/ cm soil)
*» (y3r>
Kd (mVkg)
Rt (event rainfall) (cm)
Mt (event snow melt) (cm)
CN (SCS Curve Number)
Tr (storm duration) (hrs)
VSX (storm stream flow volume) (m3)
Time (event) (yr) (m3)
IR (cm/yr)
15.0
1.0
3.15x10'
1500
1.0
400.0
0.21
0.179
0.5
1.0
0.0
0.22
100
0.500
5.0
0.0
78.0
6.0
8.63X104
2.74X10"3
1.0
9-53
-------
TABLE 9-8
Annual Fallout and Facility Life for Cadmium
Scenario Facility Life (timef, yr) Fallout (Fa, kg/km3-yr)
A 30 8.94X10"3
B 60 3.76X10"1
C 100 11.9
9-54
-------
The concentration in the receiving water for short-term
loading in Tier 1, Cish1 (mg/L) , is obtained by replacing Vfx with
VSX in Equation 9-1:
8.94xlO"3 kg/km3-yr • (15 km2 + 1 km2) 106 mg m3
8.63x10* m3/yr
kg
103 L
-3
Cish1 = 1.66x10 mg/L
For Tier 2 and 3 calculations, variables that are needed in
subsequent equations must first be obtained. These are S (Equation
9-26), DR (Equation 9-25), Q (Equation 9-28) and q (Equation 9-
29) :
S = 2.54 • [(1000/78)-10] =7.16
[5.0 cm + 0 cm - (0.2 • 7.16 cm)]'
5.0 cm+ o cm + (0.8 • 7.16 cm)
= 1.19 cm
Q = 15 km2 • 1.19 cm =17.8 km2-cm
15 km • 1.19 cm • 5 cm
hr
6.0 hr • [5.0 cm - (0.2 • 7.16 cm)] 3.6x10° s
km 109 m3
105 cm km3
qp = 11.6 m/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
9-55
-------
Xs - 2.04X106 • (17.8 • 11.55)0'56 • (0.210) (0.179) (0.5) (1.0)
Xs = 7.56X105 kg
Calculations for the long-term scenario can now be initiated.
The rate of sediment loss to the receiving water is computed from
the Universal Soil Loss Equation (Equation 9-3):
400
yr
0.21 ton 907.18 kg acre
• (0.179) (0.5) (1.0) •
acre
ton 4.047X10"3 km2
Xc « 1.68xl06 kg/km-yr
The first order loss rate, k, (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, k1t (Equation 9-7),
erosion, k1E (Equation 9-8) and degradation, k1D (Equation 9-9) :
1 cm/yr
0.22 mL/cm3 -1.0 cm •[! +(1500 mg/m3- 0.500 m3/kg)+ 0.22 mL/cm3) ]
k.. s 1.33xlO"3 yr"1
1.68X106 kg/km-yr
0.500 m3/kg • 1500 mg/m3
1500 mg/m3 • 1.0 cm 0.22 mL/cm3 + (0.500 m3/kg • 1500 mg/m3)
1E
0.112 yr
-1
In2
100
= 6.93xlO"3 yr"1
9-56
-------
k, = 1.33xlO~3 yr"1 + 0.112 yr"1 + 6.93xlO~3 yr"1 = 0.121 yr"
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady ^tate concentration to be
reached:
-1
Mm = 8.94X10"3 kg/km2-yr •
Mm = 7.21X10"2 kg/km2
1 - exp(-0.121 yr ' • 30.0 yr)
0.121 yr
-1
The contaminant also enters the water body by direct
deposition (Equation 9-5):
Mw = 8.94xlO~3 kg/km2-yr
Mw = 2.84X10"10 kg/m3
1 knr
3.15xl07 m3/yr
For long-term loading, the receiving water concentration
now be calculated using Equation 9-2:
can
Ci
1.68X108 kg/km-yr • 15 km2 • 93.7 kg/km2 105 cm
km3
Lg33
3.15X107 m3/yr • 1500 kg/m3 • 1.00 cm
km
106 m3
+ 3.78X10"7 kg/m3 •
106 mg
kg
m
103 L
CiLg23 = 4.14xlO"6 mg/L
For the short-term loading scenario, the Tier 2 and Tier 3
model calculates the receiving water concentration after a storm
9-57
-------
event. It takes into account any repartitioning of the contaminant
upon dilution. This is initiated by calculating the mean sediment
concentration in the water body, Cs, Equation 9-17:
7.56X105 kg
Cs - =8.77 kg/m3
8.63x10* m3/yr
The fraction that would be present in the dissolved, ad
(Equation 9-15), and adsorbed, aa (Equation 9-16), state can now
be determined:
0.186
1 + (0.500 mVkg • 8.77 kg/m3)
0.500 m3/kg • 8.77 kg/m3
aa = - = 0.814
1 + (0.500 m3/kg • 8.77 kg/m3)
After a storm event, the amount of contaminant entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23) , and dissolved, Da (Equation 9-24) , state:
0.220 mL/cm3
Aa » [1 + (1 + - - - )] • 7.21X10"2 kg/km2
0.500 m3/kg • 1500 kg/m3
Aa « 1.09x10"* kg/km2
0.500 m3/kg • 1500 kg/m3
Da = [1 •*• (1 + - )] • 7.21X10"2 kg/km2
0.220 mL/cm3
Da = 2.12xlO"5 kg/km2
9-58
-------
The amount of chemical deposited directly onto the water body
is a function of the annual fallout, area of the water body, and
time (Equation 9-20):
,-3
-3
Pft = 8.94x10° kg/km-yr- 2.74x10° yr • 1.0 km'
-5
Pft = 2.45x10 " kg
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, P t (Equation 9-22), is obtained as follows:
7.56xlOs kg
pxt ~
• l.09xlO"4 kg/km2 •
1500 mg/m • 1.00 cm
103 cm
km
km*
109 m3
-6
Pxt = 5.50x10"° kg
1.19 cm
• 2.12X10"5 kg/km2 • 15 km2
(5.0 cm + 0.00 cm)
Pqt = 7.52X10"5 kg
The total amount of the chemical in the dissolved state,
CHEMd (Equation 9-13), and the adsorbed form, CHEMa (Equation 9-
14) , can be determined by first calculating the total amount of
the contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the distribution factors:
TOTCHEM = 5.50X10"6 kg + 7.52xlO"5 kg + 2.45xlO"5 kg
TOTCHEM = 1.05X10"4 kg
9-59
-------
-4
CHEMd = 1.05x10"* kg • 0.186 •
CHEMd = 19.6 mg
-4
CHEMa = 1.05x10* kg • 0.565 •
CHEMa - 85.7 mg
106 mg
kg
106 mg
kg
The final concentrations for the short-term Tier 2 or 3
scenario can now be determined. CHEMH, the dissolved concentration
of the contaminant in the water body in question is obtained from
the dissolved mass and the appropriate dilution volume (Equation
9-10):
CHEMM
19.6 mg
m
8.63X104 m3
-7
103 L
CHEMW = 2.26x10"' mg/L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
CHEMgl —
85.7 mg
7.56X105 kg
CHEM§1 = 1.13X10"4 mg/kg
The mass of adsorbed contaminant per volume of water, CHEM
S
(Equation 9-12):
9-60
-------
85.7 mg
m
CHEMs2 =
8.63X104 m3
103 L
CHEMs2 = 9.93X10"7 mg/L
9.5.2.2. SCENARIO B — For Scenario B, the lifetime of the
facility (timef) changes to 60 years, and the deposition rate of
cadmium (Fa) is 3.76xlO~1 kg/km2-yr. The calculations of the
cadmium concentration for this scenario follow.
The concentration of cadmium in the receiving water for long-
term loading in Tier 1, CiLg1 (mg/L) , can be calculated using
Equation 9-1 and converting to the proper units:
CiL31 =
3.76X10"1 kg/km3-yr • (15 km2 + 1 km2) 106 mg m3
3.l5xl07 m3/yr
kg
103 L
-4
CiLg1 = 1.91x10"* mg/L
The concentration in the receiving water for short-term
loading in Tier 1, Cish1 (mg/L), is obtained by replacing Vfx with
VSX in Equation 9-1:
cishi =
3.76X10"1 kg/km-yr • (15 knT + 1 knT) 10° mg
8.63xl04 m3/yr
kg
m
103 L
-2
Cish1 = 6.97x10 " mg/L
9-61
-------
For Tier 2 and 3 calculations, variables that are needed in
subsequent equations must first be obtained. These are S (Equation
9-26), DR (Equation 9-25), Q (Equation 9-28) and qp (Equation 9-
29) :
S « 2.54 • [(1000/78)-10] = 7.16
[5.0 cm + 0 cm - (0.2 • 7.16 cm)]'
5.0 cm + 0 cm + (0.8 • 7.16 cm)
= 1.19 cm
Q = 15 km2 • 1.19 cm =17.8 km2-cm
15 knr • 1.19 cm • 5 cm
gp
hr
km 109 m3
6.0 hr -[5.0 cm - (0.2 • 7.16 cm)] 3.6xl06 s 105 cm km3
qp » 11.6 m/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
Xs =* 2.04X106 • (17.8 • 11.55)0'56 • (0.210) (0.179) (0.5) (1.0)
Xs = 7.56X105 kg
j
Calculations for the long-term scenario can now be initiated.
The rate of sediment loss to the receiving water is computed from
the Universal Soil Loss Equation (Equation 9-3):
9-62
-------
400 0.21 ton 907.18 kg
. . (0.179) (0.5) (1.0) • -
acre
yr
acre
ton 4.047xlO"3 km2
= 1.68X106 kg/km-yr
The first order loss rate, k, (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, k1z (Equation 9-7) ,
erosion, k1E (Equation 9-8) and degradation, k10 (Equation 9-9) :
1 cm/yr
0.22 mL/cm3 • 1.0 cm-[l +(1500 mg/m3' 0.500 m3/kg)+ 0.22
kn = 1.33X10"3 yr"1
l.68xl06 kg/km-yr
0.500 m3/kg • 1500 mg/m3
k1E =
1500 mg/m3 • 1.0 cm 0.22 mL/cm3 + (0.500 m3/kg • 1500 mg/m )
k1P = 0.112 yr
-1
In2
= 6.93X10'3 yr"1
100
= 4.50xlO"3 yr"1 + 0.112 yr"1 + 6.93xlO"3 yr'1 = 0.121 yr
-1
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady state concentration to be
reached:
9-63
-------
Mm = 3.76xlO"1 kg/km2-yr •
Mm = 3.11 kg/km2
1 - exp(-0.121 yr
-1
60 yr)
0.121 yr
-i
The contaminant also enters the water body by direct
deposition (Equation 9-5):
Mw = 3.76xlO"1 kg/km2-yr
Mw - 1.19xlO"8 kg/m3
1 km2
3. ISxlO7 m3/yr
For the long-term loading scenario, the receiving water
concentration can now be calculated using Equation 9-2:
Ci
Lg23
1.68X108 kg/km-yr • 15 km2 • 3.11 kg/km2
3.15X107 m3/yr • 1500 kg/m3 • 1.00 cm
105 cm km3
km 106 m3
+ 1.19X10"8 kg/m3
106 mg
kg
m
103 L
CiLa23 = 1.79x10"* mg/L
For the short-term loading scenario, the Tier 2 and Tier 3
model calculates the receiving water concentration after a storm
event. It takes into account any repartitioning of the contaminant
upon dilution. This is initiated by calculating the mean sediment
concentration in the water body, Cs, Equation 9-17:
7.56X105 kg
8.63X104 m3/yr.
= 8.77 kg/m3
9-64
-------
The fraction that would be present in the dissolved, ad
(Equation 9-15) , and adsorbed, ota (Equation 9-16) , state can now
be determined:
1 + (0.500 m3/kg • 8.77 kg/m3)
0.500 m3/kg • 8.77 kg/m3
1 + (0.500 m3/kg • 8.77 kg/m3)
= 0.186
= 0.814
After a storm event, the amount of contaminant entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23), and dissolved, Da (Equation 9-24), state:
Aa = [1 *
0.220 mL/cm
0.500 m3/kg • 1500 kg/m3
) ] • 7.2lxlO"2 kg/km2
Aa = 4.71X10"3 kg/km2
Da = [1 * (1 +
0.500 m3/kg • 1500 kg/m3
0.220 mL/cm3
)) • 7.21xlO"2 kg/km2
Da = 9.13X10"4 kg/km2
The amount of chemical deposited directly onto the water body
is a function of the deposition rate, area of the water body, and
time (Equation 9-20):
Pft = 3.76X10"1 kg/km2-yr • 2.74xlO"3 yr • 1.0 km2
Pft = 1.03X10"3 kg
9-65
-------
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, Pqt (Equation 9-22), is obtained as follows:
7.56X1CT kg
1500 mg/m • 1.00 cm
• 4.71xlO"3 kg/km2 •
103 cm
km
km"
109 m3
-4
Pxt - 2.38x10 H kg
1.19 cm
(5.0 cm + 0.00 cm)
9.13xlO"4 kg/km2 • 15 km2
Pql. » 3.25X10"° kg
The total amount of the chemical in the dissolved state,
CHEMd (Equation 9-13), and the adsorbed form, CHEMa (Equation 9-
14) , can be determined by first calculating the total amount of
the contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the distribution factors:
TOTCHEM - 2.38X10"4 kg + 3.25xlO"3 kg + 1.03xlO"3 kg
-3
TOTCHEM = 4.52x10'° kg
-3
CHEMd = 4.52x10"° kg • 0.186
CHEMd = 839 mg
-3
CHEMfl » 4.52X10° kg • 0.565
CHEMfl = 3680 mg
10° mg
kg
10" mg
kg
9-66
-------
The final concentrations for the short-term Tier 2 and Tier
3 scenario can now be determined. CHEMM, the dissolved
concentration of the contaminant in the water body in question is
obtained from the dissolved mass and the appropriate dilution
volume (Equation 9-10):
CHEMH =
CHEM.
839 mg
m
8.63X104 m3
9.72x10"" mg/L
103 L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
CHEMS1 =
3680 mg
7.56xlOs kg
CHEMs1 = 4.86X10"3 mg/kg
The mass of adsorbed contaminant per volume of water, CHEMs2
(Equation 9-12):
CHEMS2 =
3680 mg
m
8.63X104 m3 103 L
-5
CHEMS2 = 4.2 6x10 * mg/L
9-67
-------
9.5.2.3. SCENARIO C — For Scenario C, the lifetime of the
facility (timef) changes to 100 years, and the deposition rate of
cadmium (Fa) is 11.9 kg/km2-yr. The calculations of the cadmium
concentration for this scenario follow.
The concentration of cadmium in the receiving water for long-
term loading in Tier 1, CiLg1 (mg/L), can be calculated using
Equation 9-1 and converting to the proper units:
Ci,
11.9 kg/km3-yr • (15 km2 + 1 km2)
3.15X107 m3/yr
106 mg m3
kg
103 L
CiLg1 - 6.04X10"3 mg/L
The concentration in the receiving water for short-term
loading in Tier 1 Cish1 (mg/L) , is obtained by replacing Vfx with
VSX in Equation 9-1: ,
Ci
11.9 kg/km3-yr • (15 km2 + l km2) 106 mg m3
Sh1
8.63X104 m3/yr
kg
103 L
Cish1 = 2.21 mg/L ..••... .;,;
' ' • " , >'(
For Tier 2 and 3 calculations, variables that are needed in
subsequent equations must first be obtained. These are S (Equation
9-26), DR (Equation 9-25), Q (Equation 9-28) and qp (Equation 9-
29) :
9-68
-------
S = 2.54 • [(1000/78)-10] =7.16
DR =
[5.0 cm 4- 0 cm - (0.2 • 7.16 cm)]'
5.0 cm + 0 cm + (0.8 • 7.16 cm)
= 1.19 cm
Q = 15 km2 • 1.19 cm = 17.8 km2-cm
15 km2 • 1.19 cm • 5 cm
hr
109 m3
6.0 hr -[5.0 cm - (0.2 • 7.16 cm)] 3.6xl06 s 105 cm km3
qp = 11.6 m3/s
The mass of eroded soil, Xs, is obtained from the Modified
Universal Soil Loss Equation (Equation 9-27):
XS = 2.04 • 106 • (17.8._- 11.55)0'56 • (0.210) (O.179) (0.5) (1.0)
Xs = 7.56X105 kg
Calculations for the long-term scenario can now be initiated.
