Risk Analysis of TCDD Contaminated Soil


                                            PE85-1157CU
                                        EPA-600/8-84-031
                                        November  1984
   RISK ANALYSIS OF  TCUU  tONTAiljMnTEU
                 Jonn  Scnaum

          Exposure  Assessment Group
OFFICE OF HEALTH AND ENVIRONMENTAL ASSESSMENT
      OFFICE OF RESE,\RCH AND DEVELOPMENT
     U.S. ENVIRONMENTAL PROTECTION AGENCY
            WASHINGTON, DC 20A60

-------
                                                               PB85-145704
   Risk Analysis of TCDD
   Contaminated Soil
    (U.S.) Environmental Protection Agency
   Washington, DC
   Hov 84
J.S.D«
(afeoi Tedacd b&natSa Scr^ca

-------
TECHNICAL REPORT DATA
tPlcotr read Ir.nruf nun: un UK rc.irst Ociort roT.'
i RCPORT so.
CPA-600/8-S4-031
2.
4. TITLE AND SUBTITLE

Risk Analysis of TCOD Contaminated Soil
i REPORT DATE
November 1984
6 PERFORMING ORGANIZATION CODE
7. AUTI'ORIS)

John Schaum
t. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAMt AND ADDRESS
Exposure Assessment Group (RD-6S9)
US EPA, Office of Research and Development
Washington, DC 20460
10. PROGRAM ELEMENT NO.
It. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Washington, DC 20460
13. TYPE Of REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY COD«-

EPA/600/21
1S.SUPPLCMENTARY NOTES
IB. ABSTRACT
This paper provides a methodology for estimating the human exposure and
cancer risk associated with 2.3,7,8-TCDD contaminated soil. Five exposure
pathways are addressed: dust inhalation, fish ingestion, dermal absorption,
soil ingestion, and beef/dairy products ingestion. For each pathway, factors
describing contact rate, absorption fraction, and exposure duration are presented
along with the equations for calculating exposure levels and associated cancer
risk. The methodology features the use of nomographs'to provide quick and
approximate estimates of risk. More detailed procedures are also provided for
more accurate estimates. *
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
C. COSATI I ttld/GlOUp
,l|JJ|jSplSTRIBUTION STATEMENT

•. Distribute to public
19. SECURITY CLASS (T/lU fit port/

Unclassified
21. NO. Of PACES
58
70. SECURITY CLASS
-------
DISCLAIMED

Tnis report has Deen reviewed in accordance with U.S. Environmental

Protection A5ency policy ana approved for publication. Mention of trade wmes

or commercial products does not constitute endorsement or recommendation for

use.
ii
-------
TABLE OF CONTENTS

Page

Tables and Figures v

Foreword v1

Abstract vii

Acknowledgments viii

1.0 Approach 1
1.1 Estimation of Cancer Risk 2
1.2 Estimation of Human Exposure 3

2.0 Exposure Routes 4
2.1 Body Weight 8
2.2 Lifetime 9
2.3 Degradation of TCDD in Soil 11
2.4 Exposed Populations 14
2.5 Risk from Combined Exposure Pathways 15

3.0 Dust Inhalation 15
3.1 Contact Rate 16
3.2 Exposure Duration 17
3.3 Absorption Fraction 18

4.0 F1sh Ingestlon 21
4.1 Bloaccumulation 22
4.2 Consumption Rate 22
4.3 Exposure Duration 23
4.4 Absorption Fraction 23

5.0 Dermal Absorption 24
5.1 Contact Rate 24
5.2 Exposure Duration 26
5.3 Exposed Surface Area 26
5.4 Absorption 27

6.0 Scil Ingestlon 27
6.1 Contact Rate 29
6.2 Exposure Duration 29
6.3 Absorption 30

7.0 Beef/Dairy Products Ingestlon . 30
7.1 Bloaccumulation. 31
7.2 Consumption Rates 33
-------
Table of Contents continued...

7.3 Exposure Duration 33
7.4 Absorption • 36
8.0 Discussion of Uncertainty 36
9.0 How to Use Nomographs 37
References 41
Appendix - Nomographs. 44
1v
-------
TABLES AND FIGURES

Table 1. Properties of TCDO 5
Table 2. Distribution of Inspired Particles 20
Table 3. Amount of Soil Accumulated during Lifetime 28
Table 4 Beef/Dairy Products Ingestion Rates 34
Table 5. Summary of Exposure Factors Used 1n Tier 1 Calculations 39
Figure 1. Age vs Body Weight 10
Figure 2. Risk vs Half-Life . . . . . . .V. "."12"
Figure 3. Inspired Fraction vs Particle Size 19
Figure A-l Nomograph for Inhalation Exposure 44
Figure A-2 Nomograph for F1sh Ingestion Exposure 45
Figure A-3 Nomograph for Dermal Exposure 46
Figure A-4 Nomograph for Soil Ingestion Exposure 47
Figure A-5 Nomograph for Beef /Dairy Products Ingestion Exposure 48
-------
"OREWORD

The Expos-;re Assessment Group (EAG) of EPA's Office of Research and
Development has three main functions: 1) to conduct exposure assessments; 2)
to review assessments and related documents; and 3) to develop guidelines for
Agency exposure assessments. The activities under each of these functions are
supported by and respond to the needs of the various EPA program offices. In
relation to the third function. EAG sponsors projects aimed at developing or
refining techniques used in exposure assessments. This study is one of these
projects and was done for the Office of Solid Waste and Emergency Response.

Dioxin problems first surfaced in the U.S. in the early 1970's with Agent
Orange and the Missouri Horse Arenas. Since then dioxin contamination has been
found elsewhere 1n Missouri, Arkansas, Michigan, New York, and New Jersey. EPA
has become increasingly Involved in the discovery, assessment and clean-up of
these sites1. The purpose of this document 1s to provide an exposure and r^sk
estimation methodology for specific application to dioxin cohtamlnattbn sTtes.
This methodology will help us set priorities and make decisions required to
address this important problem.

James W. Falco, Director
Exposure Assessment Group
v1
-------
AtJSTKACT
This paper provides a methodology for estimating the human exposure and
cancer risk associated with 2,3,7,s-TCD0 contaminates soil. Five exposure
pathways are addressed: dust inhalation, fish inyestion, dermal absorption,
soil Inyestion, and beef/dairy products in^estion. For each pathway, factors
describing contact rate, absorption fraction, and exposure duration are presented
a Ion., with the equations for calculating exposure levels and associated cancer
risk. Tne methodology features tne use of nomographs to provide quick and
approximate estimates of risk. More detailed procedures are also provided for
more accurate estimates.
-------
ACKNOWLEDGEMENTS '
The author would like to than* the following incividuals who provided
valuable comments during tne peer review process:
Henry Falk (Center for Disease Control)
Peter Stern (tenter tor Disease Control)
George Fries'(Department of Agriculture)
Curtis Travis (Oak Ridge National Laboratory)
Dcodas Mukerjee (EHA • Environmental Criteria and Assessment office)
Michael Oourson (EPA - Environmental Criteria and Assessment -Office)
The supervision and technical advice from Charles Nauman and James Falco
were also jreotly appreciated.
-------
K1SK ANALYSIS JF TCJD CUf.TMttlNATLU SOIL*

