-------
ID
2
4-
ซ3
0.
^
to
E
L.
*
4-
ซJ
U
_
a
a.
<
^3
c
a
i
a
3
ซ^
^
C
^
^
1
X
LU
i
a
0
ซ*-
I
a
L.
4-
i
ID
a.
c
a
in
ง
a.
E
in
.
a
^>
e
4)
4-
3 ID
Q. L.
c a
Q.
T in
co co
^^ ^N
* ป
E >-
3 4)
O
ฃ X
U ID
TJ 3
m ID
O O)
CO
C^ ฃ
C a.
O
in
o
ii
in
in
4-
5
4-
UCM
a E
t -x
8 i1
^
u
outdoor
^r in
co co
* *
E >
U ID
CO X
-C
O)
.c
*ป
o
CO
Oi
^-"
E
3
ID
JZ
U
CO
c
T-,
4)
in
3
0
ao
in ซ
3
J
5
in
in
.
4)
X
X
in
4-
3
^
10
C
0
o
4)
in
ID
A
*^
in
CO
_
**
>.
4)
X 1-
a o
X X
s ?
L. ID
* >-
4} O>
- 3 C
a o
> a
in
to
in
*
ซ -o
4- V
ซJ 3
1- 0
O -o
E ID
in
CO
o\
^
^.
4)
ID
o
4)
U
in
ID
<ฃ>
r^
ป
*-^
o
t_
a
X
o
c
a
c
"o
CO
s
u.
8
d
d
4-
IO
>.
X
a
8
o
c
4-
OCM
ID E ^
4- U
5O1 ฃ
E U
1
^
^
4)
X
a
>Q
4)
ซ_
U
in
a
*o
p*-
V
**
^
o
^
t_
a
X
o
c
ID
c
o >
~o
CO X
ID
11
LU A
E -
ID >-
in 4> i_ J:
O 4-
4> X ป-
o 10 x
10 X CM
a. E 4-
m u u
O X ID
OJ *^ O) ^
c v e c
4- O
> ID CO U
C 4-
O
C 4)
Jฃ 0 t-
in o 4) ~o
4- X 3 C
3 i_ 75 "~
O O > L.
"- ซ 2
I- ป
O C T3
^ 4> 4) 4-
i- a. u
4) T3 O ID
4- 4-
ID *^ 4) C
L. ฃ > O
O 4) U
4- 'O
O L. 4-
10 O -0
4- -^ U-\ 4)
C CO t.
O W O*
O ID " X3
8
d
O CO
*
o
<0
u
10
a.
ซ i/i
8OI 4-
c
4- 'O 3
nj c > TJ
(_ ID
0 "O C
O 6 ID
in
ID
4-
ID
C ^
O L.
o 0
x
4) U
ID 4-
E 4-
IO
4-
m *-
4) O
4) 3
O
ID
X
t-" O
ID
in o
^^
O J=
O)
^- E
U
4-
4-
ID
in
2
.^
4)
X
ID
'
in
CO
d
O
CO
d
4-
C
ID
C
1 = i =
g S tn ง ,
O 4- O 4-
ID 4- ID
H- L. cm
O 4- Ifl 4- L.
C L. C O
O 4) O 4) O
t) O O T3
4- C -O C 4-
10 O C O 3
OS 0 0 O
154
-------
o
JC
4-
^
__
(O
ฃ
d)
Jit
^^
g
f-
T3
งin Em 4- in 0 Ch 6m E in Em in
ao Ooo in ao 4- co 000*003 o co ao
t_ Ch 1- O* 0O*^O* i_ Cft c uCfv L.G* o>
ซ. ^. ซ. JD O <- 10 ป. <-
^3 C
-Oซ "Oft -a-<:ซT3ft0T3ft TJซ ft
0>. 0>- CXC0X00X 0>- X
4-0 4-0 IO0". ^ \ ^ c in CT>
\ L. 1/1 L. ป 4- L. L.L.O 0
in O 4-.C QincQ o0Ef
.c -^ 04- o ฃ g o 5-1-4-
4- ioc ovtl'o -aci. -
C X 4-Q CC C C 0 X (fl
810 CE ~Ol.~ XE E
a o E 0 E m i_
uvo -^ 6 *- in vo 3 > 10
IO > Ol-eO Otnio
U4- 03 Eป^
ซ 00 ~^ >.4-in>- 3>>io in
^ >io OX ioc a* inaCM TO
X 04- tnX T3 IO"OL. Ifl-O" c
v c in xx xio ซex>.-io
in ino in>ป .* UU I.IO t_ - CM iDTtuinin om
r-4- 4-0 in CM-ซ-io <-io a. CN jz ui u. no
C3in ซx ~o m 0u cr o vo 00
E -o L. E EOL.EIO cau E
3 f 3 ฃ O 34-ฃ 34- 4-in C,34-
in ininro inc in c >c34-~m0
ino ฃCM in a in o CM ino O 4- .* in 0
in QO
CM CM IO ซป VO CM
CM
E
U
O O m o O OCMO
01 Q^ 10 r** GO r** r** t*^
in in fi in oo
CM CM - IO & IO CM
T3
0
in in
i- O
^. <-> o o.
L. L. 5 X
ฃ r *o E4-3 E O ET3T3 E3>4- 04-E
3 ฃ ~3 ~3T3 C .C C _-o~4- I-3O
t O U h-OO KOlO H- O 1 ~ O I IO ID ซC O **
155
-------
>.
IO
X
jC
4-
IO
a.
_ m
IO
E
i_
O
,.
c
o
4-
u
Q.
a
<
X)
c
IO
_J
1
^^
IO
XI
">
c
o
4)
in
^
X
ULt
4-
in
-j*
^
in
4)
"5
>
4>
i
IO
(L
o
c
a
in
ง
4_
a.
5
U)
^-*
*-
g
8
^^
w
Q
^f
CซJ
JO
^.
0)
u
0)
t.
4>
i
ง
4-
ID
C
ID
a.
X
4>
in
4>
i
JC
01
_
X
4-
IO
E
m 4-
4) U)
.
^
4)
0)
4- E
3 IO
a. i.
c 10
a.
^
c
10
c
0)
^
ง
\_
*-
c
o
4-
a.
E
3
in
4-
4)
4)
ซ*
XI
C
10
m
4)
w
ป
in
c
ID
C.
^
4-
E
in
4)
in
in
CM
u
8
CM
K"l
CM
U
8
CM
4)
in
O
a.
X
4)
C
^
V)
0
IO
4)
C
in
D
fQ
>
ID
4) in
L- OO
ID 2
4)
u
ID >
>*- 4)
3 X
m ID
ป
ON 1_
^b -
IO
a.
O)
c
._
u
3
XI
m
X
4)
{_
4)
o
-^ >ซ^
OCN
E
0
8 =
^ *_
3 JZ
o u
~~*
^r
u"i
ซ
*
o
L
T T3
CO C
O* 4)
O)
E c
3
ID 4-
J=
O U
lf> "
ป
"U l/i
?! ง
4) J=
4) m
^
in
i m
4- 4-
l_ C
O ID
j= a.
in
01 ^
c u
4>
l_ C
ID
0 C
X 4)
a.
m o
.^
c
4- ID
ซ_
3 ฃ
X) 4-
IO
2
in
4> -
E 4-
3 L.
in ~
in jc
< tn
CM
u
o
o*
CM
CM
u
o
^^
CM
w^e
O U
a. ~
X
4) 4-
C 3
^ ^
.* ID
in
*
ฐ 8
4) 4-
l_ 3
< O
4-
IO
C.
^
o
m
4)
O
o>
O
U^
co
^h
,;
4)
X
IO
X
m
ID
4)
E
ID
in
in
L.
3
"2
>_
o
O
in
O
a.
X
4>
ID
4>
ID
in
m u
f 8
M 4-
01 3
< O
CM
U
8
CD
CM
CM
U
O
Q9
CM
m CM
O E
a. u
x ~
x>
c
^ ซ^
^> *"*
a
4)
a
in
in u
i 8
3 X>
in 4-
tn 3
< O
CM
u
o
o
CM
%
U
8
CM
XI
4>
10
O
a.
X
4>
L.
C 4>
XI
X CM
in o E
u
"o tT ~*
. 8 2
4) XI
l_ C .C
< 0
.^
--^
in
co
Q\
^
4)
X
ID
in
ID
4)
E
in
in
8
XI
c
XI
4)
in
0
a.
X
4>
ID
4)
ID
in L.
i 8
3 XI
in 4-
m 3
< O
CM CM
E ~ E 01
o o o c -~
O
ซT 4- ซT ID
O\ (0 ^h ^ ^
CM
O U O C *
_ Q)
O 4- O > (J
ซT 4- T ID
O\ ID -^ CL
CM *-* CM in
XI
O E
Q. U
X >-
4)
4-
e
^ 3
^ XI
in ID
"o ฃ
IO O
4) XI
L. C
4>
j:
4- ID
E
O 4)
ซ*- Q
I/) C
C ID
O E
3
4- X
a.
E "-
3 O
in
in 4-
4>
O in
E in
E 4)
O in
O in
^.
O
c
^
m
4-
a. 10
3 JC
O 4-
O) XI
t i
i 3
in
> ID
4-
m
5 S
n
.^
ID >-
> 4>
ID
O X
ID
m x
4- JC
in o
c
1 4)
XI
CM
O J=
O
O -*
T T
C JC
CM *ป"
CM
O O
O 0
^^
4-
c
J3 ^
10 in
IO O)
> 3
10 O
0 l-
(B 4-
XI
4)
4- E
ID O
^
1- XI
O c O
ID g
JC 4)
u in E
*~ c
O x "-
Q- O i/l
c
o a ID
4- 1 l_
a. 3
4> 1 *-
1-00
3 N N
in c c
O 4> 4)
a. JD n
LU a a
.
in
4-
3
^
ID
Q
C
O
4-
Q.
O
m
J3
ID
O
J=
4-
4>
U
X
4-
m
ID
JC
c
4>
J=
O
.ฃ
c
0
c
4)
t_
XI
JC
u
ซ
^"
3
5
- in
U CO
-^ CO
z
156
-------
X
o
ฃ
10
Q.
10
5
O
01
u
c
0)
O
ซ
10
>
u
a
L.
IO
a.
o
c
u>
O
u>
in
O
ID
c
to
a.
X
V
O)
I
n>
4-
I/)
H) E
ฃ 3
O) TJ
JC IO
g
ง
O O
m
X L.
5 ^ 15
< T3 6
157
-------
the lead in household dust was 75% the concentration of lead in the outdoor soil. For his own
analysis, Hawley (1985) assumed that indoor contaminant concentrations in dust were 80% of the
contaminant concentrations in outdoor soil. The analysis of typical exposure uses a value of 80%
for the best estimate, and applies a range of 75% to 85% for the low and high estimates, respectively.
For the MEI analysis, the best estimate assumes a value of 80%, and the high estimate uses a value
of 85%.
Data Sources and Model Inputs for Contact Rate
The contact rate of soil on skin varies between outdoor and indoor exposures and among age
groups. Hawley (1985) and Schaum (1984) both described a number of studies that estimated the
contact rate of soil on the skin of children playing outdoors. Lepow et al. (1975), as cited in Hawley
(1985), estimated a contact rate of 11 mg soil per 21 cm2 of skin on the hands of young children, or
0.5 mg/cm2. Exposed skin on other parts of the body is assumed to have the same contact rate. Roels
et al. (1980), as cited in Hawley (1985), found that the mean values for quantity of dirt on one hand
of eleven-year old children ranged from 40 to 180 mg. Since the hand of a child this age has a
surface area of approximately 300 cm2, these data suggest a contact rate ranging from 0.13 to 0.6
mg/cm2. Schaum (1984) reported the upper end of the estimate for outdoor contact rate for children
to be 1.5 mg/cm2. In the analysis of typical exposure, 0.5 mg/cm2 is used for the best estimate, 0.13
mg/cm2 the low estimate, and 1.5 mg/cm2 the high estimate of contact rate for children. The MEI
analysis uses an outdoor soil contact rate of 1.5 mg/cm2 for older and younger children.