The rate of sediment loss to the receiving water is computed from
the Universal Soil Loss Equation (Equation 9-3):
400 0.21 ton 907.18 kg acre
. . (0.179) (0.5) (1.0) • -
yr
acre
ton 4.047xlO"3 km2
= 1.68xl06 kg/km-yr
9-69
-------
The first order loss rate, k., (Equation 9-6) can be calculated
by adding the loss rates due to infiltration, k,j (Equation 9-7),
erosion, k1E (Equation 9-8) and degradation, k1D (Equation 9-9) :
1 cm/yr
Mi
0.22 mL/cnr -1.0 cm •[! + (1500 mg/m3- 0.500 m3/kg)-r-0. 22 mL/cm3) ]
k,, = 1.33xlO~3 yr~1
Mi
0.500 m3/kg • 1500 mg/m3
1.68X106 kg/km-yr
1500 mg/ra3 • 1.0 cm 0.22 mL/cm3 + (0.500 m3/kg • 1500 mg/m3)
k.E « 0.112 yr
-1
In2
100
= 6.93X10"3 yr"1
kt - 1.33xlO"3 yr"1 + 0.112 yr"1 + 6.93xlO"3 yr"1 = 0.121 yr"1
The maximum contaminant mass per area of soil, Mm (Equation
9-4) is based on the assumption that the combustor has been in
operation long enough to allow a steady state concentration to be
reached:
-1
Mm = 11.9 kg/kmz-yr
Mm « 98.6 kg/km2
1 - exp(-0.121 yr" • 100 yr)
0.121 yr
-1
9-70
-------
The contaminant also enters the water body by direct
deposition (Equation 9-5):
1 km*
Mw =11.9 kg/km-yr •
Mw = 3.78xlO"7 kg/m3
3.15X107 m3/yr
For long-term loading, the receiving water concentration can
now be calculated using Equation 9-2:
Ci
1.68x10° kg/km-yr • 15 km2 • 98.6 kg/km2 105 cm
knr
Lg23
3.15X107 m3/yr . 1500 kg/m3 • 1.00 cm
km
106 m3
3.78xlO"7 kg/m3
-3
10° mg
kg
m
103 L
CiLg23 = 5.66X10'3 mg/L
For short-term loading, the Tier 2 or 3 model calculates the
receiving water concentration after a storm event. It takes into
account any repartitioning of the contaminant upon dilution. This
is initiated by calculating the mean sediment concentration in the
water body, Cs, Equation 9-17:
7.56X105 kg
8.63xl04 m3/yr
==:8.77 kg/m5
The fraction that would be present in the dissolved, ad
(Equation 9-15), and adsorbed, aa (Equation 9-16), state can now
be determined:
9-71
-------
1 + (0.500 m3/kg • 8.77 kg/m3)
0.500 nr/kg • 8.77 kg/m3
1 + (0.500 m3/kg • 8.77 kg/m3)
= 0.186
0.814
After a storm event, the amount of contaminant entering the
water body will be partitioned between the adsorbed, Aa (Equation
9-23), and dissolved, Da (Equation 9-24), state:
Aa » [1 * (l +
0.220 mL/cm
0.500 m3/kg • 1500 kg/m3
-2
) ] • 7.21x10"* kg/km'
Aa » l.SOxlO"1 kg/km;
Da - [1 -5-
0.500 m3/kg • 1500 kg/m3
0.220 mL/cm
Da « 2.89X10"2 kg/km2
•)] • 7.21X10"2 kg/km2
The amount of chemical deposited directly onto the water body
is a function of the deposition rate, area of the water body, and
time (Equation 9-20):
Pft = 11.9 kg/km2-yr - 2.74xlO"3 yr • 1.0 km2
Pft = 3.26X10"2 kg
The mass of contaminant entering the water body from the storm
runoff in the adsorbed form, Pxt (Equation 9-21), and the dissolved
form, Pqt (Equation 9-22), is obtained as follows:
9-72
-------
Pxt =
7.56xlOs kg
1500 mg/m • 1.00 cm
• l.Spxiq;1 kg/km2
10' cm
1cm
km*
109 m3
-3
Pxt = 7.52x10 J kg
1.19 cm
• 9.13xlO"4 kg/km2 • 15 km2
(5.0 cm + 0.00 cm)
Pqt = 1.03X10"1 kg
The total amount of the chemical in the dissolved state,
CHEMd (Equation 9-13), and the adsorbed form, CHEMa (Equation 9-
14) , can be determined by first calculating the total amount of
the contaminant entering the water body from all sources, TOTCHEM
(Equation 9-19), and multiplying by the" distribution factors;
.-3
TOTCHEM = 7.52x10° kg + 1.03X10"1 kg + 3.26xlO"z kg
TOTCHEM = 1.43X10"1 kg
CHEMd = 1.43X10"1 kg • 0.186 •
CHEMd = 2.66X104 mg
-1
CHEMa = 1.43x10'' kg • 0.565 •
CHEMa = 1.16X105 mg
106 mg
kg
106 mg
kg
9-73
-------
The final concentrations for the short-term Tier 2 or 3
scenario can now be determined. CHEMW, the dissolved concentration
of the contaminant in the water body in question is obtained from
the dissolved mass and the appropriate dilution volume (Equation
9-10)J
CHEMH
CHEM,.
2.66x104 mg
8.63X104 m3
3.08x10"* mg/L
103 L
The mass of adsorbed contaminant per mass of sediment in the
water body, CHEMs1 (Equation 9-11) :
CHEMs1 =
1.16X103 mg
7.56xl05 kg
CHEMs1 = 1.54X10"1 mg/kg
The mass of adsorbed contaminant per volume of water, CHEMs2
(Equation 9-12):
1.16X103
CHEMS2 =
8.63x10* m3
-3
mg/m
103 L
CHEMS2 = 1.35X10"3 mg/L
The final results for the concentration of the contaminant in
the water body for the A, B and C scenarios are summarized in Table
9-9.
9-74
-------
TABLE 9-9
Cadmium Concentration in Receiving Water for the Three
Different Scenarios
Fallout (kg/m3-yr)
Facility lifetime (yr)
CiLg1 (m9/L)
Cishl (mg/L)
CiLg23 (m9/L)
CHEMH (mg/L)
CHEMs1 (mg/kg)
CHEMS2 (mg/L)
A
8.94X10"3
30
4.54X10"6
1.66X10"3
4.14X10"6
2.26X10"7
1.13X10"4
9.93X10"7
B
3.76X10"1
60
1.91x10"*
6.97X10"2
1.79X10"4
9.72X10"6
4.86X10"3
4.26X10"5
C
11.9
100
6.04X10"3
2.21
5.66X10"3
3.08X10"4
1.54X10"1
1.35X10"3
9-75
-------
-------
10. DETERMINING EXPOSURE FROM WATER INGESTION
10.1. INTRODUCTION
Contaminants from combustor emissions that have been
transported into water can be directly ingested when humans drink
water. Contaminant intake from water is proportional to the water
concentration of the contaminant and the water ingestion rate.
10.2. DAILY INTAKE FROM WATER (DI)
This methodology assesses only risk associated with the
increase in exposure due to combustor emissions; the background
concentration of contaminants is not included in the quantitation
of DI. DI is calculated as follows:
DI = We • Cw
(Equation 10-1)
where:
DI
We
Cw
total daily intake from water (mg/kg/day)
water concentration (mg pollutant/L)
consumption rate of water (L/kg BW/day)
To determine if the contaminant adversely affects human health,
the DI is compared with the reference dose for chronic oral
exposure to systemic toxicants, such as for cadmium, or used to
determine excess risk for carcinogens, such as for benzo(a)pyrene.
This will be discussed in Chapter 15.
10-1
-------
10.2.1 Water Concentration (We). The methods for calculating
contaminant concentration in both surface water and collected
precipitation are presented in Chapter 9. The long-term Tier 2/3
water concentrations of benzo(a)pyrene (1.55xlO~7 mg/L) and cadmium
(1.79X10*4 mg/L) are used for the example calculations in Section
10.3. These values represent the total concentration of
benzo(a)pyrene or cadmium in water; concentration is not
partitioned into dissolved and adsorbed contaminant. Thus, these
values may overestimate exposure because contaminants adsorbed to
particles suspended in water would be filtered out during treatment
and would not contribute to exposure.
10.2.2. Water Consumption Rate (Cw). The U.S. EPA has
traditionally used a value of 2 L/day to represent the average
daily water consumption of adults and 1 L/day to represent the
average daily water consumption of children with body weights of
£10 kg. However in a recent document, U.S. EPA (1989a) presents
data that suggest these values may overestimate the actual water
consumption rates. These data, generated in several consumption
surveys, show that a value of 1.4 L/day may be a more accurate
average of water consumption for adults. A value of 2 L/day most
likely represents a 90th percentile value. U.S. EPA suggests that
a value ranging from 0.5 to 0.8 L/day may be a more accurate
ingestion rate for children age 5 to 14. In addition, for children
with body weights of <10 kg, the ingestion rate may be closer to
0.2 L/day.
10-2
-------
The daily water ingestion rates used in the example
calculations are expressed in the units L/kg BW/day because it can
be used for estimating ingestion rates for children over a broad
range of ages. The values selected for Scenarios A, B, and C
represent approximately the 50th, 70th and 90th percentiles,
respectively.
For children, the approximate 90th percentile of 0.1 L/kg/day
is used in Scenario C. This is based on the U.S. EPA assumption
of 1 L/day ingestion for children with body weights of 10 kg. The
approximate 50th percentile of 0.032 L/kg/day is used in Scenario
A. This value, presented in U.S. EPA (1989b) , is based on an
k
average water ingestion for children 2 years old weighing 13.5 kg
(Nelson et al., 1969). For Scenario B, a value of 0.066 L/kg/day
was chosen as a reasonable estimate of the 70th percentile.
For adults, the approximate 90th percentile of 0.029 L/kg/day
is used for Scenario C. This value is based on a water ingestion
of 2 L/day divided by a body weight of 70 kg. The approximate 50th
percentile of 0.020 L/kg/day is used for Scenario A. This value
is based on a water ingestion of 1.4 L/day divided by a body weight
of 70 kg. For Scenario B, a value of 0.024 L/kg/day was chosen as
a reasonable estimate of the 70th percentile.
10.3. EXAMPLE CALCULATIONS
This section illustrates a water ingestion exposure to
benzo(a)pyrene and cadmium represented by Scenario B. As discussed
in Chapter 2, Scenario B is defined as a child who grows up in an
area of moderate deposition and remains in the area for 30 years.
10-3
-------
A lake in the area is the only source of drinking water. The
different exposures using child and adult consumption rates are
shown below. Adjustments to exposure are presented in Chapter 15.
Exposure due to ingestion of contaminated drinking water can be
calculated according to Equation 10^1:
Child's DI = 1..55X10"7 mg/L • 0.066 L/kg/day
= 1.02x10 mg/kg/day
Adult's DI = 1.55X10"7 mg/L • 0.024 L/kg/day
= 3.72x10 mg/kg/day
Cadmium
Child's DI
Adult's DI
•i-4
= 1.79x10"* mg/L • 0.066 L/kg/day
,-5
= 1.18x10 mg/kg/day
.-4
1.79x10"* mg/L • 0.024 L/kg/day
"6
— 4.30xlO" mg/kg/day
i • . : L '"..'" -
Chapter 15 shows how these' values are used to determine if
exposure to combustor emissions poses a human health risk.
10-4
-------
11. DETERMINING EXPOSURE FROM FISH INTAKE
11.1. INTRODUCTION
Once contaminants from combustor emissions are transported to
surface water bodies, they may be incorporated into the aquatic
food chain. Humans may then be exposed to contaminants when they
eat fish raised in contaminated surface water bodies. The
concentration of contaminants in fish is determined by multiplying
the water concentration by a bioconcentration factor, which is the
ratio of contaminant concentration in fish to contaminant
concentration in water. However, for contaminants with low water
solubility that are adsorbed to suspended particles (for example,
dioxin and PCBs), use of the fish-water bioconcentration factor may
underestimate contaminant concentration in fish. Use of fish-
sediment bioconcentration factors have been suggested (Goeden and
Smith, 1989; U.S. EPA, 1988a) to overcome this problem. But
because fish-sediment bioconcentration factors are generally
unavailable in the literature, this methodology will continue to
use fish-water bioconcentration factors to estimate contaminant
levels in fish.
11.2. CALCULATING DAILY INTAKE FROM FISH
Human daily intake of contaminants is proportional to the
water concentration, the bioconcentration factor and the fish
ingestion rate. DI is calculated as follows:
11-1
-------
DI = We • BCF • Cf
(Equation 11-1)
where:
DI
We
BCF
Cf
total daily intake from fish (mg/day)
water concentration (mg/L)
bioconcentration factor (L/g)
fish ingestion rate (g/kg/day)
11.2.1. Water Concentration (We). Methods for determining water
concentration are presented in Chapter 9. The long-term Tier 2/3
concentrations of benzo(a)pyrene (1.55xlO~7 mg/L) and cadmium
(1.79xlO~4 mg/L) for Scenario B are used in the example calculations
in Section 11.3. If fish-sediment bioconcentration factors can be
located for use in a site-specific assessment, a contaminant
concentration in sediment, as calculated in Chapter 9 (CHEMs1) , can
be used in place of a water concentration to estimate contaminant
levels in fish.
11.2.2. Bioconcentration Factor (BCF). Bioconcentration is the
ability of living organisms to accumulate substances to higher than
ambient level concentrations. The degree to which a chemical
accumulates above ambient concentrations in an aquatic organism is
indicated by the magnitude of the BCF. Specifically, BCF is
defined as the ratio of the contaminant concentration in all or
part of an aquatic organism (/^g/g fresh weight) to the contaminant
concentration in the water body where the organism has been exposed
(Mg/L) . The BCF, therefore, has units of (M9/9)/ (M9/L) or L/g.
Defined in this way, the bioconcentration factor considers only the
absorption of a pollutant by an organism from the ambient water and
not consumption of contaminated food or water sources.
11-2
-------
Bioconcentration factors are specific for the compound and
the species absorbing the compound. The compounds with the
greatest tendency to bioaccumulate are those that are lipophilic
and resistant to biological degradation. Initial transport into
the organism occurs by rapid surface adsorption or partitioning to
the lipoprotein layer of cell membranes. Once in the bloodstream,
subsequent accumulation of the chemical into particular
compartments of the organism is dependent upon the metabolic
capabilities of the organism and the lipid content of the
individual organs. With continuous exposure to a compound, the
rate of excretion eventually equals the rate of uptake.
Bioconcentration factors can be estimated through laboratory
experiments, field studies, correlations with physicochemical
factors such as octanol/water partition coefficients, and models
based upon pollutant biokinetics coupled to fish energetics. In
the development of the ambient water quality criteria (AWQC), the
U.S. EPA used mostly laboratory data in the calculation of BCFs.
Field data are often less reliable than laboratory data because it
cannot usually be shown that constituent concentrations in field
situations have been held constant for a long period of time or
over the range of territory inhabited by the organism. BCFs
calculated from field data also may be greater than those
calculated from laboratory data, which is apparently due to
ingestion of the compound through prey, sediments and water, in
addition to absorption from water.
When laboratory and field data are not available, BCFs can be
estimated by several methods. Correlations between BCFs and
11-3
-------
octanol/water partition coefficients, water solubility and soil
adsorption coefficients have been documented. Veith et al. (1979)
developed the following equation using the correlation between the
BCP and the n-octanol/water partition coefficient (KOH) to estimate
BCFs to within 60% before laboratory testing:
log BCF = 0.85 log Kou - 0.70
(Equation 11-2)
The equation was developed using data from whole-body analyses of
«7.6 percent lipids (U.S. EPA, 1980e). The U.S. EPA adopted
Equation 11-2 for determining BCFs in the exposure sections of the
health effects chapter of the AWQC documents when an appropriate
BCF is not available. In a later study, Veith et al. (1980) used
the results of their own laboratory experiments and data from other
laboratories for a variety of fish species and 84 different organic
chemicals to obtain the following modification of their original
equation:
log BCF = 0.76 log KOH - 0.23 (Equation 11-3)
Equations similar to the ones developed by Veith et al. (1979,
1980) have been developed for more specific chemical classes and
particular aquatic species (Veith et al., 1979; Neeley et al.,
1974). Other investigators (Norstrom et al., 1976) have developed
more elaborate models using pollutant biokinetics and fish
energetics in addition to using octanol/water partition
coefficients to predict BCFs. Values of fish-water BCF and log
11-4
-------
Kow for many compounds of environmental concern are available in
U.S. EPA (1986d) . A BCF of 3. 6xlO'2 L/g for benzo(a)pyrene is used
in the example calculations in Section 11.3. This value is a
weighted average of values for fish (3.0xlO"2 L/g) and molluscs
(7.9X10"2 L/g) (U.S. EPA, I980b), weighted according to relative
consumption of fish (87.8%) and molluscs (12.2%). Thus, the BCF
for benzo(a)pyrene would be higher for individuals whose exposure
is through a higher percentage of molluscs or other invertebrates.
A weighted average BCF for cadmium was calculated to be 6.4xlO~2 L/g
(U.S. EPA, 1980C).
Since bioconcentration for lipophilic compounds depends on
lipid content of the fish, it is important to adjust measured or
estimated BCF values for these compounds to reflect the lipid
content of seafood in the U.S. diet. The U.S. EPA determined in
1980 that the average lipid content of freshwater and estuarine
species, weighted by average daily consumption, was 3.0% (U.S. EPA,
1980e). Since fresh and estuarine waters would be those affected
by runoff from areas of MWC deposition, a lipid content of 3%
should be assumed. The adjustment is made as follows:
LC,
BCF. = BCF,,
a U
LC_
(Equation 11-4)
where:
BCFa =
BCFU =
LCd =
LC,. =
adjusted BCF (L/kg)
unadjusted BCF (L/kg)
lipid content of dietary seafood (kg/kg)
lipid content of experimental organism (kg/kg)
11-5
-------
11.2.3. Pish Ingestion Rate (Cf). Several recent publications
have provided estimates of daily fish intake averaged over the
entire population, including fish-eaters and non fish-eaters. The
most recent fish consumption document from the U.S. Department of
Commerce (1985) reports total per capita fish and shellfish
consumption ranging from 12.8 pounds/year in 1980 to 13.6
pounds/year in 1984. The latter value is equivalent to a daily
intake of 16.9 g of fish (all kinds).