The purpose ot this report is to present a procedure for estimating me
human exposure and health ris*s associated with 2,i,7,B-TLUU (referred to as
TCUD in remainder of report) contaminated soil. This report was prepared in
response to the mandate under the Jioxin Strateyy (t^A, lye3).
1.0 APPKJACH
Tnis report provides procedures for estimating tie human exposure and
cancer risk occurring to people living around a site where the scil Is contaminated
with TCDJ. Five exposure pathways are covered: dust innalation,-f ish--vrvjestion,
dermal absorption, soil inyestion ana beef /dairy products inyestion. A two
tiered approach is used in tne assessment procedure fcr each ot these pathways.
The first tier requires minimal data and uses a nomoyraph to facilitate the
calculations. It provides a quick and very approximate estimation of upper*
hound risk. The second tier requires more data and involve: more complex
calculations, but provides more realistic estimates of r1s<,
Tne procedures described in this report involve a number of important
limitations/assumptions. Exposure calculations require knowledge of the
contaminant level at the point ot exposure, i.e. contaminant level in air where
It Is breathed, or water where It is drunk, etc. Typically these values are
either measured directly or estimated using source release rates and fate/transport
models. The presentation of these techniques are beyond the scope of this
paper. Thus, these procedures assume the user can obtain this Information
Independently.
-1-
*An Interim version of tnis report was issued in March Iyb4. This edition
modifies and significantly expands tne first report.
-------
Additionally, this paper does not discuss the nealth effects associated with
TCDD nor the derivation of the cancer potency estimate. Instead, the papsr
emphasizes how to estimate human exposure ano merely presents the mechanics of
how to estimate cancer risk. References are provided for readers desiring
further background on the health effects and cancer potency "Stimates for dioxin.
1.1 Estimation of Cancer Risk
The general procedure for calculating cancer risk, as used thrcughout *.his
report, Is as follows:
Cancer Risk • 1 - exp {-potency factor x exposure) (1)
The cancer potency factor (or 9bX upper-limit of the linear slope factor)_for
TCOD 1s .156 (ng/kg/aay)-l. The derivation of this factor 1s described 1n EPA,
1984 and further background on TCDO carcinoger.icity Is provided in EPA, 1981.
Exposure has reciprocal units to the cancer potency factor or ng/kg day in this
case.
In order to use the above equation properly, it is important that the
potency factor and exposure handle absorption 1n a consistent fasnion. The /
exposure estimates presented 1n this report represent the amount, of contaminant
absorbed Into the body. The potency factor, however, was derived on the basis
of the administered dose (total fed to animals). Thus, an adjustment Is needed
to make these terns consistent. The potency factor was derived from a study
where the TCDO was administered to rats via their feed. Fries and >.arrow (1975)
report that 50-601 of TCOD In feed Is absorbed Into rats. Accordingly, the
potency based on administered dose must be multiplied by 1.7-2 to give an
absorbed dose potency. This adjustment makes the potency and exposure estirwtes
consistent and Is used In all risk calculations in this report.
-2-
-------
1.2 Estimation OT fiumaTi Exposure
In order to far?\itate cancer risk calculations, exposure as useo in tuis
report, is expresses as a daily cose rate averted over an Individual's lifetime
and bodyweight (typical units or? ns/ky-day):
lifetime TCDO Contact Exposure Absorption
Average * Concentration t Kate x Duration x Fraction (2)
Exposure rfofly weigfit * 7U yr liJeiii»e
The TCDl) concentration refers to the concentration of TCDU in the medium
of concern at -the p^int where exposure occurs. The medium of concern varies
accordinj to the exposure pathway, in air for dust inhalation, fish for fish
ingestion, etc. Although this equation rapresents the yeneral approach used in
this report, some refinements were made. For the Tier 1 calculations, a new
term called the conversion factor was introduced. This term is defined as:
Conversion Factor « TCUD concentration in medium of concern at exposure point (3)
TtuO cuncentrstion in soil at tne original source
The orioinal source, as used in Equation 3, refers to the original source of
contamination. The product of the TCUD concentration in soil at tne original source an
the conversion factor were substituted in Equation 2 for the TlUD concentration.
This factor represents the reductions in dioxin concentration as it moves away
from the source. It was introduced to facilitate the development of nomographs,
which relate tne TCUD concentration in the soil at the site to the resulting
cancer risk levels. The nomographs simplify the mechanics of these calculations
and help decision makers analyze tne potential risk caused by a site or the
level of site cleanup needed to achieve certain risk levels. This approach
requires estimation of the conversion factor which can be very difficult.
Basically, it involves either environmental mon1toriny or fate/transport modeling
or some combination of these. A detailed discussion of these procedures is
-3-
-------
beyonc the scope of tr.is report. However, a companion report is available

which provides guidance for now to estimate conversion factors (Uawson et al.
* - • • - -;• H M. •

1984). , . ",,

The Tier 2 calculations are based on refinements to Equation 'i which allow

modifications to reflect site specific conditions and account for the temporal
-.•••>• , • .,.-...
variability in certain parameters. For example, under some circumstances TCUU

4njsoi1 degrades, which. means the exposure,1evels diminish over time.
.",;%''.". . *•***'' ' ? * ' ' •
Additionally, the behavior patterns of the exposed population may sugyest

different contact rates or exposure durations than assumed in the fief 1

approach. In summary, the Tier 2 calculations are based on general equations

and make fewer apriori assumptions regarding parameter values. This allows

adjustment of any of the parameters to match site-specific conditions. The
k/:4-. •• :' ••''«&<:• '•- •
resulting modifications to Equation 2 make jt more complex but allow more

accurate and realistic estimations of risk. The Tier 2 equations differ slightly

for each exposure pathway. The details of how to apply the Tiers 1 and 2

methods are described 1n Sections 3-7 which cover each pathway separately.

*,.•*}-•"<"' - •
fZlu .^EXPOSURE RUUTES/4U •
• • •'• ^
TCOD has a very low water solubility, low vapor pressure and^strong tendency

to sorb on solids (see Table 1). Thus, any transport from the contaminated

areas will occur almost entirely in the solid phase.* Tnis report considers

only, transport by wind blown dust and suspended solids in run-off. Although
.«-l|lf ' * «... V*- "•"'"- * •
^'.suspended solidsscambe»carried via the ground water in highly! fractured or
i\r**K\- • -i .. *»*tejL - • ' a.t~'ier>*jr n . •'. . •'»iafcfc.-...«_
•Recent unpublished work by Freeman and Schroy (1984) suggests that volatilization
of TCDU in soil occurs rapidly. Since peer review and final publication of
this work has not yet occurred, vapor exposure is not addressed 1n this report.
-------
TABLE 1. PROPERTIES OF TCDD
Structure of 2,3,7,8-TCDD:

Molecular Height: 322
*Vapor Pressure: 10~6 mmHg (estimated)
Solubility 1n Water: 0.2 ug/1
Octanol-Water Partition Coefficient: 6.9 x 106 (calculated)
Source: Mabey,et£l_. 1981
An unpublished paper by Schroy et. al. (1984) reports the vapor pressure at
25eC as 1.5 * 10-9 mm Hg.
-5-
-------
porous strata, such areas are relatively uncommon. Additionally, tne presence
of a liquid , organic phase, which may occur at disposal sites, would enhance the
'• ~ •'.*,
of fish. Humans could also contact TCDi) while swimming in contaminates .....
waters. This route is considered minor since swimming is generally a
relatively infrequent activity and contact with or ingestion of tne
•'!.- v.
^ *J». . '•*;»•>• •
sediment is minimal.
o Deposition of dust or eroded soil on residential areas, direct human
contact and dermal absorption or ingestion. Although, TCUD is tightly
<*'*•-''' '•* 'id'*
bound to soil, studies have shown tnat dermal absorption can occur
(Poiger and Schlatter, 1980). Dermal contact with soil could result
during outdoor recreation or gardening and yard work. Soil ingestion
^ can Joccur particularly amongpyoung children with mouthing tendencies.
Home grown vegetables could also become contaminated with TCUU, but this
contamination is diminished by several factors. TCUU is generally not
\*J*,:. ~"-: : .'" . .
taken up significantly in plants* and vegetables are, typically washed to
• "?T ." . 4- ' ~*
remove deposited dust before consumption. Additionally, except in dry
and disturbed areas, relatively little dust transport occurs.
"borne investigators (Cocucci et al.'1 ly/yj have found evidenced low levels or
TCDD plant uptake. However, others (Wipf et al. 1982) could not detect any
uptake.
-6-
-------
o deposition of dust or eroded soil on pastures, accumulation in cattle
and ingestion of dairy products or beef. Although this route could also
apply to other kinds of livestock, the cattle route is considered most
significant since people yenerally consume more cattle products than
other kinds of animal foods and cattle typically graze outdoors wnere
the potential for contact with contaminated soil is greatest.
Based on the above, this paper has focused on oust inhalation, fish
invest ion, dermal absorption, soil inyestion, and beef/dairy products investion.
The exposure'scenarios assumed under the Tier 1 caIculatlons,for each.pathway
are summarized below:
o Dust inhalation, soil ingestion and dermal absorption — The exposure
associated with all three of these pathways 1s assumed to occur in a
residential setting. The soil around the residence, indoor dust
deposits, outdoor suspended dust and indoor suspended dust are all
assumed to be equally contaminated. Thus the exposure is assumed to
occur Indoors as well as outdoors. People are assumed to live in this
situation for an entire 7u year life. Some adjustments are made for
climatic considerations (i.e., frozen soil) which could restrict dust
movement or soil contact and associated exposures. Also, soil ingestion
1s assumed to only occur during ages 2-6 when mouthing tendencies and
lack of personal hygiene understanding are highest.
o Fish inyestion — For this pathway, 1t is assumed that a person receives
his entire freshwater fish diet from a contaminated source over a 70
year life. This scenario would probably Involve a person who lived
near a contaminated water body and fished for subsistence purposes.
-7-
-------
o Beef/dairy products ingestion — Under this pathway it is assumed that
a person receives his entire beef and milk diet from a contaminated
source over a 70 year life. .This situation would probably Involve a *
farmer whose-fields were contaminated and derived his beef and milk
from livestock, raised on his property. Home slaughter is common at many
commercial ranches, so,this scenario-could include more than just
subsistence level farmers.
These are obviously worst-case assumptions. Many factors such as behavior
patterns, 'climatic conditions, source size, and remeaial_ac_t1pns._cpuldjLl;l -
reduce potential exposure levels. Such factors can only be considered on a
site-specific basis and, therefore, cannot be considered under the generic Tier
1 calculations. However, the Tier 2 calculations are designed to allow
considertion of such site-specific conditions and should be used to refine the
Tier 1 estimates.
The factors affecting each .exposure route are discussed and computational
techniques are presented in Sections 3-7. The remainder of this section discusses
Issues common to all of the exposure routes.