For adults, the outdoor contact rate was derived by Hawley (1985), based on assumed
thickness of the layer of soil on the skin and the density of outdoor soil. Hawley's calculations
yielded a value of 3.5 mg/cm2. This value is used in the analysis of typical exposure for "best" and
"high risk" estimates. This value is also used in the MEI analysis. For a "low risk" typical exposure
estimate, outdoor contact rate for adults is assumed to be the same as the low estimate of outdoor
contact rate for children.
Hawley (1985) estimated indoor contact rates based on assumptions regarding dustfall and
frequency of cleaning. Hawley also cited the work of Solomon and Hartford (1976), who studied
lead and cadmium levels in indoor dust. The dust values measured by these researchers ranged from
110 mg/m2 to 590 mg/m2. For his analysis, Hawley (1985) used a value of 0.056 mg/cm2 for indoor
dust contact rate, assuming a dustfall rate indoors that is 20% of the outdoor dustfall, and assuming
biweekly cleaning of surfaces. The typical exposure analysis uses Hawley's value as a best estimate,
and uses the range of values reported by Solomon and Hartford (1976, through Hawley, 1985) as low
158
-------
and high estimates. These values are used to represent contact rate indoors in living space for all
three age groups. A value of 0.06 mg/cm2 is used in the MEI analysis.
Adults may also experience dermal contact with soil when engaged in infrequent cleaning
of seldom-used spaces, such as attics. After a discussion of the relevant literature, Hawley (1985)
concluded that an adult working for a one-hour exposure in a dusty space such as an attic has indirect
dermal contact with 110 mg of dust suspended in air. In addition, the direct contact rate with dust
was estimated to be 1.8 mg/cm2, based an assumed depth of the dust layer on the skin and the density
of indoor dust particles. To assess risks from these exposures, the analysis of Hawley (1985) is
incorporated into the typical and MEI analyses.
Data Sources and Model Inputs for Area of Skin Exposed
The surface area of skin available for contact with contaminated soil will influence the
quantity of TCDD and TCDF absorbed through this pathway. The surface area available for contact
will vary depending on the clothing worn by the individual. Hawley (1985) provides a table of
surface area for various parts of the body for young children, older children, and adults. In the
following discussion, the area of the skin assumed to be exposed in each scenario for each age group
is derived from this table. The assumptions regarding the body parts exposed in each scenario are
also derived from Hawley (1985), except as noted.
For the best estimate of typical exposure, it is assumed that the feet, legs and hands of young
children are exposed to soil during outdoor play, an area of 2100 cm2; indoors, one-half of the area
of the hands, forearms and feet, or 500 cm2 is assumed to be exposed. The "low risk" estimate of
typical exposure assumes that only the child's hands are exposed both indoors and outdoors (300 cm2):
the rest of the body is covered with clothing. The "high risk" typical exposure estimate assumes that
young children's hands, arms, legs, and feet (2800 cm2) are exposed outdoors, while feet, hands, and
forearms are in contact with indoor dust (1000 cm2). To calculate MEI exposure estimates, a value
of 2800 cm2 is used for both indoor and outdoor exposures.
For older children, the typical exposure analysis uses a value of 1600 cm2 for the "best"
estimate of the surface area of skin exposed while playing outdoors. This value represents exposure
of both hands, forearms, and half of the legs (i.e., from the knees down). Indoors, older children
have 400 cm2 of skin in contact with indoor dust, an area equivalent to the area of both hands. For
-the-stew-risk" estimate of typical exposure, the analysis assumes that only hands are exposed both
outdoors and indoors. The "high risk" estimate of typical outdoor exposure is based on Keenan et al.
(1989), who assumed that children playing outdoors expose both hands, legs and feet to soil. The
159
-------
surface area corresponding to these parts of the body for older children is approximately 3200 cm2
(Hawley, 1985). The "high risk" estimate of typical exposure assumes that the hands and the forearms
of the older child, or approximately 825 cm2 of skin, are exposed indoors. The MEI analysis uses the
value 3200 cm2 for both indoor and outdoor exposure.
The hands and forearms of adults working outdoors are assumed to come into contact with
contaminated outdoor soil. The area of these body parts is approximately 1700 cm2. This value is
used in the calculation of the "best" estimate of typical exposure. As a "low risk" estimate, Schaum
(1984), citing Sendroy (1954), assumed that adults may wear a long-sleeved shirt, gloves, pants, and
shoes to work outdoors. In this case, the area exposed is 910 cm2. The "high risk" estimate of typical
exposure uses a value from Schaum (1984), citing Sendroy (1954), who assumes that adults may wear
a short-sleeved shirt with an open neck, pants, shoes, with no gloves or hat, to work outdoors. The
area of skin exposed under these assumptions is 2940 cm2.
For adults indoors, different assumptions can be made for the area of skin exposed while
the adult is in the living space and the area exposed while the adult works in an attic. For the "low
risk" and "best" estimates of typical exposure, adults working in the attic are assumed to wear an
open-neck, short-sleeved shirt, pants, shoes, and no gloves or hat, while adults in the living space
wear clothing that covers a larger area of skin and behave in such a manner that only the hands are
in direct contact with indoor dust. This corresponds to an area of 1700 cm2 in the attic, and 900 cm2
in the living space. The "high risk" estimate of typical exposure assumes that 1700 cm2 of skin are
exposed in both the attic and in the living space. For calculating MEI exposures, the analysis
assumes that 2940 cm2 of skin are exposed both indoors and outdoors.
Data Sources and Model Inputs for Exposure Duration: Indoor and Outdoor Soils
The length of time soil is in contact with the skin is an important factor in determining the
amount of TCDD or TCDF that is absorbed into the system through the skin. The following
assumptions regarding duration of dermal exposure are derived from Hawley (1985). The "low risk"
and "best" typical exposure analyses assume that young children spend 5 days a week, six months out
of the year playing outdoors. The outdoor soil is assumed to remain in contact with the skin for
twelve hours before it is washed off. Twelve hours are also spent in contact with indoor dust.
During the winter months, young children are in contact only with indoor dust, for 12 hours per day.
In the "high risk" typical exposure estimate, young children play outdoors seven days per week, six
months out of the year, with soil remaining on the skin for twelve hours. The remaining twelve
hours is spent in contact with indoor dust. The high estimate for young children also assumes that
160
-------
young children are in contact with indoor dust 24 hours a day during the six winter months. This
assumption is also used in the MEI analysis.
For the typical exposure analysis, the "low risk" and "best" exposure assessments assume that
older children spend some time outdoors everyday between May and September (5 months), and allow
the outdoor soil collected on the skin to remain there for twelve hours before washing. In addition,
older children are assumed to be in contact with indoor dust for four hours per day all year; the rest
of the time is spent at school or other locations. As a "high risk" estimate, and for the MEI analysis,
older children are assumed to spend some time outdoors every day for six months, and to allow the
outdoor soil to remain on the skin for 12 hours before washing; furthermore, these children are in
dermal contact with indoor dust for 12 hours every day year round.
Adults who live at agricultural land application sites are assumed to have dermal exposure
to contaminated outdoor soil 5 days a week, six months out of the year in the "best" typical exposure
calculations. Furthermore, the soil remains on the skin for twelve hours before it is washed off the
skin. Indoors, adults are in contact with indoor dust for twelve hours a day all year. The "low risk"
typical exposure analysis assumes that the adult lives on the farm but works elsewhere; as a result the
dermal contact is reduced to only two days per week for five months of the year; the soil is assumed
to remain on the skin for eight hours before washing. For the "high risk" typical analysis, and in the
MEI analysis, the adult experiences dermal exposure to indoor dust for 12 hours, a day during the
summer months, and 24 hours a day during the winter months.
As discussed earlier in this section, adults may also have limited dermal exposure while
cleaning seldom-used spaces such as attics. Hawley (1985) assumes that an adult spends 12 hours
in these environments during one year. This value could represent a single cleaning, where the adult
spends one twelve-hour period in the attic, or it could represent twelve one-hour cleaning sessions.
For the best estimate, it is assumed that the adult spends twelve days'in the attic, one hour each day,
and leaves the dust from attic on the skin for four hours before washing. The low estimate assumes
the adult spends one day in the attic for twelve hours, and leaves the dust on the skin for an
additional four hours before washing. For the "high risk" typical estimate, and for the MEI estimate,
the adult engages in twelve one-hour attic cleaning sessions, and leaves the attic dust on the skin for
six hours after each session.
Data Sources and Model Inputs for Dermal Absorption of TCDD and TCDF
Dermal absorption of TCDD and TCDF bound to soil involves two components: migration
of the TCDD and TCDF from the soil matrix, and absorption of TCDD and TCDF through the skin.
The Consumer Product Safety Commission (1989) reviewed data pertaining to the dermal absorption
161
-------
of TCDD from a variety of matrices. The Consumer Product Safety Commission (C'PSC)
memorandum cites studies by Poiger and Schlatter (1980), who reported absorption of TCDD from
wet soil ranging from 0.05% to 2.2%, and by Shu et al. (1988), who reported absorption of 0.65% to
1% with dry soil. Comparing these absorption rates to the rate of dermal absorption when TCDD is
applied to the skin in a methanol vehicle, CPSC concluded that from 0.3% to 15% of the TCDD in
soil is released for subsequent absorption through the skin. In the typical exposure analysis, the
recommendation of the CPSC memorandum is followed and a value of 1% is used to represent the
best estimate for this matrix effect for contaminated soil; the typical analysis uses the range of 0.3%
to 15% for the "low risk" and "low risk" estimates, respectively. For the MEI analysis, a value of 15%
is used.
CPSC also reviewed the literature regarding the percutaneous absorption of the TCDD release
from the soil matrix. Studies reviewed included studies with laboratory animals and in-vitro studies
of human skin. The animal studies report percutaneous absorption rates ranging from 40 to 48%
over 72 hours. From the in-vitro skin experiments (Weber et al., as cited in CPSC memorandum),
CPSC estimated an absorption rate of 18.5% over 17 hours of exposure, yielding a transfer coefficient
of 0.012 h"1. This value is used for the "low risk" and "best" typical exposure assessments for all age
groups. For children, Hawley (1985) states that the absorption rate through skin for children is twice
the absorption rate for adults. Therefore, in this analysis, a transfer coefficient of 0.024 hr"1 is used
for the "high .risk" typical exposure analysis and for the MEI analysis for both younger and older
children.
Data Sources and Model Inputs for Estimating the Population Exposed
In this analysis, the population exposed to TCDD through dermal contact is limited to the
population residing on the agricultural land application sites. The number of sites applying kraft
mill sludge to land is equal to the total number of acres applied with sludge in the state divided by
the average number of acres per site. Values for both the total acres and the acres per site were
obtained through conversations with state officials in Mississippi and Pennsylvania, the states where
agricultural land application is currently practiced. In Mississippi, 1000 acres are applied with the
sludge from one mill, with an estimated 100 acres per site, yielding an estimate of 10 sites in
Mississippi. Pennsylvania has 75 acres covered with sludge from one mill, with an average of 15
acres per site, giving a total of 5 sites. The total number of sites in each state is multiplied by the
number of people living on each site to obtain the exposed population. According to the 1980 U.S.
Census, the average number of persons per household is 2.7. In Pennsylvania, the exposed population
is approximately 14 persons, while in Mississippi, the exposed population is approximately 27. The
total exposed population is about 40 persons.
162
-------
2.4.2 Estimates of Exposure and Risks Estimates from Ingestion of Produce, Meat, and Dairy
Products Grown on Sludge-Amended Land
Sludge is applied to various types of land, including forest, abandoned mines, pasture, and
land used for the production of animal feed or human food crops. This section evaluates the
application of sludge to pasture and cropland, and examines the following potential pathways of
exposure:
Sludge is incorporated into the soil of farmland used for producing food crops.