U.S. EPA (1980d) presents data showing means and 95th
percentiles of monthly fish consumption for fish consumers in the
United States. For adults, the mean consumption rate was 14.3
g/day and the 95th percentile consumption was 41.7 g/day. For
children age 0-9 years (16.46 kg BW) , the mean consumption rate was
6.2 g/day and the 95th percentile consumption rate was 16.5 g/day.
This survey does not distinguish between commercially and
recreationally caught fish.
Runoff containing contaminants from combustor emissions could
affect freshwater and estuarine species, but would not affect the
marine species that constitute the greater portion of seafood in
the U.S. diet. To estimate average daily consumption of freshwater
fish, the U.S. EPA (1980d) eliminated all fish species from the
survey that were not taken from fresh or estuarine waters (Stephan,
1980). This reduced per capita consumption from 14.3 to 6.5 g/day.
Therefore, it seems reasonable to assume that in most instances
freshwater and estuarine species will constitute 50% of total
consumption.
11-6
-------
There was a great disparity in fish consumption rates from the
national average, depending on region, age, race and religion. The
New England and Middle Atlantic regions had the highest fish
consumption rates with means of 16.3 and 16.2 g/day, respectively
and 95th percentile values of 46.5 and 47.8 g/day, respectively.
Based on age, the highest consumption rates were reported for
people 60-69 years, who had a mean consumption of 21.7 g/day and
95th percentile consumption of 55.4 g/day. SRI reported fish
intake by the Black and Jewish populations to be double the average
value (U.S. EPA, I980d). The highest consumption rate, based on
race, was for Orientals, who had a mean consumption of 21.0 g/day
and 95th percentile consumption of 67.3 g/day.
The data discussed to this point have included both
commercially caught fish and data from non fish-eaters.
Commercially caught fish are usually distributed widely so that
prediction of consumption from a particular source is difficult.
Including data from non fish-eaters underestimates the actual
consumption rates for groups eating significant amounts of fish,
such as recreational fishermen. Thus, the data discussed to this
point may not be accurate for assessing exposure of recreational
fishermen from consumption of contaminated fish from one location.
U.S. EPA (1989a) presents data from two surveys that
specifically studied consumption rates of recreational fishermen.
Although both studies were conducted on the west coast, U.S. EPA
(1989a) concludes that they are representative of "consumption
rates of recreational fishermen where there is a large water body
present and widespread contamination is evident". The 50th, 70th,
11-7
-------
and 90th percentile consumption rates reported for recreational
fishermen are 30 g/day, 80 g/day, and 140 g/day respectively. U.S.
EPA (1989a) notes that these values are not recommended "for small
water bodies or for areas of localized contamination in large water
bodies". For assessments in these areas, U.S. EPA (1989a)
recommends that assessors interview local recreational fishermen
to derive site-specific data.
The daily fish ingestion rates used in the example
calculations are expressed in the units g/kg BW/day because this
approach can be used for estimating children's ingestion rates over
a broad range of ages. For calculating children's exposure due to
fish ingestion, this document uses the average or 95th percentile
rates for children age 0-9 years, based on total U.S. population
(U.S. EPA, 1980d) . The average and 95th percentile rates were
adjusted by 50% to account for the fraction of total consumption
that can be attributed to fresh water or estuarine sources. The
adjusted rates were divided by the average body weight for children
age 0-9 years, 16.46 kg (Nelson et al., 1969) . The resulting value
of 0.188 g/kg BW/day could be used for Scenario A and 0.501 g/kg
BW/day could be used for Scenario C. An intermediate value of
0.345 g/kg BW/day was chosen as a reasonable estimate of the 70th
percentile and is used in the examples of Scenario B.
For calculating adults' exposure due to fish ingestion, this
document uses the U.S. population average consumption rates (14.3
g/day) for Scenario A, and the U.S. population 95th percentile
consumption rates (41.7 g/day) for Scenario B. The 75th percentile
consumption rate of recreational fishermen (80 g/day) is used for
11-8
-------
Scenario C. All values are adjusted by 50% to account for the
fraction of total consumption that can be attributed to fresh water
or estuarine sources, then divided by 70 kg BW. The resulting
values for Scenario A, B, and C are 0.102, 0.298, and 0.571 g/kg
BW/day, respectively.
11.3. EXAMPLE CALCULATIONS
This section illustrates exposure to benzo(a)pyrene and
cadmium via fish ingestion for Scenario B. As discussed in Chapter
2, Scenario B is represented by a child who grows up in an area of
moderate deposition of emissions and remains in the area for 30
years. A lake in the area is used for recreational fishing. The
different exposures using child and adult consumption rates are
shown below. Adjustments to exposure are presented in Chapter 15.
The values of input variables chosen for Scenario B are presented
in Table 11-1.
11.3.1. Benzo(a)Pyrene. Exposure due to fish ingestion for a
child can be calculated according to Equation 11-1:
DI = 1.55X10"7 mg/L • 3.6xlO"2 L/g • 0.345 g/kg/day
= 1.93xlO"9 mg/kg/day
11-9
-------
TABLE 11-1
Values for Input Variables Used in Scenario B Example
Input Variable B(a)P
Cadmium
We (mg/L)
BCF (L/g)
Cf (g/kg/day)
1.55X10"7
3.6X10"2
0.345 - child
0.298 - adult
1.79X10"4
6.4X10"2
0.345 - child
0.298 - adult
11-10
-------
Exposure due to fish ingestion for an adult can also be
calculated according to Equation 11-1:
DI = 1.55X10"7 mg/L • 3.6xlO"2 L/g • 0.298 g/kg/day
= 1.66xlO"9 mg/kg/day
These values are used in Chapter 15 to determine excess risk
for exposure to benzo(a)pyrene in combustor emissions.
11.3.2. Cadmium. Exposure due to fish ingestion for a child can
be calculated according to Equation 11-1:
DI = 1.79xlO"4 mg/L • 6.4xlO"2 L/g • 0.345 g/kg/day
= 3.95X10"6 mg/kg/day
Exposure due to fish ingestion for an adult can also be
calculated according to Equation 11-1:
DI = 1.79X10"4 mg/L • 6.4xlO"2 L/g • 0.298 g/kg/day
= 3.41x10"° mg/kg/day
These values are compared with the RfD0 according to the
approach described in Chapter 15.
11-11
-------
-------
12. DETERMINING EXPOSURE FROM
DERMAL ABSORPTION FROM WATER
12.1. INTRODUCTION
Once contaminants in combustor emissions reach water sources,
as described in Chapter 9, humans may be exposed by absorbing them
through the skin while swimming or bathing. Dermal absorption
could be a significant source of exposure from contaminated water
compared with ingestion. Brown et al. (1984) reported that dermal
absorption of volatile organic contaminants in drinking water
accounts for an average of 64% of the dose incurred. Many of the
factors that influence dermal absorption from soil also influence
dermal absorption from water and the uncertainties associated with
the soil pathway (see Chapter 7) can be assumed to apply to the
water pathway as well.
12.2. DAILY DERMAL INTAKE FROM WATER
The daily dermal intake from water represents the increase
above background in daily human dermal intake due to combustor
emissions. As with absorption from soil, exposure due to dermal
absorption from water is likely to be influenced by the contaminant
concentration in water and the extent of contact. Chapter 9
discusses the procedures for determination of pollutant water
concentration. The extent of contact is affected by contact time,
12-1
-------
frequency and surface area. In addition, the fraction of the
contaminant absorbed from water will influence the overall dermal
exposure. These factors are presented below.
12.2.1. Contact Time and Frequency. As with dermal exposure to
soil, contact time is defined as the amount of time each day skin
is in contact with contaminated water; frequency is defined as the
number of days per year that skin is in contact with contaminated
water. Skin comes into contact with contaminated water when
individuals swim in contaminated surface water sources or bathe in
contaminated water that originates from either a surface water
source or collected precipitation.
U.S. EPA (1988b) reports data quantifying the contact time
and frequency of swimming, based on information from the Bureau of
Outdoor Recreation. The average contact time for swimming is 2.6
hours/day, and the average frequency is 7 days/yr. However, these
averages are likely to vary significantly depending on local
climate and location. Thus, risk assessors should consult the
local recreation department for site-specific data when conducting
an assessment.
U.S. EPA (1989a) reports data quantifying the contact time and
frequency of bathing. The 50th percentile bathing time is
approximately 0.12 hours/day (7 minutes); the 75th percentile
bathing time is approximately 0.15 hours/day (9 minutes); the 90th
percentile bathing time is approximately 0.2 hours/day (12
minutes). Ninety percent of Americans report bathing once a day,
and 5% report bathing more than once a day.
12-2
-------
12.2.2. Surface Area. Dermal intake of a contaminant is
proportional to the exposed surface area. For exposure via
swimming and bathing, the exposed surface area is assumed to be
the total body surface area. U.S. EPA (1989a) presents total body
surface area data for adult males, adult females and children.
A quantitative approach to estimating dermal exposure to
contaminants in water is precluded by the paucity of data available
concerning dermal absorption from water.
12-3
-------
-------
13. HAZARD IDENTIFICATION
13.1. INTRODUCTION
The purpose of hazard identification, the first step in the
risk assessment process, is to determine the type of toxicity that
is produced by a chemical agent (e.g., mutagenicity, developmental
toxicity, carcinogenicity). Hazard identification is the
qualitative risk assessment in which all available experimental
animal and human data are reviewed to determine if the agent is
likely to cause a particular type of toxicity.
The hazard identification step in this document distinguishes
two types of toxic responses, i.e., whether or not a chemical agent
has the potential to increase the incidence of cancer in exposed
humans, because the U.S. EPA«s approaches to assessing risks
associated with carcinogens and noncarcinogens are different.
Pollutants emitted from combustors may include carcinogens and
noncarcinogens.* The purpose of this chapter is to describe the
general process of hazard identification and to show how the
process can be used in evaluating the hazard of toxic pollutants
in combustor emissions.
In the analysis of data regarding the potential human
carcinogenicity of chemical agents, the U.S. EPA uses the approach
described in its Guidelines for Carcinogen Risk Assessment (U.S.
*The decision as to whether to treat a chemical as a
threshold- or nonthreshold-acting (carcinogenic) agent depends on
the weight of evidence that it may be carcinogenic to humans (refer
to section 13.3).
13-1
-------
EPA, 1986f). This follows the general format of the National
Academy of Sciences (NRG, 1983) in its description of the risk
assessment process.
For carcinogenic chemicals, the Agency considers the risk of
cancer to be linearly related to dose (except at high dose levels)
(U.S. EPA, 1986f). Therefore, while risk diminishes with dose, it
does not become zero until dose becomes zero. The decision whether
to treat a chemical as a threshold- (systemic toxicant) or
nonthreshold-acting (carcinogenic) agent depends on the weight of
evidence that it may be carcinogenic to humans. The agent's
potential for human carcinogenicity is inferred from the available
information relevant to the potential carcinogenicity of the
chemical and from judgments as to the quality of the available
studies. Chemicals that are not considered to be carcinogenic are
assumed to produce systemic effects with a dose threshold.
13.2. ELEMENTS OP HAZARD IDENTIFICATION
All relevant data providing both principal and supporting
evidence of the chemicals' toxic effects (e.g., carcinogenic,
systemic toxicity) should be included in the hazard identification
process. The studies providing principal evidence (particularly
for the purposes of this methodology) are those that contribute
most significantly to a qualitative assessment of whether or not
a particular chemical is potentially carcinogenic in humans.
Principal evidence is obtained from: (1) epidemiologic studies or
human studies that elaborate the association between cancer
incidence and exposure to the agent, and (2) long-term animal
13-2
-------
studies conducted under controlled laboratory conditions.
Supporting evidence is derived from short-term tests for
genotoxicity, toxicologic effects other than cancer, metabolic and
pharmacokinetic properties, structure-activity relationships, and
physical/chemical properties of the agent (U.S. EPA, 1986f).
In the evaluation of noncancer health effects, the U.S. EPA
assumes a threshold exists. Hazard identification for noncancer
effects involves the consideration of the toxic endpoints from all
available studies with primary attention given to identification
of a "critical effect". Principal studies, both human studies and
animal studies, contribute to the qualitative assessment of whether
an agent is potentially a systemic toxicant in humans (Barnes and
Dourson, 1988). These studies are evaluated for quality and
adequacy using many of the same criteria as for carcinogenicity
studies.
13.2.1. Human Studies/ Epidemiclogic Studies. Data from human
studies should be used to qualitatively establish the presence of
an adverse effect in exposed human populations when such
information is available. The strength of the human evidence for
both carcinogenicity and systemic toxicity depends, among other
things, on the type of analysis and on the magnitude of the
response. The weight of evidence increases with the number of
adequate studies that show comparable results on populations
exposed to the same agent under different conditions (U.S. EPA,
1986f; Barnes and Dourson, 1988).
13-3
-------
13.2.2. Animal Studies. For most chemicals, there is a lack of
appropriate data on effects in humans. In such cases, the
principal studies are drawn from experiments conducted on nonhuman
mammals (U.S. EPA, 1986f; Barnes and Dourson, 1988). These long-
term studies include those for chronic systemic effects, as well
as carcinogenicity.
Information from all of these studies is useful in the hazard
identification phase of risk assessment. Animal studies are
conducted using a variety of exposure durations (e.g., acute,
subchronic, and chronic), schedules (e.g., single, intermittent,
or continuous dosing), and routes (e.g., oral, dermal, inhalation),
and may be used to further support an identified toxic/carcinogenic
response. For example, overt neurologic problems identified in
high-dose acute studies tend to reinforce the observation of subtle
neurologic changes seen in low-dose chronic studies. Special
attention is given to studies involving low-dose, chronic
exposures, since such exposures can elicit effects absent in higher
dose, shorter exposures, through mechanisms such as accumulation
of toxicants in the organisms (Barnes and Doiirson, 1988) .
13.2.3. Supportive Evidence. These studies provide supportive,
rather than definitive, information and can include data from a
wide variety of sources. For example, in vitro studies can provide
insights into the chemical's potential for biological activity; in
certain circumstances, consideration of structurally-related
compounds can provide clues to the test chemical's possible
toxicity. Similarly, relevant metabolic and pharmacokinetic
13-4
-------
studies can provide insights into the mechanism of action of a
particular compound (U.S. EPA, 1986f; Barnes and Dourson, 1988).
13.2.3.1. SHORT-TERM AND OTHER TOXICTTY STUDIES — Short-
term tests for mutagenicity, including tests for point mutations,
numeric and structural chromosome aberrations, DNA damage/repair,
and in vitro transformation provide supportive evidence of
carcinogenicity and may give information on potential carcinogenic
mechanisms. Short-term in vivo and in vitro tests that can give
an indication of initiation and promotion activity may also provide
supportive evidence for carcinogenicity. Lack of positive results
in short-term tests for genetic toxicity does not provide a basis
for discounting positive results in long-term animal studies (U.S.
EPA, 1986f).
Toxicologic effects other than carcinogenicity or
mutagenicity, including systemic, teratogenic, and other
reproductive effects may also be relevant to the evaluation of
carcinogenicity. Prechronic and chronic toxicity evaluations and
dose-response and time-to-response analyses of these reactions may
yield information on target organ effects, pathophysiologic
reactions, and preneoplastic lesions that bear on the evaluation
of carcinogenicity (U.S. EPA, 1986f).
13.2.3.2. PHYSICAL-CHEMICAL PROPERTIES — Chemical specific
parameters relevant to carcinogenesis, including physical state and
physical-chemical properties, and pertinent information on whether
structure-activity relationships support or argue against a
13-5
-------
prediction of potential carcinogenicity also should be considered.
Additionally, relevant metabolic information such as whether the
agent is direct-acting or requires conversion to a reactive
carcinogen (e.g., electrophilic) species, metabolic pathways for
such conversions, macromolecular interactions, and fate (e.g.,
transport, storage, and excretion), as well as species differences,
should be critically evaluated (U.S. EPA, 1986f). By comparing the
metabolism of the chemical exhibiting the toxic effect in the
animal with the metabolism found in humans, it may be possible to
assess the potential for toxicity or to estimate the equitoxic dose
in humans (U.S. EPA, 1986f; Barnes and Dourson, 1988).