.2.1 Body Weight
The exposure calculation for each route requires making a body weight
assumption tor substitution Into Equation 2. The body weight selected should
reflect£the weight 'of|the exposed*1ndiv1dual(s);during the period whicHHhey
^;' l^v>* .' . •' • *' :*:-- ..'.-. V
are exposed. The Tier 1 procedures make the apriori assumption that the weight
of an average male or 70 kg (Snyder et al. 197b) will generally reflect the actual
exposure^condit1onS|for all routes except soii|ingestion.^|$1nce soil ingestion
Is assumed to be significant during only ages 2-6 the average weight for these
-8-
-------
ages ot 17 kg (Snyder et al. 197b) was used.
The Tier 2 procedures allow the user to maice his own determination of tne
most appropriate booy weignt. This decision should be based on the age of the
exposed population over the exposure period, body weight is related to age as
follows:
Age (yr) Body wciyht (kg)
0-la 3.14 kg * (3.i>2 kg/yr x aye)
>1« 7U ky
These relationships describe average male weight and were derived via a regression
analysis (Figure 1) en data presented by Snyder et al.'(197SJ.^ After"identifying
the ages of exposure, the analyst should integrate the weight over the appropriate
ages and divide by the exposure period. This value will best represent the average
weight to use in the exposure calculation. For example, if the exposure occurs
over a 20 year period when the individual is aged 8-28 the average weight would
be 69 kg:
Average
Weight
Ib
V 3.14 + 3.
52 x dx + (28-18) (7U)
120
2.2 Lifetime
The exposure calculation for each route also requires making a lifetime
assumption tor substitution Into Equation 2. For compataoility with the dose-
response estimates dervied from animal studies, this value should represent the
total lifetime of the exposed individual. Accordingly, In the Tier 1 calculations
it 1s always assumed equal to 7U yr which represents an average U.S. male. It
1s also recommended for use in the Tier 2 calculations unless site specific
-y-
-------
70
>HT
10
10 20 30 40 50 60 70
AGE (YR.)
Figure 1. Age vs Body Weight
-------
data on the exposed population suggests a different average lifetime.

2.3 Degradation of TCDD in Soil
The exposure calculations for each route require making a TCUO concentration
assumption for substitution Into Equation 2. Altnougn this concentration
represents the medium of concern at the point of exposure, it is directly
dependent on the concentration of TCUD in soil at the original site. If the
concentration at the site is chanyiny due to degradation, the concentration at
the point of contact will change as well. Accordingly it must be considered
when estimating exposure.
The degradation of TCDO in soil 1s difficult to measure. Most investigators
have found that it 1s generally resistant to biological and chemical degradation,
but susceptible to photolytlc degradation (EHA, 1984). Young (1963) measured
the half-life of TCDD 1n soil as 10-12 yr and attributes most of the degradation
to photodecomposition. This study has adopted this value as a lower limit in the
Tier 1 calculations, since 1t assumes that the TCDD 1s located at or near the
surface and consequently at least partially exposed to sun light.
However, Young states that physical mechanisms such as wind or water
erosion could also account for the observed losses. Given this uncertainty and
fort that much of the TCDD in soil may not be exposed to sun light, it appears
that under some conditions essentially no degradation would occur over tne time
frame of Interest, I.e., 71) yr. Thus, for Tier 1 the half-life 1s assumed to ranje
from 1U years to Infinity. Although, this range appears very wide, actually as
the half-life Increases over 100 years it has very little impact on the final risk
estimate as demonstrated in Figure 2. This figure represents how risk changes
when only the half-life 1s changed, I.e., all other parameters held constant.
-11-
-------
LOS OF
CANCER
RISK
1 2 3 : 4 5

LOG OF HALF LIFE OF DIOXIN IN SOIL t(YR.)

Figure 2. ' Risk vs Half-Life
-12-
-------
Figure 2 also demonstrates that the risk estimate is sensitive to the half-life
choices under 1UO yr.
Under Tier 2 the analyst should choose a naif-life most representative of
the site. If monitoriny data is available, they may show trends which can be
used to estimate degradation rates. Alternatively it is recommended that a
half-life near 10 years be chosen if the TCUU contamination is at or near the
surface and over 100 years if the TCDO is buried deeply (per the previous
discussion). Once a half-life has been selected the exposure is calculated
under the assumption *hat the concentration will vary according to first order
kinetics:
4£ kC
dt « (4)
where, O concentration
t» time
k» degradation rate constant
« loge
T1/2» half-life
The concentration at any point in time is calculated Dy solving Equation 4:
C • Co e-*t (b)
where, Co » initial concentration
Using Equation (5) the exposure can be calculated by Integrating C over the
exposure time and substituting Into Equation (2). Alternatively, exposure can
be estimated by solving for C at f recent Intervals, computing exposure, and
summing exposure values. Simple calculator programs should be used to conduct
such calculations.
-13-
-------
The effects of degradation on exposure can be determined as shown below:

Degradation Exposure , ,)Cdt ., jl-e"**)
Non-degradation Exposure Cot let (6)

In the Tier 1 calculations the exposure is multiplied by this ratio to reflect

the effects of degradation. The ratio will always have an upper limit of 1

when it is assumed that degradation will not occur (i.e., half-life equals

Infinity). The lower limit if calculated using a'10 yr half life and the upper

end of the exposure duration assumption which provides the maximum degradation.
.'.'•' '. ' * \
The above discussion assumes that degradation will occur according to

first order kinetics. Although, this assumption 1s commonly applied to these

types of problems, 1t is generally recognized as an over simplification of a

very complex problem. Recent unpublished work by Freeman and Schroy (1984)

suggests that TCOO degradation in soil follows much more complex kinetics due

largely to relatively rapid and strongly temperature dependent volatilization.

Since peer review and final publication of the work has not yet occurred, no
• *i < ,. ' • •''''"
final conclusions can be drawn. However, 1tf1s potentially very Important in

two respects:

o It may mean that a first order kinetics approach to this problem 1s

"^Inappropriate.

o It may mean that potential exposure periods are much shorter than

previously thought.

f*v "
2.4 Exposed Populations
, r . * ' .* •' • :-'i v ' .: , ' ' " * "
The monitoring data or modeling results will probably show that TCDD

concentration 1n the environment.diminishes with distance.from the original
-14-
-------
contamination source. Similarly, the exposure and risk levels will diminisn
with distance. Tne exposure and risk estimates can be plotted on a map and
isopleths constructed. These .lines snow areas with equal exposure or risk
levels and can be used to identify how many people are exposed at various
levels. In sparsely populated areas, U.S. ideographical Survey maps shoulo r>e
used since tney show individual buildings. Local officials should be consulted
to determine how many people are associated witn. such buildings, otherwise an
average of 3.8 persons/dwelling should be assumed. In more dense areas, the
best population statistics are available from the Bureau of Census. Population
estimates for counties and smaller areas are provided by the bureau at (202)
763-b002.

Z.b Hiss from Combined Exposure rtoutes
The procedures described in this study explain how to calculate the risk
associated with individual exposure routes. In situations where an individual
is exposed to TCUi) by more than one pathway, the risks should be calculated
separately and then added.