Contaminants in the sludge are drawn from the soil to the tissue of those crops, and
are then ingested by humans who consume the crops directly.
Sludge is incorporated into the soil of farmland used for producing animal feed or
pasture. Contaminants in the sludge are absorbed into the tissues of these feeds or
pasture grasses, which are then consumed by livestock. The meat and dairy products
produced by these livestock are consumed by humans.
Sludge is applied to the surface of pasture land, and adheres to the pasture grasses.
Grazing cattle or sheep ingest the sludge directly as a fraction of their pasture
consumption. Humans then consume contaminated beef or dairy products.
According to conversations with state environmental officials, sludge from two bleached
kraft mills, one in Mississippi and one in Pennsylvania, is currently applied to agricultural land.
Several hundred hectares of agricultural land are receiving the sludge at dozens of individual sites.
Though it was not possible to determine the land area and crops grown at individual sites, Mississippi
and Pennsylvania environmental officials indicated the overall uses of the sludges in each state. In
Mississippi, four crops are receiving the sludge: corn, soybeans, wheat, and pasture. In
Pennsylvania, the sludge is generally used on feed corn,
The methodology and data inputs for assessing human exposure due to sludge application to
agricultural land are discussed below. The calculations for determining risk from the dietary pathway
follow.
Methods for Estimating the Human Exposure to TCDD and TCDF through the Dietary Pathway
To estimate the concentration of contaminants in food products from land application of
paper and pulp mill sludge, a model was created which uses information regarding sludge application
163
-------
rates and sludge contaminant concentrations to calculate the uptake of contaminants by crops and by
animals feeding on crops and pasture. A "best estimate" of human exposure is calculated using data
on human dietary consumption of these meats and crops and the quantity of contaminated produce.
A "high risk" and a "low risk" scenario are evaluated. For each scenario, the model is given input data
for: sludge incorporation depth, the number of years sludge is applied, the contaminated land area,
sludge application rates, concentrations of individual contaminants in the sludge, uptake rates of soil
contaminants to various crop tissues, uptake rates of contaminants in animal feed to meat or dairy
products, the fraction of each type of feed in animal diets, the production yield of animal product
per unit of food, human dietary data, the acreage of sludge-amended land devoted to each crop, the
productivity of land for each crop, and the population to which the contaminated food is distributed.
The model returns exposure estimates which are then used to estimate risk.
Exposure and risk are calculated for both a typical individual and a most exposed individual
(MEI). The most exposed individual is assumed to be a farmer growing and applying sludge to all
crop types grown on sludge-amended land in the state in which he or she resides.
Each exposure calculation consists of three steps. First, the model calculates tissue
concentrations of contaminants in each crop as a result of the land application of sludge. Second,
the model estimates concentrations of the contaminants in meat or dairy products. As discussed
above, contaminants are assumed to enter meat and dairy products as a result of, animal ingestion
of sludge-treated crops and pasture grasses and of direct ingestion of sludge adhering to pasture
grasses. Third, the model sums the amount of each contaminant in all crops and animal products
ingested by humans to estimate typical population exposure or MEI exposure.
Method for Estimating Tissue Concentrations of Contaminants in Crops Grown on Sludge-amended
Soil
Soil concentration calculations are discussed in Appendix A. The calculations for determining
crop tissue contaminant concentrations from soil concentrations are:
CDfj = CjU,, (2.4.1)
(2.4.2)
where:
CD. - = Tissue concentration of pollutant j in crop i (mg/kg dry weight)
C- ----- -- ป --- Concentration of pollutant j in the soil (mg/kg)
U-. = Rate of uptake of contaminant j into tissue of crop i (mg/kg dry
weight per mg/kg dry weight)
164
-------
CW.. = Tissue concentration of pollutant j in crop i (mg/kg fresh weight)
KDWj = Constant for converting dry weight concentration to fresh weight
concentration for crop i.
Each crop's uptake rate is applied to the soil contaminant concentrations to estimate the
concentration of each contaminant per unit dry weight of crop tissue (2.4.1). Dry weight tissue
concentrations are converted to fresh weight concentrations to match units of human consumption
data (2.4.2).
Method for Determining Tissue Concentrations of Contaminant in Meat and Dairy Products Produced
with Sludge-amended Soil
To calculate meat and dairy product contaminant concentrations, the model derives average
concentration of each contaminant in each animal's feed mix. Average concentrations are calculated
by taking a weighted average of the contaminant concentrations in each animal's food sources. An
additional source of contaminant, direct soil ingestion, is added to the dose derived from food (2.4.3).
The feed contaminant concentrations are multiplied by animal bioconcentration factors to determine
fresh weight concentrations of contaminants in each meat or dairy product (2.4.4).
- S,(FJk CD,,) + (FDk Nj) (2.4.3)
- CFjk Ujk (2.4.4)
where
CF-k = Weighted average concentration of contaminant j across all food
sources for animal producing meat or dairy product k (mg/kg dry
weight)
Fjk = Fraction of animal k's food from crop i (unitless)
CD,- = Concentration of contaminant j in crop i (mg/kg)
FDk = Fraction of animal k's food from sludge (adherence pathway) (unitless)
N, = concentration of contaminant j in sludge (mg/kg)
CW -k = Tissue concentration (fresh wt) of pollutant j in meat or dairy product
(mg/kg fresh weight)
Ujk = Rate of uptake of contaminant j into meat or dairy product k per unit
of concentration in animal's food (mg/kg fresh weight per mg/kg dry
weight)
165
-------
Method for Determining Most Exposed Individual Dose from Contaminant Ingestion through Foods
Grown in Sludge-amended Soil
Methods of calculating MEI exposure and population exposure differ. The calculation for
MEI exposure proceeds in three steps. First, exposure through direct consumption of crops is
calculated. To obtain the exposure from direct crop consumption, the daily dietary consumption
of each crop is multiplied by the fraction of that crop produced in sludge amended soil, by the fresh
weight contaminant concentration of the crop, and by the bioavailability of the pollutant when
consumed with the crop:
Dc = E, (CWn FC1 DC. B, 10'6)
where:
B,- = Bioavailability of pollutant when consumed in crop i (unitless),
Tissue concentration of pollutant j in crop i (mg/kg fresh weight),
-
DC = Dose of pollutant j from crops produced with sludge-amended soil
(mg/kg/day),
DC,- = Daily dietary consumption of crop i (mg/kg/day fresh weight),
FCi = Fraction of dietary consumption of crop i grown in sludge-amended soil
(unitless)
4
In the second step, MEI dose of the contaminant through consumption of animals raised on
contaminated feed is determined. This step incorporates the contamination of animal products
through both consumption of contaminated feed and through grazing. The equations are similar to
those discussed for dose from crop consumption:
Da = =k
-------
Finally, dose from crop and animal consumption are summed:
D; = Dc + Da
where:
Dj = Total exposure to pollutant j from crops, meat and dairy products produced
with sludge-amended soil (mg/kg/day).
Method for Determining MET Cancer Risk
Once the daily dose estimate to the MEI is estimated, it is combined with information about
the cancer slope factors of TCDD and TCDF to obtain an estimate of lifetime risk from dietary
exposure to these contaminants. The calculation of MEI risk is:
1C = DOSEMEI q1*
where:
DOSEMEI = weighted average daily dose for an MEI, mg/kg/day
1C = individual cancer risk over lifetime from DOSEavg of TCDD or TCDF
q,,* = incremental lifetime risk per mg/kg/day dose of TCDD or TCDF
4
Method for Determining Population Contaminant Dose from Ineestion of Foods Grown in Sludge-
amended Soil
Contaminant doses are calculated for each state in which sludge is applied agriculturally and
then summed over these states to yield total population dose. The dose from the dietary pathway is
calculated by summing three inputs: dose of contaminant bioavailable in crops for direct human
consumption, dose of contaminant bioavailable in animal products contaminated by crop
consumption, and dose of contaminant bioavailable in animal products contaminated by grazing. The
following calculation describes typical population dose:
- (Dj + A; + Gp/BW/DP/DY
where:
Aj = Dose of pollutant j bioavailable in animal products contaminated by crop
consumption (mgs/year)
DP = Population over which the crops and animal products are distributed
BW = Body weight (kgs)
DY = Days per year
167
-------
D. = Dose of pollutant j bioavailable in crops for direct human consumption
(mgs/year)
G- = Dose of pollutant j bioavailable in animal products contaminated by grazing
(includes soil adherence and grass uptake) (mgs/year)
TD. = Total exposure to pollutant j from crops, meat and dairy products produced
with sludge-amended soil (mg/kg/day)
The following discussion describes the methods used to obtain each of the three components of the
estimation of total dose.
Method for Determining Population Contaminant Dose from Ingestion of Crops
The dose of contaminant bioavailable in crops for direct human consumption is calculated
by multiplying the mass of each crop grown on sludge-amended land that is directly consumed by
humans by the tissue concentration of the crop. The result is multiplied by a bioavailability factor:
Dj = (MH, CW.. B5)
where:
Bi = Bioavailability of pollutant when consumed in crop i (unitless),
CW.. = Tissue concentration of pollutant j in crop i (mg/kg fresh weight),
MHi = Mass of crop i grown on sludge-amended land that is consumed directly by
humans (kgs fresh weight/year).
The mass of each crop grown on sludge-amended land that is consumed directly by humans
is calculated by multiplying total acres of the crop receiving sludge by the crop yield. This mass is
then multiplied by the percent of the crop consumed directly by humans:
M, = A, Y,
MH, = Mi PH^
where:
Ai = Sludge-amended land area on which crop i is grown (hectares/year),
Mj = Mass of crop i grown on sludge-amended land (kgs/year),
Yi = Yield per area of crop i (kgs/hectare),
= Percent of crop i consumed directly by humans (unitless).
168
-------
Method for Determining Population Contaminant Dose from Ingestion of Animals Fed on
Contaminated Crops
To determine the dose of contaminant available through consumption of animal products
fed contaminated crops, the mass of each contaminated crop fed to animals is multiplied by an
animal product yield from each weight unit of contaminated feed. To obtain a dose to humans who
ingest the animal products, this yield is multiplied by contaminant concentration in the animal tissue,
percent of the animal that is fat, and a bioavailability factor:
Aj - ฃjk(MAJk YAk CWjk Bk PFk / FAkj)
where:
Bk = Bioavailability of pollutant when consumed in meat or dairy product k
(unitless),
CW.k = Tissue concentration of pollutant j in meat or dairy product k (mg/kg fresh
weight),
FAkj = Fraction of dietary consumption of crop i for animal k (unitless),
MAjk = Mass of crop i grown on sludge-amended land that is fed to animal k
(kgs/year),
PFk = Percent fat in animal product k (unitless),
= Yield of animal k per unit of corn-equivalent feed (kg/kg).
To obtain the mass of each crop fed to each animal the mass of each crop fed to all
animals is divided between the animals according to each animal's percentage of total consumption:
MA, = Mj - MH,.
MAik = MAj((NkCk)/Ek(NkCk))
where:
Ck = Food consumed per animal k (kgs/year)
MAj = Mass of crop i grown on sludge-amended land that is fed to animals (kgs
fresh weight/year)
Nk = Number of animal k in the state
169
-------
Method for Determining Population Contaminant Dose from Ingestion of Animals Grazing on
Contaminated Pasture Land
To obtain the dose of contaminant from animal products raised on sludge-amended pasture
land, the number of contaminated animals marketed is multiplied by an average production weight
to yield a total mass of meat available for consumption. This mass is then multiplied by the tissue
contaminant concentration, the percent of the animal that is fat, and a bioavailability factor:
G, - 2g(NSg WHg CWjk Bk PFk)
where
NS = Number of each animal, g, grown on sludge-amended land and marketed
(animals/year),
WH = Average U.S. production of grazing animal g per head marketed (kgs/animal).
