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

-------