3.0 DUST INHALATION
Dust is generated from land surfaces as a result of mechanical disturbances
(I.e., vehicle traffic) or wind erosion and dispersed via the wind. Cowherd et al.
(1984) have recently completed a manual specifically for the purpose of estimating
dust emission rates from contaminated land surfaces and resulting air concentration:
around the source. If sufficient monitoring data is unavailable, H Is highly
recommended that analysts consult this manual to model dust emissions.
TCDD, as discussed earlier, is typically tightly bound to soil particles.
-Ib-
-------
The organic carbon content and surface area of tne particles affect now mucn
•f
TCOU absorbs to particles. Since these'factors may vary between the source
soils and dust generated from tnern, the TCUU-levels may also differ. Unrortunaiei
the '•ntluence of these factors are generally not Known
-------
3.2 Exposure Duration
Oust generation and the resulting exposure is essentially eliminated when
the soil is very wet or frozen. Obviously, such conditions will vary widely
across the country. In warm arid areas such as the Southwest, the conditions
preventing dust emissions almost never occur. Whereas, in Minneapolis, the
soil Is frozen a.i average of lib days/yr (Personal communication from Don Baker,
Minnesota State Cllmatoligist, April 2, 19B4). Cowherd, et al. (1984) suggests
that dust emissions are negligible on days when precipitation exceeds .1)1 inches
which 1s reported by Cowherd as 110 aays/yr for Minneapolis. NUAA (1980)
suggests that approximately 80X of the precipitation days occur outside of the
winter months. Thus, Minneapolis has a total of 206 days/yr (118 + 80S of 110)
when the soil conditions would prevent dust emissions. For purposes of tne
Tier 1 calculations it was assumed that the arid Southwest and Minneapolis
would represent the possible range of conditions. Accordingly, under Tier 1
the exposure duration was assumed to vary from Ib9 to 365 oays/yr or a total of
ll,13U-2b,J>bO days over a 70 yr life.
Under Tier 2, the user should adjust the exposure duration to reflect the
climatic conditions of the site. Such adjustments should only be made if they
are not accounted for elsewhere. The models presented by Cowherd et. al.
(1984) adjust the emission rate estimates on tne basis of climatic conditions.
If such models are used it would be redundant to also adjust tne exposure
duration on the same basis. The behavior patterns of the exposed population
can also effect the exposure duration and should be adjusted accordingly, if the
appropriate data is available. Finally, a mass balance should be conducted to
ensure that mass of contaminant emitted does not exceed the amount present.
-17-
-------
3.3 Absorption Fraction
"*' •*?* • •*- ? . *
For this route the absorption fraction is the fraction of the contaminant

entering the lungs which is absorbed into the body, the fraction of particles

which are inspired (i.e. enter the respiratory system) depend on numerous
••-•&..
factors such as breathing rate, particle size distribution, wind speed and

whether breathinc, is done through the mouth or nose. The Internationa) Standards

Organization'(1981) has estimated the inspired fraction as a function of particle

size under'average conditions '{Figure 3)'. ""Particle sizes' <1U u are generally

considered most important In estimating health'effects. Virtually all of these

particles will be inspired. However, their fate after entering the lungs is
•j£.. ••**_. .f, • •••
less certain. Generally, the heavier particles deposit in the upper regions of

the respiratory tract, the lighter particles in the lower regions and the very

lightest are exhaled. Most of the deposited particles in the upper regions and

some in th? lower region are'clearred by ciliary action and swallowed. ' Lackinj

specific particle size distribution information the fate of inspired particles

should be assumed to follow the recommendations of the International Commission

on Radiological Protection (Table 2). Since TCDD has a Tow water solubility,

the recommendations for "other compounds" would apply.

After determining how much of the oarticles are swallowed, the overall

absorption fraction~can be furthe> refined on the' bas1s'"of'"GI tracFIFsorption.

Poiger and Schlatter (1980) found that 13.8 - 18.2% of the orally administered

TCDD (which had been absorbed to soil for 8 davs) reached the liver in 24 hr.

Assurning*that this'represents TOt^of the body burden (FHes and Harrow^ 197s) -'

the total (»I tract absorption is 20-261. McConnell, et al. (1984) also found

that the absorption of TCOD from soil In the GI tract was "highly efficient"
^, lrv

-18-
-------
FroCtiOA Net

Now or Moult. >t . Not
1.0 10
Aerodynamic Diameter -
The Alveolar Fraction represents the particles reaching the alveoli
(ie. deepest region of lungs). The Tracheobronchial Fraction
represents the particles reaching the tracheobronchial system (ie.
central region of lungs). The Extra Thoracic Fraction represents the
particles reaching the area outside the thorax (ie. nose and throat).
Figure 3. Inspired Fraction vs. Particle Size

Source: International Standards Organization. 1981.
•19-
-------
Table 2. Distribution of Inspired Particles
Readily soluble Other
compounds compounds]
U) * '• (I) -
Exhaled

Deposited in upper respiratory
passages and subsequently
swa11 owed

Deposited in the lunys (lower
respi ratory,passages) >j
2b
(this is taken up
Into the body)
bU
* Of this, half is eliminated from the lunys and swallowed in the first 24 hr,

making a total of 62.t>X swallowed. The remaining I2.bt is retained in the

lungs with a half-life of 120 days, it being assumed that this portion Is taken

UD into the body fluids.
Source: International Commission on Radiological Protection, 1968.
-------
in test animals suggesting tnat at least this much is absorbed.
In summary, the overall absorption fraction is calculated as follows:
Absorption
Fraction
Inspired j Fraction Remaining /Fraction GI Tract J
* Fraction • in Lungs * ^Swallowed * Absorption Fraction^'! (9)
Using Equation 9, an absorption fraction of .2b-.29 was derived from the
following assumptions:
Inspired Fraction « 1.0 (on basis that <10 u particles are primary concern)
Fraction Remaining 1n Lungs » .12i> (ICHP, 1968)
Fraction Swallowed • .62$ (1CKP. 1968)
til Tract Absorption » .itO-,26 (Polger and Schlatter, 1980)
This value is also recommended for Tier 2 calculations unless site-specific
data suggests otherwise.

4.0 FISH INliESTlON
Fisn contamination results from the transport of eroded soil via runoff
trom the TCOO contaminated site to local surface waters. The contaminated soil
mixes with the other sediment in the water bodj lowering tne effective TCUO
concentration. In a river or stream the TCOD moves downstream with the sediment
and the concentration of TCUO decreases further due to dilution with clean
sediment. This process 1s very complex and highly site specific. Once TCUJ
has entered a water body, it has been shown to bioconcentrate In aquatic species.
The nomograph for the Tier 1 calculations Is presented 1n Figure A-2. The
conversion factor -used for this exposure route is defined as:
Conversion Factor * TCOD concentration in Sediment Where Fish are Caught (nq/g)
TCUU Concentration in Soil at Original Source (ng/g) (10)
-21-
-------
4.1 Bioaccumulation
*

The approach taken in this study assumes that equilibrium condition* have
»

been reached and remain constant over the exposure period. This me*.is that the

TCDl) levels in the sediment remain constant at a particular point and.,that the
.. *

fish have reached an equilibrium with the environment. This further implies

that the levels of TCOD in fish from a certain area will remain constant over

time and at a-constant relationship to the TCDU level in the sediment.--. In

reality, some species such as bottom feeders will move toward equilibrium

conditions faster than others, many species may never reach equiMbriunrdue"to'
"j ... •" • ^
the'fact that they do not" spend enough time in one location, and finally some

species will bioconcentrate more TCOD than others due to greater lipid content.

For Tier 1 the ratio of TCDU in fish to TCOU in sediment is assumed to range
£ *, " " ;< f - , ' . • T

from 1 to 10 as reported by Kenaga and Morris (1983).

For Tier 2, the analyst should attempt to find data on the fish species

and,*water body -conditions that best reflect the,,site being analyzed.

4.2 Consumption Rate

, For the Tier 1 calculations the consumption rate is assumed equal to 6.t>

g/day which 1s the U.S. averge for fresh water fish (Stephen, 1S8U). The

average may be higher in areas near large fish supplies, such as the Great

Lskes .region.t|Additionalj1y,-the average consumption rate^ajiong fish eaters

will be higher than the overall average. However, we generally lack the

necessary data to reflect these phenomena. Unless site-specific data is

available it isy.recommended that 6.5fg/day be used in the iller 2 calculations

as well.
•22-
-------
4.3 Exposure Duration
For this pathway the exposure duration represents the number of days that
the exposed population eats contaminated fish. For the Tier 1 calculations it
1s conservatively assumed that this will occur every day of an entire 70 year
life or 25,550 days. In reality this value is probably much less due to several
factors: -
o Few individuals receive their entire fish diet from fish cauyht in one
location..
o Seasonal conditions may prevent catching fish from the.contanii.natfid
area.
»
o The contamination source may not last ?U years. A mass balance should
be conducted to determine how long it could last.
«
The analyst should attempt to consider these factors in selecting an
exposure duration for use in the Tier 2 calculations.