The number of each type of animal raised on sludge-amended land is the product of the
percentage of pasture land that receives contaminated sludge and the total number of animals grazed
on pasture land. The resulting number of animals is then multiplied by percent of grazed animals
that are marketed each year:
NSg = (SPC/PC) Ng PMg
where:
N = Number of grazing animal g in the counties with sludge-amended pasture,
Pc * Total pasture land in all counties with sludge-amended pasture land in state
(hectares),
PM s= Percent of grazing animals g marketed per year on a national basis,
SPC = Sludge-amended pasture land in all counties in state applying sludge
agriculturally (hectares).
Method for Determining Cancer Risk
Once the daily dose estimate is obtained, it is combined with the cancer slope factors of
TCDD and TCDF to obtain an estimate of lifetime risk from dietary exposure to these contaminants.
The calculation of typical individual risk is:
1C = DOSEayg QI*
170
-------
where:
DOSE = weighted average daily dose for an individual, mg/kg/day
1C = individual cancer risk over lifetime from DOSE of TCDD or TCDF
q.,* = incremental lifetime risk per mg/kg/day dose of TCDD or TCDF
Individual cancer risk for an typical exposed individual is converted to annual total population risk
(in cases per year) by multiplying individual exposure by the number of persons exposed to the
individual risk and dividing by the average person's lifespan, as described in the following equation:
PC = 1C POP / LS
where:
LS = average lifespan of an individual = 70 years
PC = population risk, cancer cases per year
POP = population exposed to DOSEflvg
Data Sources and Model Inputs
The values used for each model input are summarized in Table 2.4.F and Table 2.4.G for
typical individuals and the most exposed individual, respectively. The following sections describe
each input and document the data sources used to obtain values for each model input parameter.
Data Sources and Model Inputs for Soil Concentration
The methodology for calculating soil concentration is discussed in Appendix A. Estimates
of application rates and depth of soil incorporation were obtained from the Bureau of Pollution
Control, Mississippi Department of Environmental Quality for Mississippi sites and from the
Pennsylvania Bureau of Waste Management for the Pennsylvania sites. Sludge is assumed to be soil
incorporated in both states at 6 inches. Application rates are estimated at 58 dry metric tons per
hectare for Mississippi and 18 dry metric tons per hectare for Pennsylvania. Table 2.4.A displays
the average soil concentrations over a 70 year period assumed to result from land application of
sludge.
Data Sources and Model Inputs for Sludge-amended Acreage. Crops Grown on Sludge-amended
Land, and Percent of Crops Fed to Animals
The Mississippi Department of Environmental Quality estimates that approximately 1000
acres of farm land are sludge-amended in that state. The Pennsylvania Bureau of Waste Management
171
-------
^
ID
JZ
+.
^^
Q
4-
4)
5
c
0
4-
ID
E
in 4-
m ซ
'
o
L.
o
s 1
a. (_
C ID
Q.
O)
c
-t- ao
0} GO
.* O
L. 'O l_
ID O c O h- 3 ID C
_ = % 2 2 ซ"
40 < Q 4- 4)
< Q. a. 3 cn
O UJ 4- 0. XJ ID
2 i_ < X
.. o i. 4>
.C Q. "O 4- t/>
x cn a. c in
O 3 ID ป
_J X i O O
^
2
CO
ID
-'
in 4-
>
C L.
>4- O T)
o -*
V
4- Q. <
C E +-
4) 3 O C
u tf> in cn
4> O in 4>
Q. 0 ~ X
03
. 2!
4)
3 I/I
4- U T
^ -^ CO 1 O*
3 +- CT> O 00
u m o. o>
x
oi 10 ซ
< +- C 4-
C/1 ID ID C
E 4>
o 4- 3 e
ID V I IA T-~
L. m oo
4-3 -C C 0> \
a. 4- o> in 10
a 3
ซ/> L. E m i_ <
cn 3 o
3 < * 3 in z
r*- r*ป r^ o
O O* T in
O O O O O
r- f*. r- o
O O O O O
r~ rป- ^ O
O O\ -v \r\
O O O O O
tn o a
^ 4) 4)
4) "O TD *- *- 4-
_ O 4> -0 u
>- 4> 4) 4> ฃ1 A- 3
4- "*"*"*- X X ' O
u A A c .c i.
3 o 4> in a.--
0 X V. Jซ 4-
O * J< X U <ซ- ป-ID C
l_ 4) Ol - O*- 4)
0. 41 O C. ID JC
JD E J= 0 O 4-U)*-.* U
C ~ 4) CO
ID A A A O JD 4) 4) O -C
E o4-.ae.co
C 4) C.
< a. 4-
C
O
4- T3
ID C Ol
1- ID CO
C Q
4) Q
0 CJ
C 1 4-
O in
O L. 3
O O cn
ifl "" <
: in
L.
1 O =
4- U.
< 0 Q
a. ID o
LU U. t
in
in
in
4.
1 4>
C O
4) <-
O 4>
c a.
O >- c
O ID 'O O
4- U) J3 4-
4) ID ID
01 L.
C 4-
H- o O c
j: 4)
ฃ 4- X O
-------
v<1^
(Q
^
Tg
o.
Jฑ
^*
L*^
o
e
O
"~
to
^
Mtm
ex
3?
^^
_
ii
a
^^
^^
a
3
~
_
^
rj
.Z>
^
^
0
t-
*
i
a
0
1
L.
Q
a.
a
c
o
in
ง
^T
ง
in
io
ง
u
U_
S
^
0
U
C
^
O
1
c
O
4-
0
C
ซJ
"a
X
in
a
I
JC
o>
X
ฃ
a
E
in +-
0? S
5
L
4)
4-
^
4- E
3 a
a. c
c a
a.
c
o>
in co
(N C \
CO ro
O* X fN
O X
a
to ซ
CO - 0.
<0 O\ O 3
^ E O
41 * E O
O>
c < .*
ซ* 3 a, U
a O LU O
^j
.. .. L,
JC 4- 0
x O) m Q.
O ~ W ID
-J X ฃD Q.
in in in in
^ ^ ซ
O O O O
8 8 S 8
o d o o
I 8 8 8
d o c> o"
i
in ^
4) 4-
4- jc in
a o> c
L. ~ ID
4) 4- in 0
0 x ID c in jo
ID >ป JC O L. O
4- L. X O O) 10
a. -a
3
d
E
E
^
a.
LU
4-
in
0
o
c
a
X
O
o
"^
.^
^
E
C
0
0
0
ID
in
u.
O
m
>.
^
^
ซ
1
X
u.
m
in
O
O
in
O
d
LL.
tj
CQ
^
0
^
*^
^
m
u.
m
L.
o
u
IO
Lu
c
o
4-
L.
C
0
U
c
O
Q
H
m
c
0
4.
a
L.
^
0
U
C
o
u
jC
in
^
0
"o
X
ง
l_
4*
in
j^
o
o
m
JO
o\
co
CO
= "
u_
Q
O
^
c
a
ฃ5
^w
l_
^3
JO
TJ
0
4-
m
0
en
OJ
in
m
a
c
0
4-
IO
L.
4-
C
4>
U
S
U
4-
0
^
*.
0
en
4-
(^
.1
+-
LU
Q_
LU
ซ
JZ
O)
Q>
^
S
L.
o
ID
L.
4-
C
E
^
V
in
JC
0
LU
1
co
^
ซ
^.
(N
O
4-
l/l
3
U>
O
a.
X
LU
0
C
*"
L.
O
o
0
in
3
in
ง
0
O
4-
ID
L,
4-
0
U
S
8
LU
0
Ul
O
a.
X
LU
,("
LU
5
^^
co
co
5
t-
1/1
u
O)
JC
0
JC
.
4-
ID
E
4-
in
0
4-
m
0
JO
o
0
8
_^
^.
^_
(0
u
Q
"*^
A
Q.
O
L.
C3
tM fM
4- 00 CO
C Ol Ol
0
E
in in in
0 00
in
in L. u
< u- u.
co
CO
o>
~
ซc
a.
LU
0
4-
3 ID
E
c
"~ +
in
a ซ
X Jฃ
0 in
> L.
0 O
i_ r 4-
JC
O en
4- JC 3
o : o-
L. 0
0
4- JC 0
C 4- JO
1 C 0
0 "" 4-
0-0 TJ ซ
M 0 0 *.
-in E
JC 3 3 ซ-
in in 0
in in 0
10 in o
* * "*
*-ป*-* ซ-s
JC JC JC
O) CO O)
0 0 *0
XXX
JC JC JC
in in in
4-04-0 0
ID L. IO U 4- L.
0 U. U. Ol Lu
m CD r ffl x m
0
4-
10
3
er
0
0
JO
it
Q
-a a
II
in 0
S. 5
>o
*
^_
JC
;-
0
^
4-
(O JC
* u>
0
C L.
4) <.
U
O CD
T)
-------
ID
X
T^
ID
Q.
^
ID
U
C
ID
U
O
5
c
O
4-
03
a
a.
X
in
TJ
in
3 < to
C +
O f
4- >*
Q l_
E -a
in c
C ID O
O 4-
U O 4"
4- a
0)
u.
c
i_
- 4-
ui r 3 4- o
O 3
M- E 1- T3
4- O O ~ O
o * jf ซ3 a.
O E X ซJ 4)
4. 4> > a.
S 1- UJ IO
~ < 2. ~ ฐ~
a.
r-in in rOK> OOOOOOOO
ininOOOOino
\O IO O^ ON ON ON NO O^
r-m m >OKt OOOOOOOO
OOmininmOin
O VO GO 00 00 CO NO CO
rปin in ro ro OOOOOOOO
>
4-
~
IO
C
ID O
c m f 10 4- in
4>cm4) otn cc
>* .X ID 1. >-~4> 4)ซ-3ซ*-l_ J3O)4- >* Jฃ4)
C4) o) js 4- 4- 4) ciDc^t-inujOj:
OJD'OJIUOO'OJD'O
oi
3
t *
^
l_
IO
CO
>.
1.
4-
^
c
4>
4-
4-
o
r
4
O
8
o
8,
oT
S
8
O
o
*h
ON
tO
IM
|
oT
CM
c 4)
O 0
4- "O
10 O "O
l_ 4)
3 a. 4- 4>
a. 3 u
o -c .a 3
a. o ~ -o
*- vป O
fc JZ 4- l_
O x in a.
4) t- -0 c/>
N 4) Z
> in
tO O 1
r^
03
ON
.
in
4-
U
IO
^
4-
in
.a
10
u
4-
in
4-
IO
Z
Vfc.
o
4-
C
4)
E
4-
t_
IO
Q.
Q
<
O)
l_
3
n
in
i.
L.
IO
8
ฐ.
8
*r
O
8,
o
*
O
^
1
ซ
4)
U
3
O
L.
a
i
_
10
o
4-
in
4-
10
4-
to
to
3
^
4)
l_
4-
u
<"
CO
O>
in
4-
O
ID
t_
4-
in
-------
ID
j
x;
^
2
^^
u
ID
V
ซ
Q
c
O
4-
IO
Q.
Q.
<
C
a
l
ID
3
^3
>
C
ID
U
Q.
X
~
ID
^
ซ^
U)
a
ID
:>
L.
ซ
i
L.
ID
a.
o
c
ID
m
O
4.
f
in
U>
ซ*
1
u
U.
*
4>
ID
4)
U
C
l_
<0
a>
a:
c
o
4-
10
ID
"a.
X
in
4)
I
f
o>
ir
4>
4_
10
E
in -H
4> in
ffi 4)
X
o
_I
L.
4>
ป-
4)
+ E
3 ID
Q. U
C ID
a.
-t-
i/i
~j
T3 *
< 0 0
Q O
u_ u. 4-
c
-4- +- CD
O 0) E
a.
in Q O
ui a>
>> to >
to o a
ID
4) 'O ป
tr o <<
r in a.