13.2.3.3. ROUTES AND PATTERNS OF EXPOSURE — Since the U.S.
EPA often approaches the investigation of a chemical with a route
of exposure in mind (e.g., an oral exposure for a drinking water
contaminant or an inhalation exposure for an air contaminant),
consideration is given to potential differences in absorption or
metabolism resulting from different routes of exposure, and
whenever appropriate data (e.g., comparative metabolism studies)
are available, the impact of these differences on the quantitative
risk assessment are delineated. In most cases, the toxicologic
data base does not include detailed testing on all possible routes
of administration. Generally, the U.S. EPA's position is that the
potential for toxicity manifested via one route of exposure is
relevant to consideration of any other route of exposure, unless
convincing evidence exists to the contrary (U.S. EPA, 1986f; Barnes
13-6
-------
and Dourson, 1988). This may be an important consideration in
this methodology, since exposure may occur by several routes.
13.3. WEIGHT OF EVIDENCE
As the culmination of the hazard identification step, a
discussion of the weight of evidence summarizes the information
gleaned from the principal and supportive studies. Emphasis is
given to examining the results from different studies to determine
the extent to which a consistent, plausible picture of toxicity
emerges. For example, the weight of evidence involves considera-
tion of the quality and adequacy of the human studies; valid and
adequate studies strengthen the evidence for carcinogenicity.
Evaluation of individual studies in humans and animals
requires the consideration of several factors associated with a
study's hypothesis, design, execution, and interpretation. An
ideal study addresses a clearly delineated hypothesis, follows a
carefully prescribed protocol, and includes sufficient subsequent
analysis to support its conclusions convincingly.
Among the criteria for the adequacy of epidemiologic studies
are: proper selection and characterization of exposed and control
groups; adequate duration and quality of follow-up; proper
identification and characterization of confounding factors and
bias; consideration of latency affects; valid ascertainment of the
causes of morbidity and death; and the ability to detect specific
effects (U.S. EPA, 1986f). In evaluating these studies
consideration is also given to characterization of the compound(s)
under study, similarities and differences between the test
13-7
-------
individuals in the study groups, the number of study groups and the
types of observations and methods of statistical analysis.
Animal studies add to the weight of evidence that the chemical
poses a hazard to humans. The weight of evidence increases with:
an increase in number of tissue sites affected by the agent;
similar results in replicated animal studies by different
investigators; similar effects across sex, strain, species, number
of experiments and doses and route of exposure; clear evidence of
a dose-response relationship and high level of statistical
significance of the increased response in treated compared with
control groups; a plausible relationship between data on
metabolism, postulated mechanism-of-action, and the toxic endpoint;
similar toxicity exhibited by structurally-related compounds; and
some link between the chemical and evidence of the critical
endpoint in humans (U.S. EPA, 1986f; Barnes and Dourson, 1988).
Consideration is also given to the type of test species, the
spacing and choice of the dose levels tested, the age and sex of
the test animals and the route and duration of exposure. For
carcinogenicity studies, dose-related shortening of the time-to
tumor occurrence or time-to-death with tumor and the occurrence of
a dose-related increase in the proportion of tumors that are
malignant also factor into the weight of evidence.
The scheme used to categorize the weight of evidence for the
agent in question as a human carcinogen has three major steps. In
the first step, the evidence is characterized for human studies and
for animal studies. Secondly, the human and animal evidence are
combined into a provisional overall classification. In the third
13-8
-------
stage, the provisional classification is adjusted upwards or
downwards, based on analysis of the supporting evidence. The
result is that each chemical is placed into one of the following
five categories (U.S. EPA, 1986f; U.S. EPA, 1990):
Group A: Human Carcinogen. Sufficient evidence exists
from epidemiology studies to support a causal association
between exposure to the chemical and human cancer.
Group E: Probable Human Carcinogen. Sufficient evidence
of carcinogenicity in animals with limited (Group Bl) or
inadequate (Group B2) evidence in humans.
Group C: Possible Human Carcinogen. Limited evidence
of carcinogenicity in animals in the absence of human
data. .
Group D: Not Classifiable as to Human Carcinogenicity.
Inadequate human and animal evidence of carcinogenicity
or for which no data are available.
Group E: Evidence of Noncarcinogenicity for Humans. No
evidence of carcinogenicity in at least two adequate
animal tests in different species or in both adequate
epidemiologic and animal studies.
13^9
-------
Scientific judgment is necessary when deciding which
classifications constitute sufficient evidence for basing a
quantitative risk assessment on a presumption of carcinogenicity.
The quantitative estimates of carcinogenic risk of chemicals in
categories A and B at low doses have usually been made with a
linear no-threshold dose-response model, whereas those in
categories D and E have usually been assessed according to
threshold effects. Chemicals classified as C have received varying
treatment. The use of the weight of evidence classification,
without noting the explanatory material for a specific chemical,
may lead to a flawed conclusion since some of the classifications
are exposure route dependent.
In evaluating the results from both carcinogenicity and
systemic toxicity studies, consideration is given to many other
factors, including chemical characterization of the compound(s)
under study, the type of test species, similarities and differences
between the test individuals in the study groups, the number of
study groups, the spacing and choice of dose levels tested, the
types of observations and methods of analysis, the nature of
pathologic changes, the alteration in metabolic responses, the sex
and age of the test animals, and the route and duration of exposure
(U.S. EPA, 1986f; Barnes and Dourson, 1988).
13.3.1. EPA Classification of Benzo(a)Pyrene and Cadmium.
Benzo(a)pyrene and cadmium have been used throughout this document
to illustrate the application of the exposure methodology and will
be used as examples for the estimation of human health risk.
13-10
-------
Human data specifically linking benzo(a)pyrene to a
carcinogenic effect are lacking, even though lung cancer has been
shown to be induced in humans by various mixtures of polycyclic
aromatic hydrocarbons known to contain benzo(a)pyrene (U.S. EPA,
1990). It is not possible to conclude from this information that
.benzo(a)pyrene is the only responsible agent. In this case, the
human data are considered inadequate for classification of
benzo(a)pyrene. However, multiple animal studies in rodent and
nonrodent species demonstrating benzo(a)pyrene to be carcinogenic
following administration by oral, intratracheal, inhalation and
dermal routes lend support to the evidence for carcinogenicity.
In addition, benzo(a)pyrene has produced positive results in
several in vitro bacterial and mammalian genetic toxicity assays,
as well as a number of in vivo tests for DNA damage. It is
metabolized to reactive electrophiles capable of binding to DNA.
The principal evidence from human and animal studies, along with
the supporting evidence resulted in a weight of evidence
classification for benzo(a)pyrene of B2, a probable human
carcinogen (U.S. EPA, 1990). For illustrative purposes in this
document, benzo (a) pyrene will be used as the example of
determination of exposure and risk for a carcinogen emitted by
combustors.
Cadmium is classified as a probable human carcinogen (Bl)
(U.S. EPA, 1985c). Insufficient information exists to estimate an
oral potency for cadmium (U.S. EPA, 1990)„ However, an oral RfD
is available for cadmium (U.S. EPA, 1990) (see also Chapter 14).
Because carcinogens may also produce systemic toxicity and most of
13-11
-------
the exposures in this document are by oral routes, this document
will use cadmium as the example of determination of exposure and
risk for a noncarcinogen emitted by combustors.
13.4. SUMMARY AND CONCLUSION
The purpose of the hazard identification step of the risk
assessment process is to qualitatively determine the type of toxic
response that is produced by a chemical agent. Since the risk
assessment process is different for carcinogens and noncarcinogens,
for purposes of this document, hazard identification involves
determining whether or not the chemical has the potential to
increase the incidence of cancer in exposed humans. Principal
evidence is obtained from human or epidemiolpgic studies and long-
term, controlled animal studies. Supporting evidence comes from
short-term genotoxicity and other toxicity tests, studies on
metabolism and other physical-chemical properties, and structure-
activity relationships. All of the evidence is evaluated and the
chemical is placed in one of the five weight of evidence categories
for carcinogenicity according to the primary evidence, with any
adjustment to that classification resulting from the supporting
evidence.
If the chemical is classified as a carcinogen, then it will
be necessary to calculate the excess risk. If a chemical is
classified as a noncarcinogen, or threshold-acting agent, then the
assessor should determine risk by comparing exposure with a
reference dose (RfD) as discussed in Chapter 15.
13-12
-------
14. DOSE-RESPONSE ASSESSMENT
14.1. INTRODUCTION
The dose-response assessment is the second step of the risk
assessment process (NRC, 1983) and is an attempt to define the
relationship between the dose of a chemical and the resulting toxic
and/or carcinogenic effects. This assessment results in a
quantitative measure of a concentration or dose of a chemical above
which humans are at risk for systemic toxicity, or an estimate of
carcinogenic potency or risk when humans are exposed for a
lifetime. In the case of systemic toxicity, this dose is termed
the reference dose, RfD; in the case of carcinogenicity, the
estimate of carcinogenic potency is termed the slope factor. Many
RfD and cancer slope factor values are currently available on the
U.S. EPA Integrated Risk Information System (IRIS) (U.S. EPA,
1990). Chemicals that produce systemic toxicity are assumed to
have a threshold dose below which no adverse, noncarcinogenic
health effects occur, and chemicals that produce carcinogenic
effects arei assumed to act without a threshold. Therefore, the
respective dose-response assessment processes are different and
will be discussed separately. This chapter presents information
necessary for understanding the derivation process.
14.2. DOSE-RESPONSE ASSESSMENT FOR SYSTEMIC TOXICITY
Empiric observation generally reveals that as the dosage of
a toxicant increases, the toxic response also increases. In
general, this increase occurs in severity, intensity and frequency
14-1
-------
within the population. Such dose-response relationships are well-
founded in the theory and practice of toxicology and pharmacology
(Barnes and Dourson, 1988).
14.2.1. Selection of the Critical Data. Before an RfD can be
calculated for a given chemical, the pertinent information must be
selected from the available literature. Because of the
difficulties inherent in evaluating a dose-response relationship
for a given chemical, scientific judgment must be exercised during
the selection process. There must be a decision as to the critical
endpoint to measure as the "response". There must be a
determination of the correct measure of "dose". Other factors that
must be considered include interspecies variation, appropriate
units for dose, and the concept of administered dose versus
absorbed dose versus target organ dose (Barnes and Dourson, 1988).
The NOAEL (no-observed-adverse-effect level) is defined as the
highest experimental dose of a chemical at which there is no
statistically or biologically significant increase in frequency or
severity of an adverse effect in individuals in an exposed group
when compared with individuals in an appropriate control group.
Sometimes a chemical may induce more than one adverse effect
(endpoint) in the same test animal. In this case, the critical
effect selected for the dose-response assessment is the effect
resulting from the lowest NOAEL. The NOAEL is the most important
information obtained from the study of the dose-response
14-2
-------
relationship and is the measurement on which the quantitative
assessment of the human risk from systemic toxicants is based
(Barnes and Dourson, 1988).
Risk assessments based on human data have the advantage of
avoiding the problems inherent in interspecies extrapolation. An
appropriate human study with a well-defined NOAEL for a critical
effect has preference over animal studies as a basis for the RfD
determination. For most chemicals, human cohort investigations of
noncancer endpoints have not been conducted; however, data from
retrospective epidemiologic studies can be used when available.
In some cases, a well-designed and well-conducted epidemiologic
study that shows no association between known exposures and
toxicity can be used to determine an RfD. However, in many
instances, use of epidemiologic studies involves extrapolation from
relatively high doses (such as those found in occupational
settings) to the low doses found in the environmental situations
to which the general population is more likely to be exposed
(Barnes and Dourson, 1988).
When human data are not available, the most appropriate NOAEL
of the critical effect from a well-conducted study on an animal
species that is known to resemble the human in response to the
particular chemical is chosen. When the most similar species
cannot be determined, the most sensitive species (i.e., the species
showing a toxic effect at the lowest administered dose) is
selected. Use of data from controlled studies of genetically
homogeneous animals requires extrapolation from animals to humans
and from high experimental doses to comparatively low environmental
14-3
-------
exposures, as well as consideration of human heterogeneity and
possible concurrent human exposures to other chemicals (Barnes and
Dourson, 1988).
14.2.2. Reference Dose (RfD) . The RfD is a bench mark dose
operationally derived from the NOAEL by consistent application of
uncertainty factors (UFs) that reflect various types of data sets
used to estimate RfDs (Barnes and Dourson, 1988). It is defined
as an estimate (with uncertainty spanning perhaps an order of
magnitude) of-a daily exposure to the human population (including
sensitive subgroups) that is likely to be without an appreciable
risk of deleterious effects during a lifetime. The RfD is
calculated as follows:
' NOAEL (Equation 14-1)
RfD =
UF x MF
where:
NOAEL = no-observed-adverse-effect level (mg/kg/day)
UF = uncertainty factor (unitless)
MF = modifying factor (unitless)
The RfD is useful as a reference point from which to judge
the potential effects of the chemical at other doses. Usually,
doses less than the RfD are not likely to be associated with
adverse health risks, and are therefore less likely to be of
regulatory concern. As the frequency and/or magnitude of the
exposures exceeding the RfD increase, the probability of adverse
effects in a human population increases. However, it should not
be concluded that all doses below the RfD are "safe" or without
14-4
-------
risk and that all doses greater than the RfD are "unsafe" or pose
a risk to human health (Barnes and Dourson, 1988).
The oral RfD (RfD0) is expressed in units of milligrams per
kilogram of bodyweight per day (mg/kg/day). Methodology for
determination of an RfD for chronic inhalation exposure to a
chemical (RfD,.) is currently being developed (U.S. EPA, 1989c) .
Because this document deals only with indirect exposures, i.e.,
those resulting from deposition of combustor emissions on
environmental media, direct inhalation exposure is not considered.
Indirect inhalation exposure such as inhalation of resuspended
dust, is not likely to be a significant route of exposure (Figure
2-1).
14.2.3. Selection of Uncertainty and Modifying Factors.
Uncertainty factors (UFs) are reductions in the dose rate needed
to account for several areas of scientific uncertainty inherent in
most toxicity data bases. Currently, uncertainty factors of 10
each are applied to extrapolate from animals to humans, to provide
protection for unusually sensitive individuals, to expand from
subchronic to chronic exposure, to estimate a NOAEL from a lowest-
observed-adverse-effect level (LOAEL) and to reflect deficiencies
in the data base.
There may be additional uncertainties in the estimation of an
RfD such as scientific uncertainties in the key study, chemical-
specific issues, and/or deficiencies in the overall data base. In
these instances a modifying factor (MF) greater than zero but <10
14-5
-------
is applied to account for these considerations. The default value
for the MF is 1. The maximum total uncertainty factor applied in
the derivation of an RfD is 10,000.
14.2.4. Confidence in the RfD. When adequate data from acceptable
human studies are available, their use is preferred as the basis
for the RfD determination because the resulting value is more
certain, and the confidence in the RfD will be higher. In the
absence of human data, a complete data base consisting of chronic
and developmental toxicity studies in two mammalian species and a
2-generation reproductive study in one mammalian species is
preferred. The existence of such a complete data base would
indicate that there is little probability that additional data
would affect the RfD and, therefore, confidence in the resultant
RfD will be high. In some cases, however, general toxicity
information may indicate the need for special studies such as
neurotoxicity, immunotoxicity, or reproductive toxicity, making
the data base incomplete until these studies are performed. The
minimum data base needed for an RfD estimation is a single, well-
conducted, subchronic animal study; however, for such a data base,
the probability that additional data might change the estimate is
high so there is low confidence in the RfD.
14-6
-------
14.2.5. Calculation of RfD for Cadmium. Throughout this document,
sample calculations have been presented using data for cadmium as
an example of a systemic toxicant emitted by a MWC. Thus, for
explanatory purposes, it is appropriate to present the calculation
of an RfD for this chemical.
A U.S. EPA (1985b) study showed that a concentration of 0.2
mg Cd/g wet human renal cortex was the highest renal level not
associated with significant proteinuria. A toxicokinetic model
exists with which the level of chronic human oral exposure (NOAEL)
resulting in 0.2 mg Cd/g wet human renal cortex can be determined;
the model assumes that 0.01% of the cadmium body burden is
eliminated.per day (U.S. EPA, 1985b). Assuming 2.5% absorption of
cadmium from food or 5% from water, the model predicts that the
NOAELs for chronic cadmium exposure are 0.01 and 6.005 mg Cd/kg/day
from food and from water, respectively (i.e., levels that would
result in 0.2 mg Cd/g wet weight human renal cortex). A UF of 10
was selected to account for intrahuman variability to the toxicity
of cadmium in the absence of specific data on sensitive
individuals. The default MF (1) was used. Thus, the U.S. EPA RfD
Work Group has established the oral RfD for cadmium to be IxlO"3
ing/kg/day in food and 5xlO"4 mg/kg/day in drinking water. Because
the choice of the NOAEL reflects data obtained from many studies
on the toxicity of cadmium in both humans and animals and permits
calculation of cadmium pharmacokinetic parameters such as
absorption, distribution, metabolism and elimination, confidence
in the data base is high, resulting in high confidence in both RfD
values (U.S. EPA, 1990).