4.4 Absorption Fraction
For the Tier 1 calculations the fraction of TCDD in fish which is absorbed
in the GI tract was assumed to range from .b to .86. This was based on the
following two studies as reported by McConnell et al. (1984):
o 50-601 of TCDJ in diet of rats was absorbed.
o 86% of TCOl) in a mixture of acetone and corn oil fed by aavaye to
rats was absorbed.
Use of these data assumes that absorption from TCDD in fish will be similar
to absorption from TCDD in r£t food and acetone/corn oil mixture.
This range Is recommended for use in Tier 2 calculations as well unless.
more relevant data becomes available.
-23-
-------
5.0 DERMAL ABSORPTION
Deposition of contaminated dust or eroded soil in residential areas can
*.
cause human exposure by direct contact and dermal absorption.
The conversion factor used in the Tier 1- calculations for this route is
defined as:
-' ' ?
*•'#' -:i * '•'..' ,'" • •• **
does "not account for all of the above factors. Thus, althouyh it'represents
our current best estimate, much uncertainty remains.
The amount of soil which accumulates on skin was estimated from studies by
* . *
Lepow (1975) and Roels (1980K . •
Lepow (197b) found that children accumulated at least 11 my of soil on
their hands after normal olavinu in and around their,residences. This estimate
was'based on soil" samples collected by pTessing a 2115 cm? tape:against tne
hands of the children. Obviously, this method is not lUUt efficient and Lepow
Indicated that the samples collected represented only a small fraction of the
totaf soil on thUir hands. "For purposes^bf this analysis it was-assumed that
this measurement represented a lower bound estimate for the amount of soil on a
-24-
r
-------
21.5 cm2 area. Thus, the average soil level was 11 mg/21.4 cm2 or 0.5 mg/cm2.
Roels et al. (1980) measured the amounts of Ho on the hands of children by
rlnslny the palm and finyers of one hand with dilute nitric add. The levels
of Pb in the soil and dust where the children lived are also measured. Uslny
these data the amount of soil on the hands was calculated as follows:
Amount of soil Amount of Pb/hand (12)
per unit area * (Concentration ot PD in Soil) (Surface Area of Hand)
the surface area of the hand was determined by Snyder (197t>) who sugyests that
the palm and fingers of one hand comprise 1% of the total body surface area.
Snyder (1975) also gives the total surface area of 11 year old children (average
age studied by Roels) as 10165 cm?. Thus, the area of the palm and fingers was .
assumed equal to 102 cm2. Substituting this value and the Rod's data into the
above formula the amount of soil on the skin was calculated to be 1.5 mg/cm2.
This level was assumed to represent an upper estimate producing an overall
daily contact range of 0.5 to l.b mg/cm2. Additionally, 1t was assumed that
this range represents an average value for the entire exposed area of the body.
Normally hands are probably dirtier than other parts of the body, but the fact
that neither of the hand measurement techniques are 1001 efficient makes it a
more reasonable estimate for the average value ot the entire exposed area. It
was further assumed that this range applies to adults working outdoors as well
as children. Unless other data are available, 1t 1s recommended that this
range be used for Tier 2 as well as Tier 1 calculations. Since this contact
rate 1s expressed in per unit area terms, it must be used in conjunction with
estimates of the exposed surface area, which is discussed in Section b.3.
-25-
-------
b.2 Exposure Duration

For this route, the exposure duration represents the number of days that'

an individual will contact the contaminated soil. In a residential setting
1. , , •., • .'
behavior patterns and seasonal conditions will most influence this parameter.^
•\
' ¥ ,
Children who enjoy playing outdoors and adults who enjoy gardening or other

types of yard work could contact soiV very frequently. In warm climates, such

people could contact soil every day." In the coldest parts of the U.S. such as

Minneapolis, the soil is,frozen an average of 11B days/yr (Personal Communication

from Don Baker, Minnesota State Climatologist, April 2, 1984). AithouyhBother
jit'' 1-t> .-. .-i '' '•••• ~ -. -;
'^^ • *• c" ' - '' - '-,¥•' '^-'
types of inclement weather. Illness, travel and other factors could reduce the

duration period, no data could be found clearly connecting these phenomena to

the potential for soil^contact. Accordingly, the range of 247-365jlays/yr was^
.-sf' •• - ' • ^ • ' • -4

adopted for the Tier 1 calculations. Under Tier 2, the analyst should attempt

to find site specific data for refining this number. The exposure duration can
> ^ t ••* • • ,
also be affected by the-source size which should be analyzed via a*mass balance.

5.3 Exposed Surface Area

/*"•> ^. . ' •;• . •*. ' .;••-• ' ' ' '•,. ..
**¥_ The exposed surface area of an adult has been estimated by Sendroy (19b4)
*; . - " '••
as:
•-**• f 1 J 1 , • ' ,"f , * • , . ^. i ----' ."u, .
o 2940 cm*? • wearino short-sleeved, open-necked shirts, parts, shoes,
m. .- •

-t. wun no gioves or nats.

o 9lu cm^ - wearing lony-sleeved shirts, gloves, pants and shoes.

£^. The exposed surface4area of children wal, computed;by multiplyjng the adult

values by the ratio of a child's total surface area to an adult's total surface
-26-
-------
area.
Based on the above assumptions, the total amount of soil whlcn accumulates
on the exposed area of people was computed as follows:
Total Accumulated Soil « (Contact rate)(exposed surface area)(exposure duration)
Since the surface area changes with age, this calculation has to be made for
each year and summed over a lifetime. As shown in Table 3, this approach yields
an estimate of the potential lifetime soil accumulation of 7,900 - 110,ODD g
which was adopted for the Tier 1 calculations. The exposure duration assumption
will probably make this estimate unrealistically high in most situations.
Therefore it is strongly recommended that users attempt to find site specjtic
data to refine this estimate under the Tier 2 calculations.
5.4 Absorption
Poiger and Schlatter (19bO) found that O.Ob to 2.21 of the TCOt) in a soil
paste applied dermally to laboratory animals for 24 hr reached the liver.
Polger and Schlatter also reported that other Investigators found about 7U1 of
the total TCDD body burden in the liver. This suggests that the total absorption
actually varied from u.07 • 34. This ra?Qe was selected for use in the Tier 1
calculations. Unless other data is available it is recommended for use in the
Tier 2 calculations as well. Such extrapolations from animals to humans
Introduces uncertainty due to differences in skin properties.
6.0 SOIL INbESTION
Deposition of contaminated dust or eroded soil in residential areas can
also cause human exosure by ingestion. Although soil ingestion occurs throughout
-27-
-------
TABLE 3. AMOUNT OF SOIL ACCUMULATED DURING LIFETIME
Age
2-3
4-5
6-7
8-9
9-10
11-12
12-13
13-14
15-16
16-17
17-18
18-19
19-20
20-22
22-24
24-70

Exposure,
Duration*^
(days)^
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
494-730
741-1095?
741-1095*
11609-17155
„.
Total
Surface-Area*
(cm2)
5800
7200
8100
8900
9300
10300
11100
11700
13800
14800
15500
16100
16600
17000
17400
17400

;'* Surface Area
Child/Adult
0.33
0.41
0.47
0.51
0.53
0.59
0.64
* 0.67
0.79
0.85
0.89
0.93
0.95 u
0.98
1.0
1.0
•
Exposed Surface Area-*
(cm2)
300 - 980
380 - 1200
420 -1400
470 - 1500
490 - 1600
540 - 1700
580 - 1900
610 - 2000
720 - 2300
770 - 2500
810 - 2600
840 - 2700
870 - 2ROO
890 - 2900
910 - 2900
910; - 2900
t
I
i
Total Accumulated Soil
(9)
74 - 1100
94 - 1300
100 - 1500
120 - 1600
120 - 1800
130 - 1900
140 - 2100
150 - 2200
180 - 2500
190 - 2700
200 - 2800 '
210 - 3000
210 - 3100
330 - 4800
340 - 4800
5300 - 75000

Total
1. Exposure Duration
2. Snyder, 1975.
3. Lower Estimate)
247-365 day/yr x years of exposure.
7900 -
-------
a person's life, it will be most significant during cnilahood. For this reason
and lack of data on how much soil ingestion occurs among aaults, this stucy
only estimates exposure to children.
The conversion factor used in the Tier 1 calculations is identical to tnat
used for tne dermal absorption route:
Conversion Factor « TCDU concentration in soil at exposure point (no/o) (li)
TWu concentration in soil at original source (ny/^)
The nomograph is presented in Figure A-4.

6.1 Contact Kate
The amount of TCDO-contaminated soil which children may ingest as a result
of normal playing around their home is very difficult to estimate.—The~1ngestion
rates will depend on the mouthing and pica tendencies of the children.
Based on measurements of the amount of soil found on children's hands and
observations of mouthing frequencies, Lepow (197b) estimated that children
could ingest at least 100 my of soil per day. This estimate does not account
for direct ingestion of soil which could increase daily ingestion rates tc b
g/day (personal communication from Julian Chisolm, Baltimore City Hospital,
November 1V82). This range was selected for use in the Tier 1 calculations.
Unless site specific data is available, it is recommended that tnis range
be applied in Tier 2 calculations as well.