LU
CD fl) c
_
10 oo in
CN O O
o' o o
10 oo m
(N O O
o' o o
to oo in
CN O O
*
000
4)
t- c
10 in to
E C 4) V>
A U)
X ID >. ID
U (. O 1-
^3 O) tf) O)
ป*
O) Ol
O c ซ3
"o ? +- S
a to tu
O Jฃ
ฃ C L. -
4- O to (D
4) Z 0)
24- "O
tO "O 3
4- O C
C (O C/l
0)
E a. c
in a. o a
U) < - Q.
in TI 3 u
in c o
< 10 c
11-3
in t. in
5 M. Q "S
175
-------
a
Q
m
01
_
in
in
Z
*
ID
31
t^
^
(Q
Q.
ID
4)
0
C
O
4-
IO
U
^
Q.
^^
*
X3
e
ID
-i
i
^
ID
o
ป
>
ป
o
C
t^
a
U
a
h-
ID
fe
in
3
a
4)
4)
E
ID
U
ID
a.
c
ID
U)
O
a
. c
l_ O 3 O Ol Z
3 00 ซ <
O 4- 3 O 4- 3 C X
Lao- t- a o- O 4-
*- c
4- C O 4- C O in 3
c c a 4> x x
OO 6 O O E uj t-
o me o me (-
4-mo 4-uio>ป4)
ป 3 L. 4ป 3 L. 4- Q-
4- ซป 4- UI C
o m > 3 u
4-O C 4- O COO
V) fl ^ tij ^^ ^ Z uj Cj **-
O Kl CO O> O
*r CM . c >- r- +.
O 4- O 4- ID OO X
C C l_ Ol UJ
in 3 in 3 3
CO C O 4- >-
4) O 4) O - 4-
4- 4- 3 m C
XX X X U U 3
UJ 1_ LU L. Q
1- 1- l_ 4- O
x 4) x 4> 01 m
4-Q, +-Q.< X 4-
C C -t- 1C
31. 31.T3IO L.4)
qo 6 oc4- 4>oi
o
L. 4) O
Oฃ in 10 4> 4- c
HJ-^4- C 64)J= 301
~ C4) i_ 4- -a
l_4)in T7JZU IOIO C f- J=U 4)4- JZ u. tt
E 4- in e -c +-oa> in c
ง4-C3'OU -t--0 4)'O(O
4>v)4>inc m c
O > ~ ^ C ID ป- 'O in 4- 4) O
ID 4- O O in U4)IOC 4-J=
O J= o o c 01 o-^E4)4-x in
c 4-j= OO a 4)j=c -o
x m o JE j= 4>mi. 10 4) *
u e o ID a ra D c cr 6 ro
ID ncoEin 4-4>j= mo
ini-io 3-4- cj3H-xccc
4)1.3 Xl/14) 0) >- IQQIDIOO
1.4>4-ซ4>C4> O)O 4)4-
4) E in j= O JO C4)J3
t-IDOO X Cซ-4>C 4-OO
* JC4- J3 OO4)IO4- in>ซ-
o i- cm _ ป. ioc
JZOI-34-T3 !/!ปซ in E O 4)
noi()ingO4) c o J= i_ i_ ja
U13 1*-E-^E 4)ini/14)O^|D
OE4)ID 3 4- -4-C O
4-JCO4) 4-V) X >- C IO - +-
U4-l.J34)4>C 4) 4- O "0
O U. J= Ol O -O 10 1_ 4> O
4-M 13 4- O X4-OJO<ซ-ID4>4)
Sป ซ O 4-iD OEซ-E
O x 4- c c .n
ma TJ ปซ. ox dioioiDin
4) X 3 ซ ซ (J O 4) O C 4J ซ3
4-1. 4- 4- 6 ^ L>J=4>JZJ3
ID c 10 E 9 4>au mox4)
4-IOX34-3O JZa4)OJ3lDOL.
I/) T3 O E UI CO 4- H-lDin4-IO4>V>ID
oil ฃ"" 2
ป* ป* *ซ*ซป*ป* o
-
O O *O f% * O O
O O* ^" O* *""
m
TJ in
ID ซ C
in 6 **-
ID C C J3
E Q O J=
c O o mm
ID C 4- m
ID m x **-
O m ID x in o 10 ^
4-OCEOCO E <0
4^ ID 0 4) 0
4- 4) C ^ X C
10 c jo a i- m o i- too
4)UX 4)O) 4-
xom "o'-jSjcoT) o^o
O 4)
*R ^^ *^ ^^ 4- w^ **- xv^
-------
Q.
a
in
in
ซ
*
ID
X
.C
ID
a.
>.
ID
4)
a
o
^_
ID
U
a.
<
^3
c
<0
(
^
(O
3
>
^3
C
_
ID
U
*
^L
t-
10
in
l>
ID
0)
1
a.
^
c
ID
in
c
O
in
in
^^
4-
s
O
oi
4)
^3
(O
'O
u
c
4>
i
JC?
O)
I
a>
ro
E
in 4-
4> in
a 4)
X
o
,_
4)
4-
0)
S 1
a. L.
C ID
a.
OO
4- *
C 1
0) 3 1
o? ID O c t_ in
< 4) i- O O) u
L. +- t_ <
C 3 C V
O ^Q O ^ "* m
^ Cj c > '~
in ป LU 4- 4-
C > C ID
ซ 4- O *. 4-
4- C O <0
X 3 4- 3
LU O 3 ^
O 4- ID
>. 0. l_
4- > O 0) ID 3
C U 0. O 4- 4-
3 l_ C
34) -^
a 41 a in
U 4- ซ X
4- 3 C O
C 4- 4ป +- O
3 in u c
30 L. 3 l_
a. 4) O 4)
a. u a
in a in (si m
41 4- E
i_ o .c a> . TJ O
o .a a L.
c o
CJ 4) ^_ ^
L. a.
>ป 3 CL Uป >
t. in o
a u T>
a. a. ซ
in x>
o o ซ 4>
ซ ซ 4- a
o -o 4- 4>
C C Jt +"
> ซ 3 1.
E O ID 3
a o o E z
09
o
m
03
o
in
09
o
- T>
ซ E 41 "-
ID O "O 01
J3 L. C CT
* g 4t
ซ*. E "- '
O O ID
4) O
4- 0 4) 1-
C 3 O) 3
3 T3 -O 4-
งO 3 in
e ID
< a. in a.
in
i
^j 3
o
ID
l/l I- l_
O) 3
3 < +-
^,
ปป r~ o
Oป -~ iO
in
ป f- O
ov 10
Kl
v r- o
o* *o
^
m ID
T3 .C
\ c
ซ m ID
4- 4)
>. 0) ID C .O
^ ซ e x
o. in j= 5 o
O 3 x u in
L. A
O ~
177
-------
0)
u
01
u
I
3
O
4>
O.
x
a
x
.e
4-
ID
a.
>.
t.
U
C
4>
1_
4>
O)
ID
QL
3
HJ
0)
l_
m
ID
a
IO
C
4- O
C.
4) ฃ
O)
111
ID in
4- (D
_ C
O ID
u S
4) 4>
4- 4-
IO I/)
u>
-
C L.
10 3
Z n
ui
4>
O.
4}
O
O)
l_
3
ฃ*
U)
L.
l_
10
C
4)
t-
3 in
4- 03
O<
3
O
L. "
5
o
a
u
a.
a.
o
c
a
a
3
ID
U
ID
C
ID
a.
x
0
O
8
3
8
(A
> 4>
ฃ
C U)
l_ 3
3 S
1 70
-------
>.
a
O
a
u
o
4>
cc
O)
c
4- CO
O CO
jฃ O\
l_ 'O L.
IO O C IO
oo u 4) 10 T3
S. 4) E 4- 3
IO h- 3 IO C
_ r o o o
ซ/> < Q 4-
O.
01
u> a. c in
3 (O
X 40 I Q
o
O)
10
s
1)
CO
CO I G*
O* O 00
* ^L CT>
X
LU
C
10 (O
e
4- 3
41 X
in rป
in co
.c c u x
o> in >o
3 in in
O c <
L. O
.O U 4) CO
E in L. <
3 O
in z
in
10
ce a.
C T3 -O
ID 3 r 4)
4) Z - 4> -C
'X. u. in
L. 10
c o m c ป-
10 <ซ. o <- *-
Q. 41 IO
3
I
in
c in
4) O 4)
m 4>
m o 10
4> T3 4-
3
m
m
in
i
a.
a.
c
a
o
>
c
O
i
a.
I
in
in
0
a
>
L.
V CO O
O
in
CM ao r-
O
in
L.
a
a.
a
c
ซ
4-
E
in +
9 in
tn ซ
00 O
O
(f\
(M ao
ao oo
to r<-ป
in
in
O
^
O
S f
u >
C L.
* O "O
o ^
V I
3 n
Q. L.
C ID
ex
a. ^
8 8s
O)
6 tn 4)
o f
ซ O m
u
3
O
O
a.
4-
4- 10
O *
4- in ซ- *
c v
4) 4>
U 4- J3 E
L 10
ซ ฃ
a. 4-
V
c o
ซ u
U L.
c ซ
O o.
u
a
ซ in
a
y - C -
10
L.
C 4-
5 *
4- >
a. L.
6 -D
3 -
m c
COO
o +-
U O
. O c
- C. 4)
ฃ 4- ป O
00 C
I. ซ O
U. 4- O O
ex
E c
H- 3 L.
O in o
~ ง ฐ
o i
in jc.
c in ซ
10 i.
-<- L. in o a) *-
IO
T3
>. 10
O u
in
>.
3 ป- L.
4- ซ
in a> 10
(O J3 T3
a.
i
179
-------
>*
ID
2
ฃ
4-
s.
V.
L.
4-
0
Q
e
O
4-
00
u
~_
Q.
a.
o
c
^
4)
in
a
X
LU
4-
i
a
fc
in
a
>
ป
4-
L.
IO
Q.
o
C
IO
in
0
u>
in
CJ>
CM
4>
r\
a
t
o
u
c
4)
0
ฃ
c
o
4-
C
C
a
a.
X
in
4-
*
ฃ
ai
X
4>
4-
a
E
M 4-
Sin
0
u
di
Input
parameti
in
c
c O*ป
03 X
X O
in 10
C CM Q
*\
X <0 ซ-
0 or
r in
Q Q. >- 4-
= 3 4- U
O 3
g L. -O
L. .Ol.
to .* a a.
i f. ~n 0
4> > CX
E L. 0 O
000.
< a.
0. a CO c
LU a. =
o o m
rป pป Ch
O O O
O o m
r* r>- o>
O O O
4-
ฃ,
a
c
a o
IO 4-
0 ซ
J3 01 4-
c o c *-
IO ฃ O 4>
X ">
c
CT*
00
l/l X
c to
(N
X \
o ^o
.
ro Q
ao a.
O\ * 3
งQ
^
4J O
O1 ฃ
C -*
O CC O
>- LU 3!
< t.
JC -K 0)
O) tn a
x m a.
ir\ trป m o IT* ir\ irt ir\ tn
O*OปO*r^o ^ซ^-^^-
OOOOO OOOO
tTt IA ITซ O IA CN CS CN C4
O* Ok O\ f** Ov ^D O ^3 ^3
OOOOO OOOO
1
m
4) 4-
in 4- ฃ in
C C (O Q) C
0 ID L. ซ3
>. j: 4> 04- in 0
u.mu^ฃ 0xiocmA
oj >-in j^ 4>L.ซjx
-ฃO(-O
ofuin^- 4-i_zuoiin
a. -o
o
E
i
cf
LU
ป
4-
in
.
n
CT*
ao
ao
ซ
;
u.