14-7
-------
14.3. DOSE-RESPONSE ASSESSMENT FOR CARCINOGENS
The purpose of the dose-response assessment step in carcinogen
risk assessment is to define the relationship between the dose of
a chemical and the likelihood of a carcinogenic effect assuming the
chemical to be a human carcinogen. The goal of the assessment is
to estimate the potency of the chemical and to determine the "slope
factor" that is used in conjunction with the exposure estimate to
yield a numeric estimate of risk. Dose-response assessment usually
involves extrapolation from the high doses administered to
experimental animals to the much lower exposure levels expected
from human contact with the agent in the environment; it also
includes considerations of the validity of these extrapolations.
14.3.1. Selection of Data Sets. The dose-response assessment
process requires scientific judgment in the choice of appropriate
data sets. As with systemic toxicity determinations (previously
discussed), human data are preferable to animal data. In the
absence of appropriate human data, data from an animal species
whose biologic responses are similar to those of humans would be
selected. When this information is not available, data from the
most sensitive animal species/strain/sex combination are preferred.
The route of administration is also a consideration and, whenever
there is a choice, the route that most resembles the route of human
exposure is selected. When this is not possible, the route
difference is noted as a sign of uncertainty. Also, when the
incidence of tumors is significantly elevated at more than one
anatomical site, estimates of overall risk are made by determining
14-8
-------
the number of animals with tumors at one or more of these sites.
The number of benign tumors is generally combined with the number
of malignant tumors unless they are not considered to have
potential to progress to the associated malignancies of the same
histogenic origin (U.S. EPA, 1990).
14.3.2. Choice of Extrapolation Model. Since human risk due to
a carcinogen at low exposure levels cannot be directly determined,
a number of mathematical models and procedures have been developed
for use in extrapolating from high to low doses (U.S. EPA, 1990).
Extrapolation is ordinarily conducted by fitting a mathematical
model to the observed data and then extending the model (or a bound
on the risks it predicts) from the observed range down toward risks
expected at low exposure.
Cancer risk estimates based on adequate human epidemiologic
data are preferred over estimates based on animal studies.
However, in most cases, the cancer potency estimate must be
determined by applying a low-dose extrapolation model, such as the
linearized multistage procedure (the U.S. EPA's default procedure) ,
to animal data. Using this procedure, the slope factor is
determined by the upper bound of the 95% confidence limit on the
slope. (For the linearized multistage procedure, this is referred
to as the q.,*) . The units are per (mg/kg)/day.
When animal data are used as a basis for extrapolation, the
human dose that is equivalent, to the dose in the animal study is
calculated assuming that different species are equally sensitive
to the effects of a carcinogen if they absorb the same dose per
14-9
-------
unit of body surface area. This assumption is made only in the
absence of specific information about equivalent doses for the
chemical in question (U.S. EPA, 1990).
14.3.3. Route-to-Route Extrapolation. If data for one route of
exposure are positive, and there are no or insufficient data for
other routes of exposure, the chemical is assumed to be
carcinogenic by all routes and the resultant slope factor may be
adopted for all routes. In order to evaluate human risks for oral
exposure when only inhalation exposure has been tested in animals,
the slope factor for inhalation exposure can be adopted as the
slope factor for oral exposure with application of an adjustment
to reflect route-specific differences in extent of absorption.
This is accomplished by multiplying the slope factor by the ratio
of percent absorption by the oral route to the percent absorption
by the inhalation route if the values are available. If this
information is not available, assumptions regarding route-specific
differences in absorption are considered on a case-by-case basis,
and usually the assumption of equal percent absorption by the two
routes is made.
14.3.4. Confidence in Quantitation. The scientific data base used
to calculate and support the setting of cancer risk levels has an
inherent uncertainty that is due to the systematic and random
errors in scientific measurement. In most cases, only studies
using experimental animals have been performed. Thus, there is
uncertainty when the data are extrapolated to humans. When
14-10
-------
developing cancer risk levels, several other areas of uncertainty
exist, such as the incomplete knowledge concerning the health
effects of contaminants in the chemical, the impact of the age,
sex, and species of the experimental animals used, the nature of
the target organ system(s) examined and the effect of dose at the
target site in humans or experimental animals. When there is
concurrent exposure to more than one chemical, additional
uncertainty results from inadequate information concerning possible
synergistic, antagonistic, or masking effects.
14.3.5. Calculation of q* for Benzo(a)pyrene. Throughout this
document, sample calculations have been presented using data for
benzo(a)pyrene as the example of a carcinogen emitted by a MWC.
Thus, for explanatory purposes, it is appropriate to present the
calculation of a slope factor for this chemical. A carcinogenicity
slope factor for human oral exposure to benzo(a)pyrene has been
derived from a bioassay in which the chemical was given to strain
CFW mice for «110 days at dose levels ranging from 1-250 ppm in the
diet (Neal and Rigdon, 1967). From the data, the U.S. EPA (1980b)
derived a q,* of 11.5 per(mg/kg)/day. This value is not available
on IRIS, but has been published in U.S. EPA (1980b).
14.4. SUMMARY
Dose-response assessment for a chemical is an attempt to
define the relationship between the dose of a chemical and the
resulting toxic and/or carcinogenic effects. The result of this
assessment is a quantitative measure of a concentration or dose of
14-11
-------
a chemical above which humans are at risk for systemic toxicity,
or an estimate of carcinogenic potency of the chemical when humans
are exposed for a lifetime. In the case of systemic toxicity, this
dose is termed the reference dose, RfD; in the case of
carcinogenicity, this value is termed the slope factor. The amount
of confidence in the values determined for any given chemical
depends on the quality of the data on which the calculations are
based. The next chapter will use the results of the dose-response
assessment to determine a quantitative human risk from exposure to
chemicals from combustor facilities.
14-12
-------
15. RISK CHARACTERIZATION
15.1. INTRODUCTION
Risk assessments ordinarily proceed from source to receptor.
The source is first characterized and contaminant movement away
from the source is then modeled to estimate the degree of exposure
to the receptor, as described in Chapters 3-12. Health effects are
then predicted based on the estimated exposure. Risk
characterization is the process by which the significance of risk
estimates for exposure is evaluated.
The methodology presented here considers only the effects from
indirect exposures, that is, exposures following deposition of
pollutants to soil or surface waters. Although direct exposure due
to inhalation of emitted pollutants is not addressed in this
methodology, the risk characterization process would be similar to
that for indirect exposures. Human exposure to a contaminant by
ingestion or dermal contact is characterized by estimating daily
intake (DI) via food, soil, or water as described in Chapters 4-
12. For all exposure pathways, the daily intake represents the
increased amount of contaminant above background or existing levels
to which the individual is exposed due to operation of the
combustion facility. Estimates of daily intake have been
determined using various exposure scenarios as shown in Chapter 2.
Adjustments to exposure calculated for each pathway may be
15-1
-------
necessary to more accurately estimate lifetime exposure and
evaluate the potential for adverse effects on human health. These
possible adjustments to exposure are described in Section 15.2.
Risks for adverse human health effects are estimated
differently for systemic toxicants, assumed to be threshold-acting
chemicals, and carcinogens, assumed to be nonthreshold-acting
chemicals. For threshold-acting toxicants, daily intake is
compared with the reference dose for chronic exposure (RfD) to
determine if the contaminant poses a risk to human health. The
RfD is an estimate (with uncertainty spanning perhaps an order of
magnitude) of the daily exposure to the human population (including
sensitive subgroups) that is likely to be without appreciable risk
of deleterious effects during a lifetime.
For carcinogens, the daily intake is used to estimate excess
risk. The excess risk (ER) is defined as the incremental lifetime
cancer risk above background occurring in a hypothetical population
in which all individuals are exposed continuously to a
concentration equal to the daily intake of the contaminant. The
ER is derived from the daily incremental dose of the contaminant
above background and the human cancer potency factor as established
by the U.S. EPA. These approaches to estimating risk are
presented in Section 15.3. An uncertainty analysis is described
in Section 15.4.
15.2. ADJUSTMENTS TO EXPOSURE
Exposure scenarios developed for a site-specific assessment
may involve exposure via several routes and pathways. An
15-2
-------
individual may be exposed to emitted contaminants by several routes
during the time he resides in the vicinity of the combustor
facility. Daily intake, however, has been calculated independently
for each exposure pathway. In order to evaluate exposure by
several routes, the duration of exposure and relative effectiveness
of each route need to be considered to determine "total" exposure
to the individual.
15.2.1. Relative Effectiveness of Exposure (RE). Relative
effectiveness is a unitless factor that reflects the relative
toxicologic effectiveness of a chemical by different exposure
routes. RE is used to standardize effects of exposure by one route
to the effects of exposure by another route. Effectiveness is the
manifestation of the same inherent toxicity (same toxic endpoint)
by the different routes of exposure. The RE of exposure is an
experimentally determined value relating the dosages that produced
a particular toxicity. Therefore, RE can be used to express the
dose of one exposure route in terms of an equivalent dose of
another route. The value of RE may, in part, reflect observed or
estimated differences in absorption rates between different
exposure routes. It can also be related to differing sensitivity
of absorption sites to damage and/or differences in toxicokinetics
between exposure routes. In addition to route differences, RE can
reflect differences in exposure conditions or differences between
sources of exposure for the same route, such as ingestion of food
or drinking water.
In order to determine the health risk associated with daily
15-3
-------
intake by ingestion or dermal exposure to contaminants from
combustor emissions, the daily intake must be compared with a toxic
threshold or the carcinogenic potency of the contaminant by a
similar route of exposure. Before applying an RE adjustment to
exposure, the route of administration used for estimation of the
RfD or carcinogenic potency factor (such as the q.,*) must be known.
Exposures are adjusted by the RE factors with respect to the same
route of exposure as the health-based criteria. For example, if
an oral RfD (RfD0) for the chemical of interest is available, RE
factors would be applied to exposure determined by this methodology
with respect to the same route of exposure as the study from which
the RfD was derived (e.g., ingestion of food). Similarly,
exposures would be adjusted by an RE relative to inhalation for
comparison with the RfD,-. In cases of exposure from soil
ingestion, the RE values should take into account the soil matrix
if supporting data are available.
For exposure routes for which health-based criteria are not
currently available, such as for dermal exposure to toxicants
(i.e., RfD for chronic dermal exposure), the DI can be transformed
to an equivalent daily intake for the same route for which health-
based criteria are available. In the case of dermal exposure,
RE refers to the effectiveness of the absorbed dermal dose relative
to an unabsorbed ingested dose. This is because daily dermal
intake (DDI) is an estimate of absorbed dose, whereas the RfD0 is
based on external (i.e., ingested) dose. Very little information
is available concerning the toxicologic effectiveness of dermally-
absorbed contaminants. In cases where sufficient data are
15-4
-------
available to determine DDI, the DDI can be transformed to an
equivalent oral DI by multiplying the DDI by RED, or the ratio of
the dermally absorbed dose to the orally absorbed dose as shown in
Equation 15-1:
DI = DDI
REn
(Equation 15-1)
where:
DI = daily intake (oral equivalent) (mg/kg/day)
DDI = daily dermal intake (mg/kg/day)
RED = reciprocal of the absorption fraction for ingested
chemical
The RED is the reciprocal of the gastrointestinal absorption
fraction, which is equivalent to the RE for an absorbed dermal dose
relative to an unabsorbed ingested dose, assuming that the absorbed
dose is equivalently effective regardless of exposure route. The
equivalent oral DI can then be compared with the RfD0 or used to
determine the ER by the oral route. It is recognized that the
assumption of equivalent effectiveness of absorbed doses between
exposure routes does not hold in many cases, such as when effects
occur at the portal of entry or when removal, inactivation or
activation of the compound before reaching the target organ varies
with the exposure route.
In some cases, potency estimates have been derived from a
different route of exposure than that which may occur from food
chain contamination (e.g., inhalation). In these cases, the use
of RE for carcinogens is similar to that described for threshold-
acting toxicants.
15-5
-------
The RE factor should only be applied when well documented and
referenced information about the pharmacokinetics of the
contaminant are available. Frequently such experimental data are
lacking and RE may be set equal to 1; that is, exposure routes are
assumed to be equally effective. If the REs for adults and
children by the same route of exposure are known to be
significantly different, the DI for each exposure pathway can be
modified by the route-specific RE prior to adjustment for exposure
duration and calculation of total DI (see Sections 15.2.2. and
15.2.3.)-
15.2.2. Exposure Duration Adjustment (EDA). Human risk
assessments for nonthreshold (carcinogenic) toxicants can be
determined for lifetime (70 years) exposure or for exposures of
shorter duration (e.g., 16 or 30 years; see Section 2.4.4.). It
is assumed that the individual is exposed throughout this period
to a longterm average concentration of the pollutant in the
environment (i.e., soil or water) that resulted from continual
combustor emissions.
An exposure duration adjustment (EDA) may be used when
exposure duration is shorter than the assumed 70-year lifetime.
EDA, analogous to a time-weighted average in most cases, is the
exposure duration divided by 70 years. For example, in Scenario
A, an adult is exposed for 16 years and DI can be multiplied by an
EDA of 16/70 years to estimate a lifetime average intake. EDA may
also be used for exposures that may be more prevalent in children.
For example, soil ingestion in children (i.e., pica) is most
15-6
-------
prevalent up to 5 years of age. Thus, DI from soil ingestion can
be multiplied by an EDA of 5/70 years to estimate a lifetime
average.
For some exposure scenarios, an EDA may be used to estimate
childhood exposure and also exposure during part of adulthood.
For example, in Scenario B described in this document, DI is
determined for an individual during both childhood and adulthood.
An EDA of 5/30 is applied to the DI of the child and 25/30 is
multiplied by the intake of the adult to estimate total DI during
the 30 year exposure period. To further adjust for lifetime
exposure, the 30 year DI is multiplied by 30/70 years to estimate
an average daily dose over the lifetime. Equation 15-2 shows how
an exposure duration-adjusted daily intake can be determined.
DIA = [(DI (child) • EDA) + (DI (adult) • EDA)] • Ds/Ld
(Equation 15-2)
where:
DIA = daily intake-adjusted for pathway (mg/kg/day)
DI = daily intake for pathway (mg/kg/day)
EDA = exposure duration adjustment (unitless)
Ds = duration of scenario (yrs)
Ld = lifetime duration (assumed 70 yrs)
Exposure duration adjustment is used only with carcinogens
and should be carefully evaluated on a case by case basis, since
for some carcinogens, childhood may be the critical exposure period
and the carcinogenic risk would be underestimated by averaging over
the entire lifetime.
15-7
-------
15.2.2.1. EXAMPLE CALCULATION OP DAILY INTAKE-ADJUSTED—
Daily intake of benzo(a)pyrene by each pathway in Scenario B is
described in Chapters 5-7 and 9-12 and is shown in Table 15-1.
Use of exposure duration adjustments for these pathways is
illustrated by applying Equation 15-2 to the food ingestion pathway
to give the daily intake-adjusted:
DIA » [(3.61X10"5 • 5/30) + (7.23X10"6 • 25/30)] • 30/70
» 5.16X10"6 mg/kg/day
Table 15-2 shows the exposure duration-adjusted daily intake
of benzo(a)pyrene for the various pathways. The DI and DIA for
dermal exposure are not determined since data an input variable
were not available (refer to Chapter 7).
15.2.3. Estimating Total Exposure by Each Route. Total daily
intake (TDI) can be estimated by summing the DIA for all pathways
of the same exposure route relevant to the exposure scenario. In
this way, a TDI for oral (food, water, soil and fish) exposure and
a TDI for dermal (soil and water) exposure are estimated. (TDI for
inhalation exposure could also be included if incorporated into the
exposure scenario.) A limitation in this approach is that
potential interactions by multiple exposure routes are not
evaluated. However, to determine the effective TDI, the DI (DIA
for carcinogens) for each exposure route must be divided by that
route's relative effectiveness (RE) factor.
15-8
-------
TABLE 15-1
Results of Example Calculations of Daily Intake of
Benzo(a)pyrene in Scenario Ba
(mg/kg/day)
Pathway
from Soil
Soil Ingestion
Water Ingestion
Fish Ingestion
Child
4.51x10
-9
1.02x10
-8
1.93x10
-9
Adult
Food Ingestion
Dermal Absorption
3.61X10"5
NAb
7.23X10"6
NA
4.38X10
-11
3.72X10
-9
1.66x10
-9
aPresented as examples only; see text
''NA = not available (refer to Chapter 7 for explanation)
15-9
-------
TABLE 15-2
Results of Example Calculations of Daily Intake-Adjusted
of Benzo(a)pyrene in Scenario Ba
(mg/kg/day)
Pathway
Intake
Food Ingestion
Dermal Absorption
from Soil
Soil Ingestion
Water Ingestion
Fish Ingestion
All Pathways
5.16X10"6
NA°
3.39x10
-10
2.06x10
-9
7.31x10
-10
5.16x10
-6
"Presented as example only; see text
"NA » not available (refer to Chapter 7 for explanation)
15-10
-------
Thus, TDI is dependent on the DI (or DIA) for each pollutant, h,
and RE for each exposure pathway, k, and can be derived using the
following equations:
TDIhk =
DIhkx(food) DIhkx(water) DIhkx(soil) DIhkx^fish)
RE(food)
RE(water)
RE(soil)
RE(food)
where:
TDI
DI
RE
(Equation 15-3)
hk = total daily intake for the' hth pollutant by kth
route (mg/kg/day)
hk = daily intake of hth pollutant from a given
exposure route/pathway, k (mg/kg/day)
= relative effectiveness of the route of exposure,
with respect to the pertinent route of exposure
(e.g., food) (unitless)
h denotes route, which is oral in this equation
A is substituted for x for carcinogens (DIA should be used)
TDI
hk
where:
DIhkx (dermal-soil)
RE(dermal-soil)
DIhkx (dermal-water)
RE(dermal-water)
TDI
(Equation 15-4)
hk = total daily intake for the hth pollutant by kth
route (mg/kg/day)
DIhk = daily intake of hth pollutant from a given
exposure route/pathway, k (mg/kg/day)
RE = relative effectiveness of the route of exposure,
with respect to the pertinent route of exposure
(e.g., dermal) (unitless)
k denotes the route, which is dermal in this equation
A is substituted for x for carcinogens (DIA should be used)
Since fish is a food group the relative effectiveness for food
is used in the determination of a total daily oral intake.