6.2 Exposure Duration
For this pathway, the exposure duration represents the number of days that
a child consumes contaminated soil. Ubviously this number can vary tremendously
depending on individual behavior patterns, access to contaminated areas, soil
-29-
-------
conditions, etc.
The children studied by Lepow ranged from 2-0 yr old. Lacking other oata,
It was assumed for purpose of the Tier 1 calculations that this represents the
ages that mouthing tendencies and lack of understanding of personal hygiene
• •• ' >" * -
will cause the most significant soil ingestion. As with the dermal absorption
pathway, it was assumed that the soil could be unfrozen from 247-3bb oays/yr
depending on location. Although other types of inclement weather, illness,
„* .
travel and other factors could reduce and potential duration period, no data
couH be found ret'lectiny such phenomena. Accordingly for the Tier 1 calculations,
the exposure duration was assumed to .last 247-36!> days/yr from aaes 2-6 tor a
tii " ' " " **'•'' '' •"'"
total of 1240-1830 oays.
This assumption probably represent: a severe worst-case for most situations.
For the Tier 2 calculations it Is strongly recommended that the analyst,attempt
.p ..,*;•• V'V '
to find site specific data leading to more realistic estimates. Amass balance
should be conducted to confirm that the source emissions can last at least b years.
6.3 Absorption
"dr-. •
The 61 tract absorption of TCUD in soil has already been discussed under
Section 3.3. In summary, an absorption fraction of .2U - .26 (Poiger and
Scnlatter, 198U) was used..for the Tier,,! calculations and is also recommended
„ ,^
for the Tier 2 calculations.

7.0 BEEF/DAIRr PKOOUCTS INGESTIUN
The deposition of contaminated dust or eroded soil on pasturelands can
lead to uptake in the human food chain and eventual human exposure. Tne
consumption of^soil by cattle has been^measured to^average .72 kg/day (Fries,
Thus?r1f the soil"is contaminated, beef and milk can also become
-30-
-------
Conversion Factor « TCDD Concentration in Pasture Soil (ng/g) (14)
TCDD Concentration in Soil at
contaminated. This process can occur relatively quickly. Fries (1982) reports
that PCB levels in milk reached steady state three weeks after it was introduced
Into the diet.
The conversion factor for the Tier 1 calculations is defined as:
Soil (ng/g)
Original Source (ng/g)
The nomograph used to facilitate the Tier 1 calculations is illustrated in
Figure A-5;
As explained below, the assumptions made for this exposure pathway reflect
a situation where a person would obtain his entire beef and milk diet from
livestock raised on his property. This is obviously a worst case situation and
users should understand that commercial marketing practices would reduce such
exposures for most people (discussed below).
7.1 BJoaccumulation
A number of studies have been conducted on chemicals similiar to TCDD such
as PCB, PBB, and DDT, which relate the level of the contaminant in the diet to
the resulting level 1n body fat or milk fat. Fries (1982) reports that these
compounds reach an upper estimate, steady state fat/diet ratio of approximately
5. Jensen et al. (1981) conducted similar studies using 2,3,7,8-TCDD and found
the steady state fat/diet ratio to be approximately 4, which suggests that 1CDD
behaves similarly to PCB, DDT, and PBB. Using a fat/diet ratio of 5 and data
regarding the soil content of the diet. Fries estimates the milk fat/soil
raMo as .7 and the tissue/soil ratio as .23. These estimates are based on
data from New Zealand where animals are typically grazed throughout the year.
-31-
-------
In the U.S. grazing is normally less frequent and supplemental feeds are commonly
' _ ., ... ,v4.. • , .•.'*• i.- • '
used. Such feeding practices could alter tne amount of soil consumption.

Fries and Jacobs (1983) conducted another study where cattle were kept In a

feed lot situation containing PBB contaminated soil. Under these conditions,
. i

the beef fat/soil ratio averaged .3y and milk fat/soil ratio averageo .40.
, < .. ' -'•• 4 :*•>•»"•' - •:

Since these conditions more typically represent conditions in the U.S., they

were selected for use in this study. However, the .extrapolation of these }.

results to this assessment Involves several important assumptions:

o TCDD wil) be metabolized in a similiar fashion to PB8.

o The"portion of soil in the diet is the same in the exposure scenario

as the study.

o Cattle will ingest more TCDO from soil on the ground than from

foliage deposits. This is the situation occurring during the

experiment from which Fries and Jacobs (1963) derived the .4

fat/soil ratio. Where run-off represents the dominant transport

route, most of the TCDO will be on the ground rather than on

foliage and this assumption should be valid. However, in dry and
j * - ( ' f i' i .*••'.-
disturbed areas/significant dust transport may occur causing
• f- •

cattle to obtain more TCDD from foliage deposits than ground

deposits (personal communication from Curtis Travis, Oak Ridge

National Laboratory, Feb.iZ, 1984).-Since, the bibavailability
5*~- " ''•*•"'- , ^
* . '
of TCDD in soil may differ from that in foliage deposits, the
. • s -, .. •• • .'.' '.
fat/soil ratio may differ.

4n summary,fa'fat to soil ratio of^4 was adopted for useTn the Tier 1
•32-
-------
calculations and is also recommended for use in the Tier 2 calculations.
7.2 Consumption Rates
Average beef ana milk fat consumption rates and fat content data are
presented in Table 4.
Thivs data suggests an average of 62 g/person-day beef and milk fat are
consumed. This value was adopted in the Tier 1 calculations and is recommended
for Tier 2 as well, unless site specific data suggest otherwise.

7.3 Exposure Duration
The exposure duration for this route refers to the number of days an
Individual will consume contaminated beef or dairy products. This value can
vary tremendously depending on how the contaminated food is distributed. Some
Individuals may derive all of the beef and milk from the same source which means
their exposure duration could potentially last every day of a lifetime. However,
most people obtain beef and milk commercially. The production and marketing
practices of commercial food operations can greatly reduce exposure durations.
For example, milk from a number of dairies may be collected in one truck.
Assuming only one dairy is contaminated, the contaminant level is diluted ai a
result of mixing with uncontaminated milk. Further mixing and dilution may
occur at the processing and bottling plant. The milk from one plant probably
represents a small portion of the total local market. Thus, an Individual is
unlikely to buy only milk which was contaminated. Accordingly, that individual's
exposure is much less .nan that suggested by his total milk consumption. Thus,
the "dilution" effects of production and marketing reduce individual exposure
-33-
-------
TABLE 4. BEEF/DAIRY PKUDUCTS INGESTION KATES-

Total
Consumption Kate Percentage Fat Consumption Kate
(g/person-day) Fat (g/person-day)

Beef 124 Ib 19

Dairy Products 55U 7.8 43

^.. , _-„„. . - _ . .. .„

Total 62

*EPAt-1981
-34-
-------
levels. This reduction is best estimated using local data. However, the
potential dilution is illustrated by following data for the Mid Atlantic Region
(Personal Communication from John Buche, Statistical Reporting Service, U.S.
Department of Agriculture, Beltsville, MD, November 7, 19fa4):
Une month's production of Class I rail* 2.2 x 1UB ID
One month's average production per producer 3.2 x 104 Ib
These figures suggest a potential "dilution" in exposure level of 6.9UU times.
Obviously, these numbers represent averages over a large area and different
statistics may apply to Individual markets. Further study of the-dairy"arfd
other food Industries 1s needed to more accurately predict the dilution effects
caused by production and marketing practices.
These dilution effects can be accounted for by either lowering the assumed
contamination levels in food products or reducing the effective duration of
exposure to fully contaminated products. For milk 1t 1s probably more logical
to reduce the contamination levels and assume milk 1s consumed every day. For
beef, It would be better to reduce the duration estimate since the marketing
practices will reduce the number of days contaminated beef 1s eaten rather than
the level in the beef.
Unfortunately, we lack the data to characterize these effects and have
Ignored them in the Tier 1 calculations where it was assumed that the exposure
could occur every day of a 7U yr life or 2t>,5SU days. However, the analyst
could attempt to consider dilution effects in the Tier 2 calculations, since
i
they appear to have potentially very significant Impacts on exposure levels.
Other site specific factors such as the source size and accessab111ty could
also reduce the exposure duration.
-35-
-------
7.4 Absorption

* As discussed in Section 4.4, the til tract absorption used in Tier 1 was

assumed to vary from .b to .86 on-basis of studies reported by McConnel et al.
* '
(1984). Some uncertainty is introduced by the fact that these studies used rat
**-•-' . . •
feed and a mixture of corn oil and acetone rather than beef and milk. Unless

better data become available later, it is recommended that this range be

assumed for the Tier 2 calculations as well.
* '' *

8.0 OISCUSS1UN OF UNCERTAINTY

Users of the methodology des.. >oed in this report should understand that
» < * t. *••
. .', T ' i V ' "..- "'.•
it Involves considerable uncertainty. The uncertainty is derived from the

numerous assumptions which may not accurately reflect actual conditions:

o The assumptions regarding body weight, lifetime, and contact rates were
•„ 4K - >;'
based on national averages and may not be representative for specific