Q
CJ
h
TJ
C
ID
Q
f j
^_
,_
5
.
t/)
~J
^
^
0)
*
in
4)
0)
O)
in
v>
(O
c
0
4-
a
^_
4-
0
U
o
u
4-
ซ
**"
O
4-
O)
c
4-
ID
E
4-
m
LU
LU
"ง>
1
0)
4-
ง
L.
<^
O
4-
<0
L.
4-
C
4)
E
o
4)
in
.c
in
<ซป
0
*~
^
a.
LU
i
CO
ซ
p^
rO
ซ
CM
O
m
L.
3
in
O
a.
X
LU
ID
ฃ
*"
^
o
^,
0
in
3
in
o
g
ID
E
O
4-
ID
4-
0
U
C
O
1
<:
a.
LU
1)
L.
3
Ul
O
a.
X
LU
ซ
<^
LU
I
O
ซ
ao
CO
8
4
I/)
L.
0
i_ r
O O)
4- ฃ
ID ฃ
L.
0
C 4-
0
e c
*o
0 -o
in 0
m
in
in
*** ปป
CM CM
co co
O^ O^
k ป
in m
0 0
i- i_
u. u.
t
10 in
T ซ
4- 4-
ฃ ฃ
cn a>
0 0
X X
ฃ ฃ
in in
4-0 4-0
ID L. ID L.
ซซ->*- *^ ป*-
0 U. U.
00 u
CD CD ซE OP
180
-------
>
fO
3C
J=
4-
O
0.
u
e
^
o
4>
g
^
X
Ul
4-
1
L.
o.
XJ
tz
IO
u>
1
1
V)
*
CM
O
J3
ID
^
4)
U
4)
ฃ
CC.
C
O
ID
C
in
m 4>
u
4)
Q. 1_
C ID
a.
o o
+- 4-
_ _
10
4> 4)
J3 J3
4- -f-
xป in 41
< JB < ja
o
ป *
^ ^.
0) 0)
4) 4)
X X
+"
J= ID ฃ
in >*- in
+- L. C L.
O * 41 *
U
Ol U. U.
6O ฃ U
I o u m
_
-t-
ui >~
Ol Ol
-J O C ซ3
-o oo
o ป- o c +- 01
< O O T3 ID 41
Q O O Jt:
u. u. +- JT c u ป
c *- o * e a c
4- 4) in Q. o 10
ซ3 O Q Ul in u m
Q 41 C O *
u. tr - ce -* Q o
fM O O
O O O
ซ co in
(MOO
O O O
l_
4)
4- in
4- C
ID m ID
E c 41 in
ja in
>. a >ป ID
t. u o i.
xi oi in 01
ซ- c
O 41
E
jt: i. 01 u)
ui O c ui
n- 41 41
Qฃ 4- \O 41 Ul Ul
>. 41 ao xi in 10
>** Ol JC Ol
O 4) 4- 3 * L.
2 ID C OOlO
4) O O W) h- 4-
> 4- ~ 4) 4)
4) C 4- o-~
O4)Ct3iD -~inxi
r E a. J3 a. * E
in < *- 10 E
in u u O <- -
LU ซC _l Q Z LUQ.t/1
.
'
^ K^ ^ IO GO Ol O*
^ ^ o% cปj T o in in
CMr^-ov* ป
f t^\ ^~ ^r 10 oo 01 ^ft
KioปoปCMปOmin
O ซ3 T in 10
CMP'O'T'*' > ซ
_ฃ.
in
c >*- ~
X e ^ 4- o
O O > 4- ID
L. in > ID ซJ>*-4-CC
Ol Ul 41 L. 4-XI H- ซJ 4) IO
C E 4- CLX4- >- ซ- ^ 41
ง~3J^4- E Ol 10 C -ซ- 1- (JJ3J=
ID O) ID 3 3 ^ 41 1- 41 Ol >~in
^ 41 41 O WIXJ=O4)IOOJ=O~
J= Ol E E O. CO1XUJ3XJJZUU1H-
ป* O ซ-
181
-------
IO
~
c
IO
>
^
in
e
1
.
^*
10
X
.c
4-
IO
0.
L.
ID
4>
5
4-
U
CL
<
C
a
i
3
o
'>
o
e
ซ
1
, X
UJ
4-
u>
i
10
t.
o
in
4>
"5
4)
1
ID
a
a.
c
(A
O
4-
a.
ง
u>
A
H-
4>
U
C
4>
4)
i
c
0
4-
IO
C
ID
"a
in
0
i
O)
4-
IO
4- -
in 4-
0 in
4)
4-
4)
v e
3 a
O. L.
C .
ecu
O 10 3
0 X ฃ>
in
0 4> 41
4- +- JC
ซ in
4- IO
ซ/> X X
0
0
ID
0
l_
a
O
01
^
in
o ~
4- ID
O -C
(U
*- u
o
3 -ซ-
4- ID >
ง s ^
0 Z J3
000
4- 4- Jฃ
ID U)
4- ID
V) 3 X
O
O
8
en
c
u
o
o
a.
10
41
10
ป- C
0 b
ปซ u
*- u
O
3 *-
ID O
O) CL
4- 6 (U <1)
O W ฃT Q
ID O) 0
4- ID >. O) t.
C C I. I. 3
O <0 3 34-
U Z -O ฃ)
in m 3
ODD) 0
4- 4- Jฃ V. ~
ID in ^ U L.
4- ID ID O)
in x -x i <
8 8
8 8
in
__
ID >-
E L.
c 10
10 "O
0 O
c c
O O in
0 0 g
ปป ป* o
182
-------
_
t_
tf)
U)
U)
>
2
5
^
Q.
L.
4-
4)
^T
^^
tf
C
"
jT
U
*
CL
a.
^^
_
^
5
**
.
1
^
^~
"
^
c
~
I
LU
in
X
4-
E
4- ^
in v
4) M
CO
L.
^_
^
4- E
3 a
a. L,
c a
a.
ป. 'o
o
3 ID
4) *^ 4- L. *ซ-
D . CD
CO >io
O 4- 3 O 4-
l_ 0. O L. Q.
4- C O 4- C O
ID O 4- (DO
4- a. c 4- a.
c c a, ซ) c c CL
O O E O O
o m c o in
4- 10 o 4- in
4> 3 i. 4) 3
4- in 4- in
ซ> in > 10 m
4- O C 4- O
CO Q. Z LU CO Q- 3!
in ซซ ซ<
O to co
*T (N CM
O Kl 00
0)
c
a
4) u
L. 3
a -a
O
4) L.
o> a.
o
3 ID
4)
m L. 4-
10 ID C
- ^ ^ ฃ 0
V ซ) O X U
O ฃ
* 4- 4-
4" 4) CO 4) CO
ซC ^
ID
3 C >- C >-
O O 4- O 4-
C C
10 3 U) 3
4- 4) CJ 4) O
C 4- 4-
4) X > X >-
E LU U LU L.
C L. L.
O > 4) > 4)
t, 4- a. 4- a.
C C
> 3 l_ 31.
COO O O
ซซ ซซ ป* ป*
o> o o o
o\ o o o
in
ID
U) E
in
>. in 4) t- >
O ID ฃ O O
in a * o in
4- C
C 0
4) CO
< C
4)
c >. r- 4-
O 4- ID CO X
C l_ O> LU
in 3 3
C Q 4- >
4) O 4-
4- 3 ซ C
X >. O U 3
LU L. ~ O
t- l_ 4- O
> 4) o> in
4- O. < >. 4-
C 4- L. C
3 L. TJ ID L. 4)
O O C 4- 4) O)
o
in 4) 4-
1. 4) ซ- O
0-C 10 ID 4) 4- C
ID-ป- 4- C E4)ฃ 3in
C 4) l_ +- T3
O E C U _* 4-ClO 4II-IO
i_d)in -o j; u IOID c m E
4IฃU 4)4- ฃ l_ CL
E 4- in E -c 4-O4) in c
E 4-C3T3O 4-T3 4)T3IO
O > ^E C O 4- XI in 4- 4) O
ID 4- O O in L. 4) ID C 4-ฃ
O ฃ ซ u c b) o ซ- e 3) +- >. in
C 4- ฃ O O O. 4>ฃC XI
>> in u ฃ ฃ 4) in L. ID 4) ป-
O ID O 4- 1_ C O)3IO<*-4-
L. E "O ID O. IDOCO'E ID
a acogin 4-4)ฃ in o
in L. ID 3 ซ- C JD h- > C C C
4) t- 3 >.u)4> 4>>> ID O ID ID O
l-4)4-ซ4>C4) O)O 4)4-
41 E in ฃ o ฃ> ID in c 4) J3
ฃ L. 3 OI 4- U 'OXIOฃ>-'O
4-IDOO > Cป-4>C 4-O4)
* ฃ 4- JO OO4>IO4- U)H-
13 L. C ID <- ID c
ฃ Ol ป 3 4- -O mปซ in E O 4)
ID O) ID in O O I) COฃI-L. ฃ
in D ป- 6 +- e 4) ui in 4) o T3 ID
O E 4) ID 3 4- .__H-C o
4-ฃO4> 4- in X>ป*- C ID ซ 4-
O 4- L. -O 4) O C 4) _ +. O XI
d U. ฃ O) O . 4) ID l_ 4) XI
4- in XI 4- U >>4-Uj3-*-- 4- in eemซ 3
o o ซ .a - 3 amaccin
in a. xi ป* Ox o> o ID o in
4) > 3 4) ซ UO4>OC 4>a
4- L. +- +- g ฅ t->ฃ4)ฃ.a
o c ซ S 8 4> a. L, in o > a>
4- ซJ ป 3 4- 3 O ฃQ.4)Or>lDOI-
co xi o E in to 4- f o in 4- ID 41 m ID
>O Ki O O
SS^S 8
x> m
4) C
ซ- 10
4)
C JD
O S" ฃ
u in in
in
m x ซ-
ID X V) O ID 4-
6 O c o E ID
0 4) 0
C J* >. C
ID - in o t. ID O
JO) 4-
O ฃ 10 -*-
ฃ U XI O XI
O 4)
183
-------
estimates that about 50 acres of agricultural land in that state are amended with sludge. The
Mississippi Department of Environmental Quality stated that the major crops grown on sludge-
amended land are wheat, corn, and soybeans and that a small amount of sludge is applied to grazing
pasture. This analysis assumes that 90% of the Mississippi sludge-amended land is split between
wheat, corn, and soybeans and that 10% is pasture land. The wheat, corn, and soybean percentages
of land are proportioned according to the land use of the two Mississippi counties which receive the
sludge. This data was obtained from the county extension agent from one of the counties. In this
analysis 23% of the sludge-amended land in Mississippi is assumed to produce wheat, 28% to produce
corn, and 39% to produce soybeans. Estimates of the percentages of these crops that are consumed
directly by humans were also obtained from the county extension agent. One hundred percent of the
wheat, 10% of the corn, and 60% of the soybeans are assumed to be consumed directly by humans.
The Pennsylvania Bureau of Waste Management stated that the major crop grown on sludge-amended
land was corn. One hundred percent of this corn is assumed to be animal feed.
Data Sources and Model Inputs for Animals Consuming Sludge-amended Corn and Soybeans and
Percentage of Crop to Each Animal
Specific animals consuming sludge-amended corn and soybeans in Mississippi were
determined through discussions with a Mississippi county extension agent. This analysis assumes
that Mississippi animal corn is fed to beef cattle, hogs, and chickens (broilers). The percentage of
contaminated corn fed to each of these animals was estimated as the product of the number of each
type of animal in Mississippi and the average consumption of each animal divided by the sum of
these products for all animals consuming contaminated feed. These data are displayed in Table 2.4.E.
The resulting assumptions are that 96% of the corn fed to animals is fed to beef cattle, 3% to hogs,
arid 1% to broilers.