15-11
-------
15.2.4. Total Background Intake of Pollutant by Each Route (TBI).
It is important to recognize that there may be sources of exposure
other than the combustor emissions, and that total exposure should
be maintained below the RfD or a specific risk level. These
sources of exposure include background levels (whether natural or
anthropogenic) in drinking water, food or air. Other types of
exposure that are due to occupation or habits such as smoking might
also be included, depending on data availability and regulatory
policy. These exposures are summed to estimate TBI.
Data for estimating background exposure usually are obtained
from analytical surveys of surface, ground or tap water, from FDA
market basket surveys, and from air monitoring surveys. These
surveys may report means, medians, percentiles or ranges, as well
as detection limits. Estimates of TBI may be based on values
representing central tendency or on upper-bound exposure
situations, depending on the desired degree of conservatism.
Data chosen to estimate TBI should be consistent with the
value of body weight. Where background data are reported in terms
of a concentration in air or water, ingestion or inhalation rates
applicable to adults or children can be used to estimate the daily
background intake value. Where data are reported as total daily
dietary intake for adults and similar values for children are
unavailable, conversion to an intake for children may be required.
Such a conversion could be estimated on the basis of relative total
food intake or relative total caloric intake between adults and
children. Background intake may also be adjusted for duration if
such information is available for children and adults.
15-12
-------
The TBI is the summed estimate of all possible background
exposures except exposures resulting from combustor emissions.
The effective TBI is determined by dividing the background intake
values (BI) for the pollutant, h, for each exposure route, k, by
the relative effectiveness (RE) factor of that route or source.
Thus, after all the background exposures have been determined, TBI
values can be calculated for oral and dermal exposure respectively,
using the following equation:
BI
TBI
hk
RE(food)
BIhk (water)
RE(water)
BI
RE(soil)
BIhk(fish)
RE(food)
(Equation 15-5)
where:
.th
TBIhk = total background intake rate of h pollutant
by kth route (mg/kg/day)
BIhk = background intake of h pollutant from the kth
route (mg/kg/day)
RE = relative effectiveness, with respect to the
pertinent route of exposure (e.g., food)
(unitless)
k denotes route, which is oral in this equation
BIh,(soil)
TBI
hk
where:
RE(soil)
BIhk (water)
RE(water)
(Equation 15-6)
th
TBIhk = total background intake of h pollutant by k
route, (mg/kg/day)
th
,th
BIhk = background intake of h.pollutant by k route
(mg/kg/day)
RE = relative effectiveness, with respect to the
pertinent route of exposure (unitless)
k denotes route, which is dermal in this equation
15-13
-------
A limitation in determination of the TBI is that potential
interactions (e.g., synergism or antagonism) by multiple exposure
routes are not evaluated.
Total background intake should be derived on a site-specific
basis. Lacking such data, background intake for the particular
geographic area is preferred over national averages. The degree
of urbanization where the exposed individual lives should also be
considered since the contribution to exposure from ambient
concentrations of certain pollutants (e.g., lead) may differ.
When TBI is subtracted from the RfD, the remainder (after
adjusting for RE) defines the increment that can result from
combustor emissions without exceeding the threshold. That is, the
TDI plus the TBI should not exceed the RfD. If upper-bound data
(such as 95th percentiles) were used to estimate TBI, then an
increase in exposure corresponding to this increment, if realized,
would cause the RfD to be approached or exceeded in a relatively
small percentage (5%) of the exposed population. If central-
tendency data (the median) were used to estimate TBI, such an
increase would cause the RfD to be approached or exceeded in about
half of the exposed population. If TBI were set at zero for lack
of exposure data, the allowed increase would result in exceeding
the RfD to an unknown degree, depending on whether other sources
of exposure exist.
Total background intake is not considered for nonthreshold
toxicants in this methodology since ER is determined and is an
estimate of incremental risk and not total carcinogenic risk (see
Section 15.3.3.).
15-14
-------
15.3. RISK ESTIMATION
The scenarios presented in this document do not determine
exposure or risk for an entire exposed population (aggregate risk) .
Rather, they determine the risk for an individual in a defined
exposure scenario without reference to the numbers of individuals
at a given exposure level. For estimation of risk to threshold-
acting toxicants (i.e., noncarcinogens), DI by ingestion (food,
water and/or soil) or dermal exposure (soil) is compared with the
RfD to characterize the risk to human health. A DI exceeding the
RfD would be a basis for concern that adverse health effects may
occur in those individuals.
For carcinogens, the DI is used to estimate excess
carcinogenic risk. The excess risk (ER) is defined as the
incremental lifetime cancer risk above background to a hypothetical
population in which all individuals are continuously exposed to a
concentration equal to the daily intake of the contaminant. The
ER is derived from the daily incremental dose of the contaminant
above background and the human cancer potency factor (e.g., q,*) as
established by U.S. EPA methods (U.S. EPA, 1986f). For example,
an ER of 10"6 means that lifetime exposure to that specific
concentration (or administered dose) would have an upper-bound
excess cancer risk of one case in one million individuals.
Exposure to members of a chemical class or a chemical congener
may be estimated from emission data using the proportion of the
total mass that the congener comprises of the chemical class.
Unless congener-specific data are available, all the congeners of
the chemical class (e.g., PAHs or PCBs) are assumed to have the
15-15
-------
same transport and fate characteristics. Furthermore, for the
purposes of risk assessment, the congener is assumed to comprise
the same proportion of the total chemical class in deposition,
soils, plants and animals as it does in the emissions. This allows
estimation of exposure to the congener as well as to the total
chemical class. Potential health effects resulting from exposure
to the congener or the chemical class can be evaluated using the
RfD or ER for the congener.
15.3.1. Comparison with the RfD. For threshold-acting toxicants,
risks due to the DI from the various pathways in the exposure
scenarios are assessed by comparison with the RfD. Values of DI
from the oral route (ingestion of food, soil, water and fish) are
summed to determine TDIhk (where k is the oral route) and compared
with the oral RfD. Pathways of exposure by dermal absorption of
contaminants in soil (and water, if available) may be converted to
an equivalent oral DI, summed and compared with the oral RfD.
One approach for assessment of risk to systemic toxicants by
multiple routes of exposure is the use of an approach analogous to
a hazard index (HI) (Stara et al., 1987), wherein the total daily
intake for the pollutant (h) is expressed relative to the RfD for
the pollutant by the same exposure route as shown in the following
equation:
15-16
-------
= TDIhk/RfD,
'hk
(Equation 15-7)
where:
HIhk = hazard index for the htn pollutant by the kth exposure
route (unitless)
TDIhk = total daily intake for the hth pollutant by the kth
route (such as inhalation, dermal) (mg/kg/day)
RfDhk = reference dose for chronic exposure to the h
pollutant by the k route (k is the same route as the
TDI) (mg/kg/day)
If RfDs for several routes of exposure are available,, a multi-
route (total) hazard index for a chemical, h,. can be determined
according to the following ec[uation:
HI, =
in ;<
sk=1TDIhk/RfDhlc
(Equation 15-8)
where:
HIh
TDI
RfD,
= hazard index for the hth chemical for m exposure
routes, k (unitless)
hk
'hk
= total daily intake for h chemical by k route
(mg/kg/day)
= reference dose for chronic exposure to hth
chemical by kth route (where k is the same route
as the TDI) (mg/kg/day)
It is assumed that no hazard exists if the HIh is less than
1, since the TDI does not exceed the RfD. If the HI is greater
than 1, the risk is defined, only in terms relative to the RfD.
This approach can only be used if the route-specific RfDs (i.e.,
RfD0, RfDj) are not based on pprtal-of-entry effects.
For multiple chemical exposures, such as with mixtures, by
several routes, a hazard index for each chemical by multiple routes
could be determined (HIh) . These values of HIh may then be summed
for groups of chemicals affecting a given target tissue according
to Equation 15-9:
15-17
-------
HI1
Mix
HI.
(Equation 15-9)
where:
HIHix
HI,
- hazard index for a mixture of n chemicals (unitless)
« hazard index for multiple route exposure to hth
chemical (unitless)
The summing of HIh for pollutants can only be used if the RfDs
for those chemicals are based on effects in the same target organ.
15.3.1.1. EXAMPLE CALCULATION: COMPARISON WITH RfD FOR
CADMIUM. The definition and derivation of the RfD was discussed
in Chapter 14. The total background intake (TBI) of cadmium for
adults has been estimated to be 27 jug/day (3.85xlO~4 mg/kg/day)
(Federal Register, 1985) . The oral RfD for cadmium in food is
IxlO"3 mg/kg/day and the oral RfD for cadmium in water is 5xlO"4
nig/kg/day (U.S. EPA, 1985b; U.S. EPA, 1990). It is implied that
children would not be adversely affected when exposed to the same
food and water that resulted in this threshold level in adults.
The DI of cadmium for the various pathways in Scenario B are shown
in Table 15-3. Scenario B is described in Chapter 2, and DI values
for cadmium were derived in example calculations presented in
Chapters 3-12. Although based on current research, these
calculations are provided only as examples and should not be taken
as definitive determinations of cadmium exposure for any particular
combustion facility.
15-18
-------
TABLE 15-3
Results of Example Calculations of Daily Intake of Cadmium
in Scenario Ba
(mg/kg/day)
Pathway
Child
Adult
Food Ingestion
Dermal Absorption
from Soil
Soil Ingestion
Water Ingestion
Fish Ingestion
6.62xlO"5
NAD
2.39X10"5
1.18x10
-5
3.95X10
-6
1.80X10
NA
2.32X10
4.30X10
3.41x10
-5
-7
-6
-6
^Presented as examples only? see text
''NA = not available (refer to Chapter 7 for explanation)
15-19
-------
Since an RfD for cadmium is available for both food and water
ingestion, pathways involving exposure from the same sources would
be compared with the respective RfD. DI from ingestion of food,
soil and fish are summed and the TDI compared with the oral RfD for
food, whereas the DI from water ingestion is compared with the oral
RfD for water.
The TDI for cadmium for food ingestion and water ingestion
routes is determined according to Equation 15-3. An RE of 1
relative to food is assumed, and adult values are used.
TDI
food
-5
TDI
water
1.80X10
2.16X10"5 mg/kg/day
4.30xlO"6 mg/kg/day
2.32x10
-7
3.41X10
-6
Since there is not a RfD for chronic dermal exposure, the DDI
of cadmium in soil could be converted to an equivalent oral DI,
summed with oral DIs and the TDI compared with one of the oral
RfDs. However, data were not available for the absorption fraction
of cadmium in soil for human skin and DDI was not calculated (see
also Chapter 7).
Hazard indices are calculated for food ingestion and water
ingestion since there are RfDs for these routes. The hazard
indices are summed to determine overall risk relative to the RfD.
The TBI for cadmium can also be subtracted from the RfD to show
the increment that can result from combustor emissions without
exceeding the threshold.
15-20
-------
The HI for cadmium for food ingestion according to Equation
15-7 is:
2.16X10"5
1X10
-3
= 2.2x10
-2
The HI for cadmium for water ingestion is:
4.30X10"6
= 8.6x10
-3
5x10"
The HI for cadmium for these routes is:
HI = 2.2X10"2 + 8.6X10"3
= 3.0X10"2
If there were several chemicals with the same target organ,
HIh could be determined for each chemical across its routes of
exposure and HI determined for the mixture according to Equation
15-9. Since for cadmium HI <1, the appropriate conclusion is that
cadmium does not present a risk of noncarcinogenic effects by
Scenario B. As stated in Section 13.3.1., cadmium is classified
as a probable human carcinogen, but insufficient information exists
to derive a quantitative estimate of carcinogenic risk for ingested
cadmium.
15.3.2. Determination of the Excess Risk Level (ER) for Non-
Threshold Toxicants (Carcinogens) . For chemicals classified as
carcinogens, the excess risk, ER, for a particular route of
exposure is determined as follows:
15-21
-------
ERhk - l-exp[-(TDIhk • RE - BHhk)]
(Equation 15-10)
where:
total daily intake for the hth pollutant by the kth route
ERhk » excess risk for the hth pollutant by kth exposure route
(unitless)
TDI
IIK _
(mg/kg/day)
RE = relative effectiveness, with respect to pertinent
route of exposure (unitless)
BHh|c = human cancer potency (BH) for the hth pollutant by k
route [per(mg/kg)/day]
.th
The definition and derivation of each of the parameters used
to estimate ER for carcinogens have been discussed in Sections
15.2.1., 15.2.2. and 15.2.4. and Chapter 14. TBI is not taken into
account in the above equation, since ER is an expression of
incremental risk due to the facility and not an expression of total
carcinogenic risk. It may be useful, however, to compare TBI with
DIA or TDI to determine the fraction of exposure to a given
carcinogen that is contributed by the facility.
ER is determined for each exposure route if a human cancer
potency estimate is available for that exposure route. Risk across
routes can be summed to determine the multiroute exposure risk for
chemical i by the following equation:
ERh
(Equation 15-11)
where :
ERhk
excess risk for hth chemical for m exposure routes
(unitless)
excess risk for hth chemical for kth exposure route
(unitless)
15-22
-------
This equation assumes independence of action for each route
and is equivalent to the assumption of route addition (dose and
response addition across routes). A hazard index approach using
a specific level of risk could also be applied, somewhat analogous
to that for systemic toxicants (U.S. EPA, 1986h).
For a mixture of carcinogenic chemicals, excess risk can be
estimated by the following equation (Stara et al., 1987):
ERMix = 1-
hn (i- ERH)
(Equation 15-12)
where:
ER'
ERh
Mix
excess risk for a mixture of n chemicals for
multiroute exposure (unitless)
= excess risk for the
routes (unitless)
.th
chemical for m exposure
This approach assumes independence of action by the
carcinogens in a mixture. Because of uncertainties in estimating
dose-response relationships for both single and multiple chemicals,
the excess risk should not be regarded as an estimate of absolute
risk. The assumptions, limitations and uncertainties in conducting
risk assessment of mixtures are detailed in U.S. EPA (1986h).
15.3.2.1. EXAMPLE CALCULATION OP ER FOR BENZO(A)PYRENE.
Excess risk calculations for each pathway of Scenario B are shown
below. The human cancer potency for oral exposure (q.,*) for
benzo(a)pyrene (B(a)P) has been calculated by the U.S. EPA to be
11.5 per(mg/kg)/day (U.S. EPA, 1980b). For purposes of
illustration, it will be assumed that this potency value is the
15-23
-------
same for all routes of exposure. The TBI (i.e., intake from
background sources of benzo(a)pyrene exposure not related to the
combustor) for adults is estimated to be «0.88 /ng/day (1.26xlO~3
mg/kg/day) (U.S. EPA, I980b).
Risk due to TBI is calculated using Equation 15-10:
Risk = 1 -exp[-(1.26xlO"3 • 1 • 11.5)]
= 1.44xlO"2
The DI. for benzo(a)pyrene has been calculated for each
" , ^
pathway in Scenario B (see Table 15-2). Scenario B is described in
Chapter 27 and DI values for benzo(a)pyrene were derived in example
calculations presented in Chapters 3-12. Although bases on current
research, these calculations are provided only as examples and
should not be taken as definitive determinations of benzo(a)pyrene
exposure for any particular combustion facility. Values of ER for
each pathway are calculated from the respective DIA as follows:
For the Food Ingestion Pathway:
ER = l-exp[-(5.16xlO"6 • 1 • 11.5)]
-5
= 5.93x10
For the Dermal Absorption from Soil Pathway:
ER - Cannot be guantitated as input data are not available for
calculation of DI
For the Soil Ingestion Pathway:
ER = l-exp[-(3.39xlO"10
=* 3.90X10"9
1 • 11.5)]
15-24
-------
For the Water Ingestion Pathway:
ER = l-exp[-(2.06xlO"9 • 1 • 11.5)]
= 2.37X10"8
For the Fish Ingestion Pathway:
ER = l-exp[-(7.3lxlO'10 • 1 • 11.5)]
= 8.41X10"9
The excess risk for benzo(a)pyrene all routes of exposure is:
ER
5.93X10'5 + 3.90X10'9 + 2.37X10'8 + 8.41xlO'9
= 5.93X10"5
These values represent a plausible upper-bound for excess risk
resulting from Scenario B. it can be seen that risk due to TBI
(i.e., that due to background sources of benzo(a)pyrene not related
to the combustor) is much greater than the calculated excess risk
(i.e., that due to the combustion facility).