Individuals*

o The absorption data were derived from animal studies and assumed,.
••• -•" * •_» . ,*.

applicable to humans, such extrapolations introduce uncertainty due to

differences in the exchange membrane (skin, til tract, aveoli) properties

between animals and humans and differences, between the human exposure

scenario and experimental design. Such absorption fraction estimates

also assume steady-state conditions which may not be achieved in the

actual, human exposure scenario.

o The exposure duration parameters are based on assumptions regarding

behavior patterns and various physical phenomena. These factors are

very difficult to^estimate, especially infa general father than site-

-36-
-------
specific basis.
The parameter values were typically selected from wide ranges. Tnese
ranges were carried through the Tier 1 calculations so that calculated risks
show an even greater range of uncertainty. Thus, the final risk estimates
reflect the uncertainty associated with the direct parameter assumptions.
The magnitude of this uncertainty is expressed by the difference between the
low and high estimates of risk given in the nomographs. These differences are
summarized below for each exposure pathway:
Exposure Pathway Orders of Magnitude' Uhc'ert'aTnty
Inhalation Exposure 1
Fish Ingestion Exposure 2
Dermal Exposure 2.5
Soil Ingestlon 2
Beef/Dairy Products Exposure 1
In addition to the direct assumptions associated with the various parameter
values, 1t 1s Implicitly assumed under Tier 1 that all site conditions which
could Influence conversion factors (terrain features, climatic conditions,
etc.) remain constant over the exposure period.
The uncertainty associated with the Tier 2 calculations should be much
less than the Tier 1 calculations since Tier 2 involves fewer apriori assumptions
regarding site conditions.
9.0 HOW TO USE NOMOGRAPHS
Nomographs have been developed to fadllate the Tier 1 calculations. One
-37-
-------
for each of the exposure routes is provided in Appendix A:

Title Figure Number
* *• ~
Dust Inhalation A-l

F1sh Ingest ion ' A-2

Dermal Exposure to Soil A-3
* :••» ' :'
Soil Ingestion A-4

Beef/Dairy Products Fat Ingest ion A-&

Each nomograph consists of 3 axes: cancer risk, soil concentration (TCOO

level-in soil at original source) and conversion factor. The intersection points

of a straight line drawn through the axes provides the solution to the problem.

Two of the three quantities must be determined before solving for the third.

Typically, the conversion factor and soil concentration are known and a risk

estimate is desired. This would involve plotting the conversion factor and

soil concentration and drawing the line through the points to the risk axis.*

The intersection point on the risk axis Is the risk corresponding to the

predetermined soil concentration and conversion factor.

The nomoaraphs were developed by combining Equations 1 and 2 into one
tf' . '
eouation for risk and making assumptions for all oarameter values except the"'

risk, conversion factor and soil concentration. The overall equation, parameter
1 ..-'.. * ' .„.
assumptions and conversion factor definition are listed on each nomograph. The

parameter assumptions are also summarized In Table 5.

Since7some of the parameter values span a range, a range of risk values
- f-^jw&." r •-..&&••£ A •-•• -' " -.k. ~ -,.;*..'

can be calculated as well. Thus, the risk axis has two scales. The upper risk

estimate was derived from parameter values chosen from the ranges to maximize
-38-
-------
TABLE 5. SUWARY OF EXPOSURE FACTORS USED IN TIER 1 CALCULATIONS
I
OJ

Dust
Inhalation
Fish
Ingestlon
Dermal
Exposure
to Soil
Soil
Ingestlon
Contact Absorption
Rate Fraction
23 m3/day .25-. 29
6.5 g/day .5-. 86
7900-110.000 g
life .0007-. 03
.1-5 g/day .2-. 26
Exposure
Duration
11,130-25,550
days
25,550 days
17,290-25.550
days
1240-1830
days
Body
Weight
70kg
70kg
70kg
17kg
Degradation
Effects
Ratio
.2-1
.2-1
.2-1
.84-1
Miscellaneous
Factors

Fish -Sediment
Distribution
Factor = 1-10

Beef/Dairy
Products Fat
Ingestlon 62 g/day
.5-.86
25,550 days
70kg
.2-1
Animal Fat to
Soil Blocon-
centratlon
Factor « .4
-------
the risk estimate. Conversely, the lower risk estimate was derived from
parameter values chosen from the ranges to minimize the risk estimate. Although,
• •«, * ••.- f
the resultingvjrange does not represent all possible uncertainty, 1t doesr reflect
1t to some extent.
In situations where the TCDD concentration is known in the medium of consrn
(ie. level in air at point inhaled, level in sediment at point where fish caught,
etc.), use a conversion factor equal to 1 and the soil concentration (middle axis)
equal to the known TCUU concentration in the medium of concern. Uepenainy on
^
the exposure pathway, the middle axis may represent something other than soil
concentration. In order to clarify what the middle axis-would-repres-ent'when
using this approach the following chart is provided.
Pathway Middle Axis Representation
, ' A *
Dust Inhalation; Concentration of TCOO in air at exposure point (ng/m3)
Fish Ingestion Concentration of TCOU in sediment where fish are caught (ng/c,)
Dermal Exposure Concentration of TCUU in soil at exposure point (ng/g)
Soil Ingestion Concentration of TCOD in soil at exposure point (ny/g)
Beef/Dairy Products Concentration of TCOU in pasture where animals yraze (ng/g)
Ingestion
Finally, it should be noted that the risk equation given on each nomograph
1s presented in the linear rather than the exponential form as given in Equation
1. "Kase two are equivalent for low exposure values since:
1-e-qd n q(j t when qd < 10'3
where, q,«'cancer potency factor*-
* ~ • .£ttA. 4 * ^dXtrr&mfe*
d • dose or expjsure
Thus, the nomographs give mathematically correct results only when the risk is
less than about 10-3.
-40-
-------
REFERENCES

Abraham, S. 197y. Height ana Heiyht of Adults 18-74 Years of Age, U.S. 1971-
1974.

Cocucd, S., 01 Gerolamo, A. Borderio et al. 1979 Absorption and Translocation
of TCUu by plants from polluted soil. Experientia 35:482-484, 1979

Cowherd, C., (i. Muleski, P. Enylehart and D. Gillette, 1984. Rapid Assessment
of Exposure to Particulate Emission from Surface Contamination Sites, EPA
contract #08-01-3116.

Dawson, G. et al. 1984. Conversion Factors for Assessing Potential Exposure
to Dioxin. EPA Contract No. 68-01-6861.

Freeman, A. Raymond and Jerry M. Schroy. 1984. Environmental Mobility of Uioxins.
Presented at 8th ASTM Aquatic Toxicology Symposium. April Ib, 1984. Fort
Mitchell, ICY.

Fries, G.F. '1982. Potential Polychlorinated 81 phenyV Residues Tn'Animai
Products from Application of Contaminated Sewage Sludge to Agricultural Land.
J. Environ, (jual. Vol.11, No.l, pp. 14-20.

Fries, G.F. and L.W. Jacobs. 1983. Tissue Residues in Livestock Confined to
Unpaved Lots with Soil that Contained PBB. J. Animal Sci. (in press).

Fries, George F. and George S. Marrow, 1975. Retention and Excretion of 2,3,7,8-
TCOD by Rats. Journal of Agricultural Food Chemistry. Vol. 23, No.2, pp. 2bd-
269.

International Commission on Radiological Protection. 1968. Report of Committee
IV on Evaluation of Radiation Doses to Body Tissues from Internal Contamination
due to occupational Exposure. ICRP Publication 10. Per^amon Press. New York

International Standards Organization (ISO). 1981. Recommendations on Size
Definitions for Particle Sampling. Report of Ad Hoc Working Grouu to Technical
Committee 146 - Air Quality, ISO. Am. Ind. Hyg. Assoc. J. (42), May 1981.

Jensen, D.J. et al. 1981. A Residue Study on Beef Cattle Consuming 2,3,7,8-
TLifi. J. A.jric. rood Chem. V. 29, 265-269

Kenaga, E.E., and »..A. Norris. 1983. Environmental Toxicity of TCOO. Human
and Environmental Risks of Chlorinated Dioxins and Related Compounds. Edited
by R. Tucker, A. Young, and A. Gray. Plenum Press.

Lepow, M. et al. 1975. Investigation Into Sources of Lead in the Environment
of Urban Children. Environ. Res., 10:415-426.
-41-
-------
Maybe, W.R. et al. 1981. Aquatic Fate Process Data for Organic Priority
Pollutants. EPA 440/4- 81-014.