Though soybeans grown in the two Mississippi counties that apply contaminated sludge may
be fed to hogs, chickens (broilers), and laying hens, the majority of the soybeans fed to animals are
fed to catfish (Mississippi county extension agent). Since additional information on the percentages
fed to catfish and to other animals is unavailable, 100% of the contaminated soybeans fed to animals
are assumed to be fed to catfish. This assumption will affect exposure results to the degree that
bioconcentration factors (BCF), the contaminated percentage of the animal, and animal product yield
per unit feed differ between catfish and the other animals that may consume contaminated soybeans.
Bioconcentration factors, (discussed below) for hogs and broilers are higher than the bioconcentration
factor for catfish in "best estimates". No uptake rate assumption for eggs has been made but fat
content of eggs is assumed to be the same as the fat content of chicken (Pocchiari, et al, 1986).
Because bioconcentration factors used in the "best estimate" are higher for hogs and chickens than
184
-------
C + Q.
DOE
O t- 3
t-
a
a
E
in
00
CM
O
8
03
oo
in
m
+-
a
3
O
c
a
u
5
a
o
a
O
185
-------
for catfish, the assumption that 100% of soybeans are fed to catfish may understate risk. However,
catfish yield per unit of food (discussed below) is higher than the yields for the other animals.
Furthermore, the BCF for catfish reflects the contaminant concentration in the filet whereas the BCF
for hogs and chickens estimates the concentration in the animals' fat. The fat percentage of the
animal is a small percentage of the total animal product consumed. Therefore, the assumption that
all contaminated soybeans fed to animals are fed to catfish probably results in risk estimates
approximately equal to the risk estimates from a scenario where some soybeans are fed to other
animals.
Data Sources and Model Inputs for Crop Yields
In order to determine the quantity of contaminated food available for consumption, this
analysis calculates the crop yield specific to each state for the planted acres (U.S. Department of
Agriculture, 1985). These numbers are reported in Table 2.4.F.
Data Sources and Model Inputs for Animal Feed Mixes
The level of animal tissue contamination depends on the percent of the animals' diet that is
contaminated. Commodity Maps from the U.S. Department of Agriculture (US Department of
Agriculture, 1982) provide information on these feed mixes. From these data, and from pasture
data described below, the percentage that each type of crop contributes to an animal's total diet can
be derived.
Estimates for pasture consumption were derived from AGDATC data base maintained by
the Oak Ridge National Laboratory. Dividing an estimated national total of 202.7 million metric
tons of pasture consumption per year among beef cattle, milk cattle and lambs, estimates for the
pasture fraction of beef cattle and milk cattle were obtained. Table 2.4.H summarizes national feed
totals for beef and milk cattle, chickens, and hogs. Catfish data are not included in the AGDATC
data base. Instead, a Mississippi catfish feed mill was consulted to obtain the fraction of a catfish
diet that consists of soybeans. According to this source, soybean meal constitutes approximately one
half of the diet of farmed catfish (Delta Western Feed Mills, 1989).
Grazing animals - beef cattle and milk cattle in this analysis - also directly ingest some
sludge due to sludge adherence to pasture grasses. U.S. EPA (1988b) describes various studies that
have attempted to quantify the amount of sludge that grazing animals ingest. Studies have found that
approximately eight percent of forage consisted of sludge 7-21 days after sludge application. EPA
selected this value to approximate the sludge fraction of grazing animals' diets when setting national
criteria for municipal sewage sludge regulations. This value was used in the current analysis for the
186
-------
4>
a
at
10
^^
in 4-
o> e
x ป*
c
^ *"^
O 4-
-C E
0 ~
.* 4-
_ 5
X
JM^
- 4-
0 E
O E
ffl *-
O
Ol CM
1 1
fO 1 1 CM
T
O O
in oo
i I
Ot 1 1 O
O Q O
rซ O >O
1
Oi 1 V (O
Kl
O O O
1 >
ซO 1 Kl CM
r*
o
0
0
u.
C 1.
C 0
i. a >- .c
3l_ ซJ 4-
O X O
o
t_
10
X
i_
s:
0
i
i
I
i
Ji
t
IT*
(N
O
^^
^
0
4-
10
E
4-
0
0
L.
3
4-
in
1 1 ...
1 1 CM O f>
in
(N
^
c
0
E
0
in
a. 4-
Q. 0
3
V)
0 in
c a. c
(0
0 a. 0 4- _i
4- 0 .a 10 <
O 01 > 0 h-
i- -c o -c O
a. oo 3 i-
ed to be approximately 202
4-
c
O
4-
0
3
T3
O
L.
a.
^
'g
D)
c
O
E
IO
0
a
_
>
o
in
in
^3
h-
L.
.
^
0
a.
in
c
O
4-
O
l_
4-
0
E
C
O
'I
ฃ)
_J
O
4-
in
d
o
4-
^x
en
oo
.
ฐ
o
in
c
O
u
in
T3
0
ซ^.
t-
0
_c
4-
o
1^_
o
0
0)
IO
c
c
O
4-
10
4-
o
187
-------
"best estimate" of typical and MEI risk. Ranges of soil ingestion estimates used in this analysis are
from 1.5% in the "low" risk estimate (NCASI, 1987) to 10% in the "high" risk estimate (U.S. EPA,
1988b). In all cases, the percentages of other foods in the grazing animal's diet are reduced to
accommodate estimated dietary percentage from sludge ingestion.
Data Sources and Model Inputs for Animal Yields per Unit of Food Consumed
The estimate of the quantity of contaminated food that is available for human consumption
uses a relationship between animal food intake and animal product produced. Data from the U.S.
Department of Agriculture (1985) provides feed consumption per unit of production in equivalent
feeding value of corn for milk cows, beef cattle, broilers, and hogs. These data are used to obtain
animal product yield for Mississippi and Pennsylvania animals consuming corn. In the absence of
more specific knowledge and information on animal yields from consumption of corn silage, this
analysis assumes that in Pennsylvania and Mississippi the corn fed to animals is grain. The animal
yield data are listed in Table 2.4.F.
Since the corn-equivalent feeding value of grass was not known, the production units of
grazing animals were calculated separately. Sludge is only applied to pasture in Mississippi. Of the
1000 acres receiving sludge in Mississippi, 10 percent, or 100 acres, are assumed to be pasture land
(Mississippi agricultural extension agent). The two counties in Mississippi, applying sludge
agriculturally have a total pasture land of 43,000 acres according to conversations with a Mississippi
agricultural extension agent. Therefore approximately 0.2% of the pasture land is estimated to be
sludge-amended. The two counties have approximately 24,000 beef cows and cattle. Assuming the
cattle are grazed equally qver all the pasture land, 48 cattle would be grazed on sludge-amended land.
Data from USDA (1985) indicates that of the national beef cow and cattle holdings for 1985, 54%
were marketed in 1985. Applying this percentage to Mississippi beef cows and cattle grazed on
sludge-amended land, the estimate of the number marketed per year is 26. Finally, USDA data
indicate that the average beef cattle produces 433 pounds per head which, when multiplied by the
26 marketed cows and cattle, results in 11,258 marketed pounds (5100 kgs) of cows and cattle grazed
on sludge-amended pasture.
To obtain catfish yield per unit of feed, a Mississippi catfish feed mill was consulted.
According to this source, one unit of feed yields approximately one-half unit of catfish production
(Delta Western Feed Mill, 1989).
188
-------
Data Sources and Model Inputs for Uptake Rates to Plant Tissue
U.S. EPA has reviewed studies of plant uptake and has estimated above ground plant uptake
at 2 percent (U.S. EPA, 1988a). This estimate is based largely on the data presented in Sacchi et al.
(1986) and Wipf et al (1982). A more recent EPA memo evaluated new studies to determine the need
for modification to this uptake rate and concluded that there was no support for modification. A
2% uptake rate (dry weight) is used for the aboveground crops for typical and MEI "best estimates"
in this evaluation.
Uptake rates, however, are uncertain. Studies have variously found no TCDD in edible
aboveground portions (fruits and grains) of plants in contaminated soil (Wipf, 1982) (soil level
approximately 10,000 ppt; detection limit 1 ppt); and uptake rates of 15 percent (Young, 1983). In
addition, TCDD and TCDF volatilization and subsequent adsorption onto plant cuticles may be
occurring but not captured in some experiments. In this analysis, an uptake rate of 0.001 is used
for typical individual "low risk" and 0.15 for typical individual and MEI "high risk".
Data Sources and Model Inputs for Animal Bioconcentration Factors
From crop uptake rates and from information on the typical diets of livestock, the average
dry weight concentrations of contaminants in animal feeds can be calculated. The rates at which
animals incorporate these feed contaminants into their tissues are then used to calculate
concentrations of the contaminants in meats or dairy products. Bioconcentration factors for animals
are listed in Table 2.4.F for typical individuals and 2.4.G for the MEI. The uptake rates for TCDD
and TCDF are assumed to be equal to each other for all animals except fish.
Beef and milk fat uptake rates for halogenated hydrocarbons are reviewed by Fries (1982).
Fries assumes that the disposition of ingested residues is the same regardless of the dietary component
that contains the residue. He evaluates information on the relationship of dietary residue to the
resulting product residue and applies this relationship to all routes of dietary exposure.
Fries reports that a given intake level of contaminant will product similar steady-state residue
levels in the milk fat of dairy cows and the body fat of non-lactating animals. He suggests an uptake
rate of 4-5 times the dietary concentration (plant dry weight concentration to animal fresh weight
concentration) at all dietary concentrations for milk fat at steady state (reached within 60 days).
Several long-term studies involving cattle and sheep fed halogenated hydrocarbons indicate a steady-
state uptake rate of 5-6 times the concentration in the diet (plant dry weight concentration to animal
fresh weight concentration) with those compounds that produced the highest tissue concentration
relative to diet concentration. Cattle study durations were 712 and 476 days, with steers found to
189
-------
reach a steady-state concentration in about 280 days (Bovard et al, 19^1 and Rumsey and Bond,
1974). Hog fat and chicken fat bioconcentration factors are assumed to be equal to beef fat. This
analysis assumes a bioconcentration factor of 4 in the typical and MEI "best estimates" for all animals
except catfish.
"Low" and "best" bioconcentration factors for fish based on dietary intake are taken from
U.S. EPA (1989h). The mean fish BCF for TCDD from contaminated dietary sources was given as
0.0967 (whole body, wet" weight basis) in the EPA memo. This BCF was based on seven
measurements. U.S. EPA referenced one study of bioconcentration of TCDF from dietary sources.
This study measured bioconcentration for a warm water species as 0.1538 on a whole body, wet
weight basis.
Since contaminated sediment can result in fish being exposed to a contaminant through the
dietary pathway, fish to sediment ratios also estimate bioconcentration from dietary sources. U.S.
EPA (1988a) reviewed a variety of studies of fish to sediment ratios and,observed that the ratios
typically range from 1:1 to 10:1. A dietary bioconcentration factor of 10 is used in this analysis for
the typical individual and MEI "high" risk estimates for both TCDD and TCDF. To estimate fish filet
concentrations from whole fish concentration, bioconcentration factors are multiplied by 0.5 (U.S.
EPA, 1989h).
t
Data Sources and Model Inputs for MEI Dietary Consumption
Since the methodology for calculating typical population exposure computes the quantity of
contaminated food available for consumption and then distributes this food over the consuming
population, it is not necessary to estimate the percent of dietary consumption which is contaminated
with sludge. The MEI analysis, on the other hand, estimates exposure as a percent of MEI
consumption and requires dietary information.
For the MEI the estimates of dietary consumption are drawn from the U.S. EPA's Office of
Pesticide Programs' Tolerance Assessment System (TAS) dietary data base (U.S. EPA, 1987a). The
TAS data base contains statistics for average daily consumption (in fresh weight grams per kilogram
of body weight per day). For estimates of MEI exposure, the consumption data for a non-nursing
infant are used. This level of consumption is unlikely to be maintained over a lifetime and will yield
a conservative estimate of risk. However, the TAS database reports average consumption rates for
several age groups and does not represent individuals with unusually high consumption rates per unit
of body weight. Therefore, the conservative method described above was chosen to compensate for
outlying high individual rates of consumption that are not represented by TAS data.