15.4. CHARACTERIZATION OP UNCERTAINTY
There is an inherent uncertainty in the models presented in
this document to determine exposure to combustor emissions. An
uncertainty analysis can be used to estimate a range of model
values from which a best estimate can be derived. Such an
uncertainty analysis was undertaken for the terrestrial food chain
and human soil ingestion pathways by Belcher and Travis (1989).
Pollutant deposition rates were modeled for 24 operating MWCs in
15-25
-------
the United States. These rates and other input variables were
expressed as distributions and were sampled using a stratified,
random sampling procedure referred to as Latin Hypercube Sampling.
A range of contaminant intake values was determined from many
iterations of the procedure. Because of differences between
pollutant deposition rates used in Scenario B of the present
analysis and those included in the Belcher and Travis (1989)
analysis, the results of the two analyses in terms of pollutant
intake are not directly comparable. Details of the latter analysis
will not be presented here. However, the results for cadmium and
benzo(a)pyrene, presented below, serve to illustrate the potential
variability in computed pollutant intakes that can result from
variation of input parameters.
The Belcher and Travis (1989) analysis also examined the
percentage contribution of each component of the studied pathways
to total intake for each contaminant. The resulting breakdowns for
cadmium and benzo(a)pyrene are also presented below.
15.4.1. Uncertainty Analysis for Cadmium. Cadmium intake from
the terrestrial food chain and soil ingestion pathways was due
primarily to root uptake by vegetation (63% of total daily intake)
and consumption of contaminated meat and dairy products (21%).
Tables 15-4, 15-5 and 15-6 show these contributions. A probability
distribution of daily human intake by these pathways found a
geometric mean of 2.1xlO~3 /zg/day (Figure 15-1), with a range of
15-26
-------
uncertainty (from approximately the 2nd to the 98th percentile of
frequency) spanning approximately 2.7 orders of magnitude.
15.4.2. Uncertainty Analysis for Benzo(a)pyrene. Human
consumption of contaminated vegetation was the primary source of
human exposure to benzo(a)pyrene, contributing 72% of the total
for the pathways analyzed. The contribution of various vegetative
and meat groups to DI were also analyzed. Garden fruits were found
to make the greatest contribution (33%) of vegetative groups and
beef the greatest contribution (14%) for meat and dairy products.
Results are shown in Tables 15-7, 15-8 and 15-9. Figure 15-2 shows
the probability distribution for the daily intake of
benzo(a)pyrene. The range of uncertainty was approximately 4
orders of magnitude, or somewhat greater than for cadmium.
A similar approach to an uncertainty analysis can also be
conducted for other chemicals and the other pathways in this
methodology.
15.5. CONCLUSION
The risk due to indirect pathways of exposure to combustor
emissions is characterized differently for carcinogens and
noncarcinogens. Risk of carcinogens is determined by calculation
of an upper-bound estimate of excess carcinogenic risk. Risk of
noncarcinogens is calculated by comparison with an RfD using a
hazard index approach. Both approaches can be applied to the same
15-27
-------
TABLE 15-4
Breakdown of Cadmium Exposure for the Terrestrial Food Chain
and Human Soil Ingestion Pathways
(% contributions to average daily intake)*
1. Direct ingestion of contaminated soil:
2. Consumption of contaminated vegetation:
Atmospheric deposition component
Root uptake component
Air-to-plant transfer component
3. Consumption of contaminated meat and
dairy products:
Soil-Animal-Human Route
Plant-Animal-Human Route
12.9%
3.3%
62.8'
0.0%
13.3%
7.7%
Source: Belcher and Travis, 1989
15-28
-------
TABLE 15-5
Contribution to Daily Intake of ^Cadmium by Vegetation
Food Groups*
Source
Percentage of the
Total Intake
Garden fruits
Root vegetables
Potatoes
Leafy vegetables
Dried legumes
Grains
Fresh legumes
32.3%
19.6%
6.9%
4.7%
1.7%
0.6%
0.3%
Source: Belcher and Travis, 1989
15-29
-------
TABLE 15-6
Contribution to Daily Intake ^of Cadmium
by Meat Food Groups*
Source
Percentage of the
Total Intake
Beef liver
Dairy products
Poultry
Beef
Pork
Eggs
Lamb
12.3%
3.7%
3.1%
1.4%
0.3%
0.2%
Source: Belcher and Travis, 1989
15-30
-------
0.1
cd
0.01 -
ti
g 0.001
CO
o
0.0001 -
Percentage
- o.oi
10 20 30 40 50 60 70 BO 00
- o.ooi
- o.oooi
Figure 15-1. Probability distribution for cadmium daily
intake values by the Terrestrial Food
Chain and Human Soil Ingestion pathways
Source: Belcher and Travis, 1989
15-31
-------
TABLE 15-7
Breakdown of Benzo(a)pyrene Exposure for the Terrestrial Food
Chain and Human Soil Ingestion Pathways ^
(% contributions to average daily intake)
1. Direct ingestion of contaminated soil:
2. Consumption of contaminated vegetation:
Atmospheric deposition component
Root uptake component
Air-to-plant transfer component
3. Consumption of contaminated meat and
dairy products:
Soil-Animal-Human Route
Plant-Animal-Human Route
4.3%
7.2%
15.0%
50.2%
3.3%
20.0%
*Source: Belcher and Travis, 1989
15-32
-------
TABLE 15-8
Contribution to Daily Intake of Benzo(a)Pyrene by Vegetation
Food Groups
Source
Percentage. of, the
Total Intake
Garden fruits
Leafy vegetables
Potatoes
Dried legumes
Fresh legumes
Root vegetables
Grains
32.9%
12.8%
12 .1%
11.0%
2.6%
0.8%
<0.1%
Source: Belcher and Travis, 1989
15-33
-------
TABLE 15-9
Contribution to Daily Intake of Benzo(a)Pyrene
by Meat Food Groups
Source
Percentage of the
Total Intake
Beef
Dairy products
Beef liver
Lamb
Pork
Poultry
Eggs
13.7%
5.8%
3.5%
0.2%
0.1.%
Source: Belcher and Travis, 1989
15-34
-------
GM = 8.16E-6
30 40 50 60 70 60 80
0.001
- 0.0001
- 0.00001
- o.oooooi
- o.ooooooi
Percentage
Figure 15-2.
Probability distribution for benzo(a)-
pyrene daily intake values for the
Terrestrial Food Chain and Human Soil
Ingestion pathways
Source: Belcher and Travis, 1989
15-35
-------
chemical across multiple exposure routes. These approaches can
also be applied to multiple chemicals in a mixture across multiple
exposure routes, with some constraints.
Uncertainty analysis, such as may be carried out by a
stochastic sampling approach, is a useful tool for examining the
influence of particular assumptions, input values or pathways on
the overall results, and for estimating overall ranges of
uncertainty associated with the use of the methodology. Analyses
of this type are particularly helpful whenever the interpretation
and use of results derived from this methodology are important
components of a risk management process.
15-36
-------
16. REFERENCES
Bacci, E., D. Calamari, C. Gaggi and M. Vighi. 1989.
Bioconcentration of organic chemical vapors in plant leaves:
Experimental measurements and prediction. University pf Siena:
Siena, Italy.
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-
AC05-84OR21400.
Barnes, D.G. and M. Dourson. 1988. Reference dose (RfD):
Description and use in health risk assessments. Reg. Toxicol.
Pharmacol. 8:471-486.
Belcher, G.D. 1989. Office of Risk Analysis, ORNL. Memorandum
to R.J.F. Bruins, U.S. EPA, Cincinnati, October 23, 1989.
Belcher, G.D. and C.C. Travis. 1989. Modeling Support for the
RURA and Municipal Waste Combustion Projects: Final Report on
Sensitivity and Uncertainty Analysis for the Terrestrial Food
Chain Model. Prepared under XAG-1824-A020-A1 by ORNL for U.S. EPA,
Office of Health and Environmental Assessment, Environmental
Criteria and Assessment, Cincinnati, OH.
Bidleman, T.E. 1988. Atmospheric processes.
Technol. 22(4): 361-373.
Environ. Sci.
Binder, S., D. Sokal and D. Maughan. 1985. Estimating the amount
of soil ingested by young children through tracer elements. Draft.
Atlanta, GA: Centers for Disease Control, Center for Environmental
Health, Special Studies Branch.
Binder, S., D. Sokal and D. Maughan. 1986. The use of tracer
elements in estimating amount of soil ingested by young children.
Arch. Environ. Health. 41: 341-345.
Bohn, H., B.L. McNeal and G.A. O'Connor. 1985. Soil Chemistry.
Second edition. John Wiley and Sons, Inc. New York.
Boone, F.W., Y.C. Ng and J.M. Palms. 1981.
of radionuclide particulates. Health Phys.
Terrestrial pathways
41: 735-747.
Bowers, J. F. and A.J. Anderson. 1981. An Evaluation Study for
the Industrial Source Complex (ISC) Dispersion Model. H.E. Cramer
Co,Inc. U.S. EPA, Office of Air Quality Planning and Standards,
Research Triangle Park, NC. EPA-450/4-81-002.
16-1
-------
Brown, H.S., D.R. Bishop and C.A. Rowan. 1984. The role of skin
absorption as a route of exposure for volatile organic compounds
(VOCs) in drinking water. Am. J. Publ. Health. 74(5).
Burke, P. 1986. Household Mobility. Division of Housing and
Demographic Analysis. (In-house HUD document).
GARB (California Air Resources Board). 1986.
calculating dry deposition velocities using
Subroutines for
Sehmel's curves.
Prepared by Bart Croes, California Air Resources Board.
CAST (Council for Agricultural Science and Technology). 1980.
Effects of sewage sludge on the cadmium and zinc content of crops.
Report No. 83. ISSN 0194-4088.
Clausing, P., B. Brunekreef and J.H. van Wijnen. 1987. A method
for estimating soil ingestion by children. Int. Arch. Occup.
Environ. Health. 59: 73-82.
Cowerd and Guenther. 1976. Development of a Methodology and
Emission Inventory for Fugitive Dust for the Regional Air Pollution
Study. U.S. EPA, Office of Air and Waste Management, Office of Air
Quality Planning and Standards, Research Triangle Park, NC. EPA-
450/3-76-003.
Elliott, H.A., M.R. Liberati and C.P. Huang. 1986. Competitive
adsorption of heavy metals by soils. J. Environ. Qual. 15(3):
214-219.
ERT (Environmental Research and Technology). 1987. User's Guide
to the Rough Terrain Diffusion Model (RTDM), Rev 3.20. ERT
Document No. P-D535-585. Environmental Research and Technology,
Inc., Concord, Ma. (Docket Reference No. IV-D-5).
Farino, W., P. Spawn, M. Jasinski and B. Murphy. 1983. Evaluation
and selection of models for estimating air emissions from hazardous
waste treatment, storage and disposal facilities. Prepared by GCA
Corporation, Bedford, MA, under Contract No. 68-02-3168, for the
Office of Solid Waste, U.S. EPA, Washington, DC.
Federal Register. 1985. National primary drinking water
regulations; synthetic organic chemicals, inorganic chemicals and
microorganisms, proposed rule. 50: 46936-47022.
Federal Register. 1989. Standards for the disposal of sewage
sludge. 54(23): 5746-5902.
Feldman, R.J. and H.I. Maibach. 1974. Percutaneous penetration
of some pesticides and herbicides in man. Toxicol. Appl.
Pharmacol. 28: 126-132.
16-2
-------
Fradkin, L., R.J.F. Bruins, D.H. Cleverly, et al. 1988. Assessing
the risk of incinerating municipal solid waste: The development and
application of a methodology™ In; Municipal Waste Combustion and
Human Health. CRC Press, Palm Beach, FL. (In press)
Freeman, H.M., Ed. 1988. Incinerating Hazardous Wastes.
Technomic Publishing Co., Inc., Lancaster, PA.
Freeman, H.M., Ed. 1989. Standard Handbook of Hazardous Waste
Treatment and Disposal. McGraw Hill, New York, NY.
Freeze, R.A. and J.A. Cherry. 1979. Groundwater. Prentice Hall,
Inc., Englewood Cliffs, NJ.
Giordano, P.M., A.D. Behel, J.E. Lawrence, J.M. Solleau and B.N.
Bradford. 1983. Mobility in soil and plant availability of metals
derived from incinerated municipal refuse. Environ. Sci. Technol.
17: 193-198.
Goeden, H.M. and A.H. Smith. 1989. Estimation of human exposure
from fish contaminated with dioxins and furans emitted by a
resource-recovery facility. Risk Analysis. 9(3): 377-383.
Harter, R.D. 1983. Effect of soil pH on adsorption of lead,
copper, zinc, and nickel. Soil Sci. Soc. Am. J. 47: 47-51.
Hawley, J.K. 1985. Assessment of health risk from exposure to
contaminated soil. Risk Analysis 5(4): 289-302.
Hillel, D. 1980. Fundamentals of Soil Physics. Academic Press,
Inc. New York.
Hoffman, F.O. and C.F. Baes. 1979. A statistical analysis of
selected parameters for predicting food chain transport and
internal dose of radionuclides. ORNL/NUREG/TM-882.
Jury, W.A. 1986. Volatilization from the soil. In: Vadose
Zone Modeling of Organic Pollutants. S.C. Hern and S.M. Melancorn,
eds. Lewis Publishers, Inc. Chelsea, MI.
Jury, W.A. and R.L. Valentine. 1986. Transport mechanisms and
loss pathways for chemicals in soils. In: Vadose Zone Modeling
of Organic Pollutants. S.C. Hern and S.M. Melancorn, eds. Lewis
Publishers, Inc. Chelsea, MI.
Karickhoff, S.W. 1981. Semi-emperical estimation of sorption of
hydrophobia pollutants on natural sediments and soils. Chemosphere.
10 (8): 833-846.
Kimbrough, R.D., H. Falk, P. Stehr and G. Fries. 1984. Health
implications of 2,3,7,8-tetrachlorodibenzodioxin (TCDD)
contamination of residential soil. J. Toxicol. Environ. Health
14: 47-93.
16-3
-------
LaBauve, J.M., J. Kotuby-Amacher and R.P. Gambrell. 1988. The
effect of soil properties and a synthetic municipal landfill
leachate on the retention of Cd, Ni, Pb, and Zn in soil and
sediment materials. J. Water Pollut. Control Fed. 60(3): 379-
385.
Lagerwerff, J.V. 1971. Uptake of Cd, Pb, and Zn by radish from
soil and air. Soil Sci. ill: 129-133.
LaGoy, P.K. 1987. Estimated soil ingestion rates for use in risk
assessment. Risk Analysis. 7(3): 355-359.
Lepow, M.L. , L. Bruckman, M. Gillette, S. Markowitz, R. Robino and
J. Kapish. 1975. Investigations into sources of lead in the
environment of urban children. Environ. Res. 10: 415-426.
Logan, T.J. and R.L. Chaney. 1983. Utilization of municipal
wastewater and sludge on land — metals. in; Proc. 1983 Workshop
on Utilization of Municipal Wastewater and Sludge on Land.
Riverside, CA. p. 235-323.
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt. 1982. Handbook of
Chemical Property Estimation Methods. McGraw-Hill Book Co.: New
York, NY.
Maibach, J.I., R.J. Feldmann, T.H. Milby and W.F. Serat. 1971.
Regional variation in percutaneous penetration in man. Arch.
Environ. Health. 23: 208-211.
McKone, T.E. and P.B. Ryan. 1989. Human exposures to chemicals
through ^food chains: An uncertainty analysis. Livermore,
California: Lawrence Livermore National Laboratory Report. UCRL-
99290.
Miller, C.W. and F.O. Hoffman. 1983. An examination of the
environmental half-time for radionuclides deposited on vegetation.
Health Phys. 45(3): 731-744.
Myers, R.E. 1989. U.S. EPA, Washington, DC. Memorandum to Files:
8.41.5. July 12.
NAS (National Academy of Sciences). 1987. Predicting Feed Intake
of Food-Producing Animals. National Research Council, Committee
on Animal Nutrition, Washington, DC.
Neal, J. and R.H. Rigdon. 1967. Gastric tumors in mice fed
benzo[a]pyrene: A quantitative study. Tex. Rep. Biol. Med.
25:553.
Neeley, W.B., D.R. Bronson and G.E. Blau. 1974. Partition
coefficient to measure bioconcentration potential of organic
chemicals in fish. Environ. Sci. Technol. 8: 1113-1115.
16-4
-------
Nelson, W.E., et al.(eds.) 1969. Textbook of Pediatrics. 9th Ed.
W.B. Saunders Co., Philadelphia.
Ng, Y.C., Colsher, C.S. and S.E. Thompson. 1982. Transfer
coefficients for assessing the dose from radionuclides in meat and
eggs. U.S. Nuclear Regulatory Commission. Final Report.
NUREG/CR-2976.
Norstrom, R.J., A.E. McKinnon and A.S.W. DeFrietas. 1976. A
bioenergetics-based model for pollutant accumulation by fish.