McConnell, E.G. Lucier, R. Rumbaugh, P. Albro, D. Harvan, J. Mass and M. Harris
.1984.. Oioxin 1n Soil: Bioavailaoility after Ingestion by Rats ana buinea
Plgs.^i Science. March 9, 1984.

National Oceanic and Atmospheric Administration. 1980. Climates of the States.
2nd Edition. Gale Research Co. Detroit, MI.
ir
Poiger, H. and C. Schlatter. 1980. Influence of Solvents and Adsorbents on
Dermal and Intestinal Adsorption of TCDD. Food and Cosmetic Toxicology. Vol.
18, pp. 477-481.

Roelsl^Harvey et'al. 1980. Exposure to Lead by the Oral an3 Pulmonary Routes
of Children Living in the Vicinity of a Primary Lead Smelter. Environ. Research
Vol. 22. pp. 81-94.

Schrog, J.M., F.E. Hilenan, and S.C. Cheng. 1984. Physical and Chemical
Properties of 2,3,7,8-TCDO, The Key to Transport and Fate Characterization.
Presented at 8th ASTM Aquatic Toxicology Symposium April 15. 1984. Fort Mitchel, KY

Sendroy;-j; and L.P. Cecchinl. 1954. Determination of Human*Body Surface Area
from Height and Weight. J. of Appl. Physiology. 7(1):1-12.

Skldmore, E.L., and N.P. Woodruff. 1968. Wind Erosion Forces in U.S. and
Their Uses in Predicting Soil Loss. Agricultural Handbook No. 346. USOA.
Agricultural Research Service, Washington, DC.

Snyder, M.S. 1975. Report of the Task Group on Reference Man. International
Commission of Radiological Protection No. 23. Pergammon Press?: NY.

Stephanirc.E. 1980. Memorandum to J. Stara, U.S. EPA July 30, 1983.

U.S. EPA. 1981. Risk Assessment on (2,3,5-trichlorophehnoxy) acetic acid
(2,4,5-T), (2,4,5-trichlorophenoxy) propionic acid, 2,3,7,8-tetrachlorodibenzo-
p-d1ox1n (TCDD). Carcinogen Assessment Group. EPA-600/6-81-003.

U.S. EPA. 1983. Dloxin Strategy. Washington, DC

U.S. EPAJ%1984. Ambient Water Quality*Criteria Document for^ 2,3,7,8 TCDD.
Environmental Criteria and Assessment Office. Cincinnati, Ohio. EPA 444/5-84-
007.

-42-
-------
Wipf, R.K., Homberyer, E. Neuner, N. et al. iy«2 TilW Levels in Soil and Plant
Samples and Related Compounds - Impact on the Environment. Peryamon Series on
Environmental Science Vol. b. Oxford. iy«2. pp. lli-126.
Young, A.L. 1983. Long Term Studies on the Persistence and Movement of
1n a National Ecosystem. In: Human and Environmental Risks of Chlorinated
Dloxlns and Related Compounds, R.E. Tucker et al., Ed. Plenum Publishing Corp.
NY..
-43-
-------
Appendix - Nomographs
-------
FIGURE A-l. NOMJWAPH FOR INHALATION EXPOSURE
Cancer
Risk
Low Hitjh

10-9J10-B
4lU-6

10-6

1U-4

10-4 io-3
Soil Concentration
(ng/s or ppb)
10-b

10-4

10-2

1.0

102

10*

106
Conversion Factor
(g/m3)
l.U

1 -1

10-2

10-3

10-4
10-6

10-7

io-a
Cancer Risk. * (Potency Factor x Soil Concentration x Conversion Factor x Exposure
Duration x Inhalation Kate x Absorbed Fraction x Degradation Effects
Ratio)/(Body Weight x Lifetime)

where, Potency Factor » .26-.31 (ny/kg-day)-l
Exposure Duration » ll.l3U-2b,biO days
Inhalation Kate * 23 m^/day
Absorbed Fraction > .25-.29
Degradation Effects Ratio * .2-1
Body Weight * 70 kg
Lifetime « 70 yr

Conversion Factor • Concenxration of TCDD in Air at Exposure Point (nQ/tn3)
Concentration of Ttuu in Soil at Original Source
-44-
-------
FIGURE A-2. NvJMOGKAPHFOR FISH INGtSTlON EXPOSURE
Cancer
RISK
Low High
1U-7
10-7

10-6
10-i>

iu-4

10-3
Soil Concentration
or ppb) ,•;
10-6

10-4

1U-2

102
106
Conversion Factor
(unities*)
1.0

10-1

10-2

10-3
10- *

10- *

10-7

10-8
Cancer Risk » (Potency Factor x Soil Concentration x Conversion Factor x Exposure
Duration x Fisn Ingestion Kate x Fish-Sediment Distribution Factor
x Absorption/Fraction x Oegradatipn|Effects Ratio)/(Body .weight x
Lifetime)^
1-1U g so11/g fish
where. Potency Factor « .26-.31
Exposure Duration > 2b,5bO days
F1sh Ingestion Rate • 6.5 g/day
Fish-Sediment Distribution Factor
Absorption Fraction » .6-.86
Degradation Effects Ratio - .2-1
Body weight * 70 kg
Lifetime « VD yr

^Conversion Factor * Concentration of TCDD in Sediment Where Fish Are Caught
- 4
-------
FIGURE A-3. NOMOGRAPH FOR DERMAL EXPOSURE
Cancer
Risk
Low

10-10.
10

10
-9
-8
io-'

10-6
High

5 x IO-7

5 x lO'6
-5
5 x 10
5 x lO'4

5 x ID"3
Soil Concentration
(ng/g or ppb)
io-4

10-2

102

IO4

106
Conversion Factor
(unitless)
1.0
10-2
10
.4
10
-6
Cancer Risk » (Potency Factor x Soil Concentration x Conversion Factor x
Lifetime Soil Accumulation x Absorption Fraction x Degradation
Effects Ratio)/(Body Weight x Lifetime)

where* Potency Factor « .26-.31 (ng/kg-day)"*
Lifetime Soil Accumulation « 7900-110,000 g
Absorption Fraction * .0007-.03
Degradation Effects Ratio * .2-1
Body Weight » 70 kg
Lifetime « 70 yr
Conversion Factor « TCDD Concentration In Soil at Exposure Point (ng/kg)
TCDD Concentration in Soil at Original Source (ng/kg)
-46-
-------
FIGURE A-4. NOMO!aRA»>H FOR iCIL 1NGESTION EXPOSURE
Cancer
Risk
Low Hi gh
1U-9

10-8
10-6
10-7

10-6
10-*

10-3
Soil Concentration
(ng/g or ppb)
Ur3

10*1

• 101

103

105
Conversion Factor
(unitless)
Jl.O *

10-2

10-4

1U-6
Cancer Risk » (Potency Factor x Soil Concentration x Exposure Duration x
'* ' Conversion Factor 'x Soil Ingestion Rate x Absorption^Fraction x
Degradation Effects Ratio)/(Boay Weiyht x Lifetime) "';'•
!f
where* Potency Factor « .26-.31
Exposure Duration « 124U-183U days
Soil Ingestion Rate « .1-b g/day
Absorption Fraction « .2-.26
Degradation Effects Ratio * .84-1
Body Weight » 14 ng
Lifetime « 70 yr

Conversion Factor * TCDg Concentration in Soil at Exposure Site (ng/g)
TCDD Concentration in Soil at Original Source (ny/y)
-47-
-------
FIGURE A-b. NOMOGRAPH FOR BEEF/DAIRY PRODUCTS INbtSTION EXPOSURE
Cancer
Risk
Low High
10-9
10-8
10-7
10-6
10-5

10-8
10-7
10-6
10-5
10-4
10-3
Soil Concentration
(ng/y or ppb)
10-8

10-6

10-4

10-2

1
Conversion Factor
(unltless)
104
1.0

10-2

10-4

10-6
Cancer Risk * (Potency Factor x Soil Concentration x Conversion Factor x Exposure
Duration x Beef/Dairy Products Fat Ingestlon Rate x Animal Fat to
Soil B1 concentration Factor x Absorption Fraction x Degradation
Effects Rat1o)/(Body Weight x Lifetime)

where. Potency Factor « .26-.31 (ng/kg-day)"l
Exposure Duration • 25.550 days
Beef/Dairy Products Fat Ingestlon Rate » 62 g/day
Animal Fat to Soil B1oconcentrat1on Factor • .4
Absorption Fraction * .5-.86
Degradation Effects Ratio » .2-1
Body Weight • 70 kg
Lifetime « 70 yr
Conversion Factor
TCDD Concentration In Sol] at Pasture (ng/g)
TCOU Concentration in Soil at Original Source (ng/g)
•48-
-------