Simulation of PCB and methylmercury residue levels in Ohawa River
yellow perch (Perca flayescens). J. Fish. Res. Board Can. 33:
248-267.
NRC (National Research Council). 1983. Risk Assessment in the
Federal Government: Managing the Process. NAS Press, Washington,
DC.
NRC (National Research Council). 1984. Mineral tolerances of
domestic animals. Washington, DC: Subcommittee on mineral toxicity
in animals.
Page, A.L., T.J. Logan and L.E. Summers. 1986. Report on the
workshop on effects of sewage sludge quality and soil properties
on plant uptake of sludge-applied trace contaminants. November,
1985. Las Vegas, NV. Draft. ,
Paine, R.J. and B.A. Egan. 1986. Results of additional evaluation
of the rough terrain diffusion model (RTDM). For presentation at
the 79th Annual Meeting of the Air Pollution Control Association
(ACPA), Minneapolis, MN, June.
PEI Associates, Inc. and H.E. Cramer Company, Inc. 1986. Air
quality modeling analysis of municipal waste combustors. Prepared
for Monitoring and Data Analysis Division, Office of Air Quality
Planning and Standards, Research Triangle Park, NC under contract
68-02-3890.
Poe, S.H. 1988. Equilibrium vapor phase adsorption of volatile
organic chemicals on dry soils. J. Hazard. Mater. 19: 17-32.
Poiger, H. and C. Schlatter. 1980. Influence of solvents and
adsorbents on dermal and intestinal adsorption of TCDD. Food
Cosmet. Toxicol. 18: 477-481.
Risk Assessment Study of the Dickerson Site. 1988. Prepared by
the Maryland Power Plant and Environmental Review Division.
Preliminary Draft Report for Panel Review.
Roels, H.A., J.P. Buchet, R.R. Lauwerys, et al. 1980. Exposure
to lead by the oral and the pulmonary routes of children living in
the vicinity of a primary lead smelter. Environ. Res. 22: 81-
94.
16-5
-------
Ryan, J.A., H.R. Pahren and J.B. Lucas. 1982. Controlling cadmium
in the human food chain: A review and rationale based on health
effects. Environ. Research. 28: 251-302.
Schaefer, V.J., V.A. Mohnen and V.R. Veirs. 1972. Air quality of
American homes. Science. 172: 173-175.
Sehmel, G.A. 1980,
Int. 4: 107-127.
Particle resuspension: a review. Environ.
Shor, R.W., C. Baes III and R. Sharp. 1982. Agricultural
Production in the United States by Country: A Compilation of
Information from the 1974 Census of Agriculture for Use in
Terrestrial Food-Chain Transport and Assessment Models. ORNL-
5786.
Shu, H., P. Teitelbaum, A.S. Webb, et al. 1988. Bioavailability
of soil-bound TCDD: Dermal bioavailability in the rat. Fund. Appl.
Toxicol. 10(2): 335-343.
Simmonds, J.R. and G.S. Linsley. 1981. A dynamic modeling system
for the transfer of radioactivity in terrestrial food chains.
Nucl. Saf. 22(6): 766-777.
Skidmore, E.L. and N.P. Woodruff. 1968. Wind erosion forces in
the United States and their use in predicting soil loss.
Agriculture Handbook No. 346. U.S. Department of Agriculture,
Agricultural Research Service, Washington, DC.
Slade, D.H. 1968. Meteorology and Atomic Energy. U.S. Atomic
Energy Commission, Oak Ridge, Tennessee.
Slinn, W.G.N. 1984. Precipitation scavenging. In: Atmospheric
Science and Power Production, D. Randerson, Ed. DOE/TIC-27603.
Technical Information Center, U.S. Department of Energy.
Smith, W.J., F.W. Whicker and H.R. Meyer. 1982. Review and
categorization of saltation, suspension, and resuspension models.
Nucl. Saf. 23(6): 685-699.
Snyder, W.S. 1975. Report of the task group on reference man.
International Commission of Radiological Protection. No.
23,Pergamon Press: New York, NY.
Solomon, R.L. and J.W. Hartford. 1976. Lead and cadmium in dusts
and soils in a small urban community. Environ. Sci. Technol. 10-
773-777.
Sparks, D.L. 1989. Kinetics of Soil Chemical Processes. Academic
Press, Inc. San Diego, CA.
16-6
-------
Spencer, W.F., W.J. Farmer and W.A. Jury. 1982. Review: Behavior
of organic chemicals at air, water interfaces as related to
predicting transport and volatilization of organic pollutants.
Environ. Toxicol. Chem. 1: 17-26.
Stara, J.F., R.J.F. Bruins, M.L. Dourson, et al. 1987. Risk
assessment is a developing science: Approaches to improve
evaluation of single chemicals and chemical mixtures. In: Methods
for Assessing the Effects of Mixtures of Chemicals. V.B. Vouk,
G.C. Butler, A.C. Upton, D.V. Parke and S.C. Asher, Eds. SCOPE
1987.
Stephan, C.E. 1980. Personal memo. U.S. EPA. July, 1980.
Tesche, T.W., R. Kapahi, R. Honrath and W. Dietrich. 1987.
Improved dry deposition estimates for air toxics risk assessment.
Prepared by the Radian Corporation for presentation at the 80th
Annual Meeting of the Air Pollution Control Association (ACPA), New
York, NY.
Thorton, I. and P. Abrams. 1983. Soil ingestion: A major pathway
of heavy metals into livestock grazing contaminated land. Sci.
Total Environ. 28: 287-294.
Travis, C.C. and A.D. Arms. 1988. Bioconcentration of organics
in beef, milk, and vegetation. Environ. Sci. Technol. 22(3): 271-
274.
Travis, C.C. and H.A. Hattemer-Frey. 1988. Uptake of organics by
aerial plant parts: A call for research. Chemosphere. 17(2): 277-
283.
->
Travis, C. C. and M.W. Yambert. 1989. An evaluation of models
used to compute particulate dry deposition velocity to pasture,
agricultural land, and similar surfaces. In; EPA/ORNL Workshop on
Risk Analysis for Municipal Waste Combustion: Deposition, Food
Chain Impacts, Uncertainty, and Research Needs. Cincinnati, OH
(In Press).
Travis, C.C., C.F. Baes, III, L.W. Barnthouse, et al. 1983.
Exposure assessment methodology and reference environments for
synfuel risk analysis. Oak Ridge National Laboratory. ORNL/TM-
8672. Prepared for U.S. EPA, Office of Research and Development
under Interagency Agreement EPA 82-D-X0533 and DOE-40-740-78.
Turner, D." B. 1986. Fortran computer code/user's guide for
COMPLEX I Version 86064: An Air Quality Dispersion Model in Section
4. Additional Models for Regulatory Use. Source File 31 Contained
in UNAMAP (VERSION 6) (NTIS PB86-222361/AS).
U.S. Bureau of the Census. 1986. Current Population Reports,
Series P-20, No. 407, Geographical Mobility: March 1983 to March
1984. U.S. Government Printing Office, Washington, DC.
16-7
-------
U.S. Bureau of the Census. 1987. Statistical Abstract of the
United States: 1987. 107th edition. Washington, B.C.
U.S. Department of Agriculture.
Agricultural Handbook No. 8.
1975. Composition of foods.
U.S. Department of Agriculture. 1983. "Food Consumption:
Households in the United States,; Seasons and Year 1977-1978.
Consumer Nutrition Division, Hyattsville, MD. NFCS 1977-78
Report No. H-6.
U.S. Department of Commerce. n.d. Local Climatological Data
Annual Summaries for 1977. National Oceanic and Atmospheric
Administration/Environmental Data Service/National Climatic Data
Center. , .
U.S. Department of Commerce. 1985. Fisheries of the United
States, April, 1985. Current fisheries statistics No. 8360.
Washington, DC.
U.S. EPA. 1979. Industrial Source Complex (ISC) Dispersion Model
User's Guide, Vol. I. Prepared by the H.E. Cramer Company, Salt
Lake City, Utah for the Office of Air Quality Planning and
Standards, Research Triangle Park, NC. EPA-450/4-79-030. NTIS
PB80-133044. '
U.S. EPA. 1980(a). User's Guide for MPTER: A Multiple POJ nt
Gaussian Dispersion Algorithm with Optional Terrain Adjustment.
Environmental Sciences Research Laboratory, RTP, North Carolina
EPA-600/8-80-016. NTIS PB80-197361.
U.S. EPA. 1980(b). Ambient Water Quality Criteria for Polynuclear
Aromatic Hydrocarbons. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment
Office, Cincinnati, OH for the Office of Water Regulations and
Standards, Washington, DC. EPA-440/5-80-069. NTIS PB81-117806.
U.S. EPA. 1980(c). Ambient Water Quality Criteria Document for
Cadmium. Prepared by the Office of Health and Environmental
Assessment, Environmental Criteria and Assessment Office
Cincinnati, OH for the Office of Water Regulations and Standards
Washington, DC. EPA-440/5-80-025. NTIS PB81-117368.
U.S. EPA. 1980(d). Seafood Consumption Data Analysis: Final
report. Prepared by SRI International, Menlo Park, CA, under
Contract No. 68-01-3887.
U.S. EPA. 1980(e). Water Quality Criteria Documents.
Availability. Federal Register. 45(231): 79318-79379.
16-8
-------
U.S. EPA. 1982 (a). MPTER-DS: The MPTER Model Including Deposition
and Sedimentation. Prepared by the Atmospheric Turbulence and
Diffusion Laboratory (NOAA), Oak Ridge, TN for Environmental
Sciences Research Lab, Research Triangle Park, NC. EPA-600/8-82-
024. NTIS PB83-114207.
U.S. EPA. 1982(b). Pesticides Assessment Guidelines Subdivision
O: Residue Chemistry. Office of Pesticides and Toxic Substances,
Washington, DC. EPA 540/9-82-023.
U.S. EPA. 1982(c). Dermatotoxicity. Office of Pesticides and
Toxic Substances, Washington, DC. EPA-560/11-82-002. NTIS PB 82-
224320.
U.S. EPA. 1984(a). Air Quality Criteria for Lead. Office of
Health and Environmental Assessment, Environmental Criteria and
Assessment Office, Research Triangle Park, NC. EPA-600/8-83-028F.
NTIS PB84-142378.
U.S. EPA. 1984 (b). Risk Analysis of TCDD Contaminated Soil.
Office of Health and Environmental Assessment, Washington, DC.
EPA-600/8-84-031. NTIS PB85-145704/AS.
U.S. EPA. 1985(a). Rapid Assessment of Exposure to Particulate
Emissions from Surface Contamination Sites. Prepared by Midwest
Research Institute under contract 68-0303116. Office of Health
and Environmental Assessment, Washington, DC. EPA/600/8-85/002.
NTIS PB85-192219.
U.S. EPA. 1985(b). Environmental Profiles and Hazard Indices for
Constituents of Municipal Sludge: Cadmium. Office of Water
Regulations and Standards, Washington, DC.
U.S. EPA. 1985(c). Drinking Water Criteria Document for Cadmium.
Office of Drinking Water, Washington, DC. NTIS PB89-19140/AS.
U.S. EPA. 1985(d). Updated Mutagenicity and Carcinogenicity
Assessment of Cadmium: Addendum to the Health Assessment Document
for Cadmium. Office of Health and Environmental Assessment,
Washington, DC. EPA 600/8-83-025F. NTIS PB85-243533/AS.
U.S. EPA. 1986(a).' Reference Values for Risk Assessment.
Prepared by the Office and Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH for
the Office of Solid Waste, Waishington, DC.
U.S. EPA. 1986(b). Air Quality Modeling Analysis of Municipal
Waste Combustors. Prepared by PEI Associates, Cincinnati, OH.
U.S. EPA. 1986(c). Guideline on Air Quality Models (revised).
Office of Air Quality and Planning and Standards, Research Triangle
Park, NC. EPA-450/2-78-072R. NTIS PB86-245248.
16-9
-------
U.S. EPA. 1986 (d) .
Office of Emergency
EPA/540/1-86/060.
Superfund Public Health Evaluation Manual.
and Remedial Response. Washington, D.C.
U.S. EPA. 1986(e). 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. EPA. 1986(f). Guidelines for Carcinogen Risk Assessment.
Federal Register. 51 (185): 33992-34003.
U.S. EPA. 1986(g). Guidelines for Health Risk Assessment of
Chemical Mixtures. Federal Register. 51(185): 34014-34025.
U.S. EPA. 1987(a). Methodology for the Assessment of Health Risks
Associated with Multiple Pathway Exposure to Municipal Waste
Combustor Emissions. Prepared by the Office of Health and
Environmental Assessment, Environmental Criteria and Assessment
Office, Cincinnati, OH.
U.S. EPA. 1987(b). Air Dipersion Modeling of a Municipal Waste
Combustor in Rutland, Vermont. Prepared by PEI Associates,
Cincinnati, OH.
U.S. EPA. 1987(c). MWC Study: Assessment of Health Risks
Associated with Municipal Waste Combustion Emissions. Office of
Solid Waste and Emergency Response, Washington, DC. EPA/530-SW-
87-021g. NTIS PB87-206132/REB.
U.S. EPA. 1988(a). Estimating Exposures to 2,3,7,8-TCDD. Office
of Health and Environmental Assessment, Office of Research and
Development, Washington, DC. EPA/600/6-88/005A. NTIS PB88-
231196/AS.
U.S. EPA. 1988(b). Superfund Exposure Assessment Manual. Office
of Emergency and Remedial Response, Washington, DC. EPA/540/1-
88/001. NTIS PB89-167985.
U.S. EPA. 1988(c). Treatment Potential for 56 EPA Listed
Hazardous Chemicals in Soil. Robert S. Kerr Environmental Research
Laboratory, Ada, OK. EPA/600/6-88/001.
U.S. EPA. 1988(d). Distribution of Intake of Foods in Seven Food
Groupings. Prepared by Technical Assessment Systems, Inc. for the
Office of Health and Environmental Assessment, Environmental
Criteria and Assessment Office, Cincinnati, OH.
U.S. EPA. 1989(a). Exposure Factors Handbook. Office of Health
and Environmental Assessment, Exposure Assessment Group,
Washington, DC. EPA/600/8-89/043. NTIS PB90-106774.
16-10
-------
U.S. EPA. 1989 (b) . Development of Risk Assessment Methodology for
Land Application and Distribution and Marketing of Municipal
Sludge. Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH. EPA
600/6-89/001. NTIS PB90-135740/AS.
U.S. EPA. 1989(c). Interim Methods for Development of Inhalation
Reference Doses. Office of Health and Environmental Assessment,
Washington, DC. EPA 600/8-88/066F.
U.S. EPA. 1990. Integrated Risk Information System (IRIS).
Online. Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH.
Valentine, R.L. 1986. Nonbiological transformation. In; S.C.
Hern and S.M. Melancon, eds. Vadose Zone Modeling of Organic
Pollutants. Lewis Publishers, Inc. Chelsea, MI.
Valentine and Schnoor, 1986. Biotransformation. In; S.C. Hern and
S.M. Melancon, eds. Vadose Zone Modeling of Organic Pollutants.
Lewis Publishers, Inc. Chelsea, MI.
Veith, G.D., D.L. Foe and B.V. Bergstedt. 1979. Measuring and
estimating the bioconcentration factor of chemicals in fish. J.
Fish Res. Board Can. 36: 1040-1048.
Veith, G.D., M.J. Macek, S.R. Petrcelli and J. Carroll. 1980. An
evaluation of using partition coefficients and water solubility to
estimate BCFs for organic chemicals in fish. In; Aquatic
Toxicology, J.G. Eaton, P.R. Parrish, A.C. Hendricks, Eds. p. 116-
129. ASTM STP 707.
Wischmeier, W.H. and D.D. Smith. 1978. Predicting rainfall
erosion losses — A guide to conservation planning. USDA Handbook
No. 537.
Yang, Y.Y. and C.B. Nelson. 1986. An estimation of daily food
usage factors for assessing radionuclide intakes in the U.S.
population. Health Phys. 50: 245-257.
16-11
-------
-------
APPENDIX A
A-l
-------
= radius from source (m)
FIGURE A-l
The concentric rings used in the COMPDEP model.
Radii not drawn to scale.
A-2
-------
TABLE A-l
Location of Receptor
Receptor Number
Radius
(m)
1-16
17 - 32
33 - 48
49 - 64
65 - 80
81 - 96
97 - 112
113 - 128
129 - 144
145 - 160
200
500
1000
2000
5000
10000
20000
30000
40000
50000
A-3
-------
APPENDIX A
Equation to Calculate Areal Average Deposition
Daa = S [ (
CD, -
)Z (D2 -D3)
n-1
n-1 - Dn)
where:
Daa = areal average deposition (g/m/yr)
D = concentration at ith radius
r = radius from source (m)
r ~
r ~
10
200 m
500 m
1000 m
2000 m
5000 m
10,000 m
20,000 m
30,000 m
40,000 m
50,000 m
A-4
U.S. GOVERNMENT PRINTING OFFICE: 1990— 7 If 8 - 1 S 9 / 00"tl9
------- |