190
-------
As a less conservative alternative to the consumption rates of a non-nursing infant, the
consumption rates listed in TAS as the average consumption for the U.S. population (48 states, all
seasons) could be used to estimate risk. Replacing non-nursing infant consumption rates with U.S.
average consumption rates decreases the estimates of MEI risk from the dietary pathway (land
application disposal method) by approximately a factor of two.
The consumption rates used in each case are listed below.
non-nursing U.S.
infant average
(mg/kg/day) (mg/kg/day)
wheat 1031 1411
corn 693 408
beef fat 191 372
dairy fat 1424 429
pork fat 146 208
chicken 508 379
soybean oil 1359 322
fish 59 253
The TAS consumption value used for soybeans is that given for "soybean oil" since no value
for "soybeans" is given. However, soybean oil consumption probably approximates soybean
consumption since most human consumption of soybeans is in the form of oil. Since TAS does not
break down fish consumption by species of fish, the analysis assumes that all fish consumed by the
MEI is catfish. Beef, pork, and dairy consumption values include only the animal product fat, since
the TCDD concentrations are calculated for animal fat. Pork, beef, and milk fat consumption values
are given in TAS. Chicken total consumption values are taken from TAS and then reduced by
multiplying by percent fat (NCASI, 1987). Since the fish BCF calculates the concentration in muscle
(filet), no adjustment is made to the fish consumption figure.
Data Sources and Model Inputs for Animal Product Fat Percentages
As discussed above, chicken fat percentages are needed in the MEI estimate to convert chicken
consumption to chicken fat consumption. In addition, animal product fat percentages are used in the
population exposure estimates to convert from total animal production units grown on contaminated
land to animal fat production units grown on contaminated land. This conversion is necessary since
bioconcentration factors reflect the concentration in the animal product fat, except for catfish.
191
-------
Kimbrough et al. (1984) gives beef fat percentages at between 8 and 12 percent; pork fat
between 6 and 8 percent; and milk fat at 4 percent. The National Council of the Paper Industry for
Air and Stream Improvement uses a milk fat percent of 3.7% and a beef fat percent of 12% (NCASI,
1987). The State of Wisconsin Department of Natural Resources suggests 12.6% beef fat, 4% dairy
fat, 8% hog fat, and 10% chicken fat. The animal product fat percentages used in this analysis are
listed in Table 2.4.F. for typical individuals and 2.4.G for the MEI.
Data Sources and Percent of MEI Diet that is Home-Grown
MEI consumption estimates are adjusted to reflect the percentage of the diet which is grown
on sludge-amended land. The MEI is defined as a rural farmer applying sludge to all crops that are
assumed in this analysis to be grown on sludge-amended land. The farmer is also assumed to feed
the crops to all animal species that consume contaminated food according to this analysis.
The MEI is assumed to obtain only a portion of food consumed from her or his farm. U.S.
EPA has estimated the percent of annual consumption which is homegrown for various foods for
rural farm households (U.S.D.A. 1966, cited in U.S. EPA, 1988b). These data are used in this analysis
and are presented in Table 2.4.G.
Data Sources and Model Inputs for Bioavailabilitv
Human exposures are adjusted for bioavailability of the contaminant via ingestion. The
bioavailability of 2,3,7,8-TCDD for fatty or oily foods and for other foods is estimated by U.S.
FDA (1989). This analysis uses the ranges of bioavailability given by FDA for the "low" and "high
risk" estimates and the midpoint of the range for the "best estimate" in the typical individual
scenarios. The high value given by FDA is used in both the "best estimate" and "high risk" MEI
scenarios.
Data Sources and Model Inputs for Population to which Contaminated Produce is Distributed
To determine typical exposures, total available quantity of contaminated food is divided by
the population across which the crop or animal product is distributed. Though this calculation
indicates the population at risk, the distribution population will not affect total population cancer
risk (assumed to be a linear function of dose) since the total pollutant dose has already been
determined. According to the county extension agent in one of two Mississippi counties applying
sludge to agricultural land, the grains from the counties are distributed nationally. The Mississippi
Cattle Industry Board stated that beef cattle are also distributed nationally. Catfish, chicken, and
192
-------
hogs produced in Mississippi are assumed to be nationally distributed. The only sludge-amended
agricultural products from Pennsylvania are dairy products. Pennsylvania dairy products are generally
distributed throughout the New England and Mid-Atlantic regions according to the Harrisburg
Department of Agriculture. The estimated sizes of the populations to which contaminated animal
products and crops are distributed are shown in Table 2.4.F.
2.4.3 Estimates of Exposure and Risks from Direct Ingestion of Sludge
Direct ingestion of soil can occur when sludge is applied to sites where people may live and
work, such as a family farm. To model the risks from the direct ingestion of sludge contaminated
with TCDD and TCDF, this analysis adapts a model developed by Hawley (1985) which accounts
for differences in exposure to indoor and outdoor concentrations of soil contaminants. Children
ingest far more soil on average than adults; however, adults may also inadvertently ingest soil that
adheres to food or cigarettes.
Description of Calculations
The calculation of risks from direct ingestion of soil is straightforward. First, the soil
concentrations outdoors and the dust concentration indoors are estimated. The outdoor contaminant
concentration is multiplied by the quantity of dirt consumed outdoors, while the indoor contaminant
concentration is multiplied by the quantity of indoor dust ingested daily. Risk is estimated based on
the daily quantity of soil and dust ingested, the gastrointestinal absorption of TCDD and TCDF from
soil, and the cancer slope factors of TCDD and TCDF.
Description of Calculations for Estimating Exposure
The concentrations of TCDD and TCDF in outdoor soil are estimated as described in Appendix
A. To obtain an estimate of indoor dust contaminant concentrations, the following calculation is
performed:
where :-
Cin * concentration of contaminant in indoor dust, mg/kg
_ ฃ,^ .. concentration of contaminant in outdoor soil, mg/kg
F = ratio of the contaminant concentration in indoor dust to the contaminant
concentration in outdoor soil
193
-------
Once the indoor dust and outdoor soil contaminant concentrations are computed, the daily
dose of contaminant is calculated for persons in three age groups: young children (ages 1-6), older
children (ages 7-11), and adults (ages 12 and older). The daily dose is calculated as:
DOSEg = [(C0 DCg FBfOUt) + (C-n DCg Fg>J] ABg / BWg
where:
AB = systemic absorption rate from gastrointestinal tract (expressed as a fraction)
BW = body weight of individual in age group g
CQ = concentration of contaminant in soil, mg/kg
Cjn = concentration of contaminant in indoor dust, mg/kg
DC = daily soil ingestion rate for individual in age group g, g/day
DOSE = daily dose to individual in age group g, mg/kg/day
F_ ,_ = fraction of ingested soil from indoor sources, adult
g, in
F QUt = fraction of ingested soil from outdoor sources, older child
First, for each age group, the concentration of TCDD or TCDF in outdoor soil is multiplied by total
quantity of soil ingested each day and by the fraction of ingested soil from outdoor sources for that
age group. The same calculations are performed for indoor dust ingestion. The total daily quantity
of ingested soil-bound TCDD or TCDF is the sum of the indoor and outdoor quantities ingested. The
model then adjusts the total quantity of ingested soil-bound TCDD or TCDF by the fraction absorbed
into the system through the gastrointestinal tract, and divides by the body weight of an individual
in that age group to obtain an average daily dose in mg/kg/day for that age group.
The weighted average daily dose of contaminant over an individual's lifetime is calculated as
the sum of the daily doses for each age group weighted by the fraction of the individual's lifespan
spent as a member of that age group, as described in the following calculation:
where:
DOSEavg= SFg DOSEg
DOSEavg = average daily dose over lifetime, mg/kg/day
DOSE = daily dose for individual in age group g, mg/kg/day
F = fraction of an individual's lifetime spent in age group g
194
-------
Description of Cancer Risk Calculations
Once the daily dose estimate is obtained, it is combined with the cancer slope factors of TCDD
and TCDF to obtain an estimate of lifetime risk from direct ingestion exposure to these contaminants.
The calculation of individual risk is:
1C = DOSEavg q/
where:
DOSE = weighted average daily dose for an individual, mg/kg/day
1C = individual cancer risk over lifetime from DOSEavg of TCDD or TCDF
q.,* = incremental lifetime risk per mg/kg/day dose of TCDD or TCDF
Individual cancer risk for a typical exposed individual is converted to annual total population risk
(in cases per year) by multiplying the number of persons exposed by the individual risk and dividing
by the average person's lifespan, as described in the following equation:
PC - 1C POP / LS
where:
LS = average lifespan of an individual = 70 years
PC = population risk, cancer cases per year
POP = population exposed to DOSEavg
Data Sources and Model Inputs
The values used for each model input for "low risk," "best" and "high risk" typical exposure
estimates are summarized in Table 2.4.1. The values used to derive the MEI "best" and "high risk"
exposures are found in Table 2.4.J. The best MEI exposure estimate is derived by combining
estimates of behavioral input parameters with the best estimates of physical/chemical properties of
TCDD and TCDF. The "high risk" estimate of MEI exposure uses the same behavioral inputs, but
combines them with the high estimates of physical and chemical parameters of TCDD and TCDF.
The following sections describe each input and documents the data sources used to derive the
values for the parameters for both the typical and MEI analyses. Where parameter input values differ
for the "best" and "high risk" MEI exposure estimates, these differences are discussed. For those
behavioral input parameters that do not vary between the "best" and "high risk" MEI calculations, a
single value for the MEI analysis is discussed.
195
-------
X
"O
c
^
ID
U
"a.
t
0
o
in
4)
3
IO
4)
4-
i
IO
ID
a.
>o
c
ซ
in
S
1
in
in
.
*
CN
4)
ID
1-
4)
0
C
4)
L.
4)
4)
SX.
C
O
4-
C
IO
a.
X
in
i
j=.
ir
ซ
4
ID
^_ _
in 4-
4) in
03 41
X
o
u
4)
+* 6
3 ID
O. C.
C ID
O
0)
T u r
c -
O -a in
in -o o
00 4-
ON 3 T3 L.
o a: 4) o
4) 4- 00
> C L. CTi
10 0 O
4- c a.
O 4- -
4) u. in c r-
L, 4) O CM
E O) 4-
Q c in x
L. C l_
Ct 4) S3
m 4- 3 3
3 C - C
co o re
Or co -5 5
14
0
4) O
O)
C
10 C
u, O
4) 4-
4- 4)
U) C
OH
LU
3 O
to m
O
x O
c r
4) in
* 3
O)
a
CN >
o i i
o s.
0:0
CN
O
CN
O
^_
0
*
^
ID
t
O)
"o
in "a
4)
V 4-
c in -o
3 4)
O O)
O 0 1/1- in - ui -
<*- -*- QJQ. fli r> q\ rt
L. 3 L. D l_ 3
*!) QJ 3*O ^ O 3"*O
T u T o in *c L. in o> otraicTv OI'-E OI"-E 01 " E
4) 1-00 0) -t-co cam cQm cam
> c i_ o^ > ~ c L. c^ . *~ (~^ i/i *^ Q in LJ i/i
-u_incr- 4>u.incr- oo < ->- co < .-oo<
U 4)OCN L. 4)OCN 4-ป IO 4-ป (O *-ป 03
E O) 4- e O) 4- inr-4)i- i/ir-4)!- mr4)L.
Q c in >. Q c m > uj-i-Q Lu-i-Q UJซI_Q
fy Q) to Qฃ Q) *^ ."Q in * i/i * ui
LU4- 33 LLJ4- 33 -CNOCO -CNQOO -CNOCO
3C C 3C C < Q.O3 < Q.CO < Q.OO
co ^ O 10 co "- O !Q d. O x o* n o x C7\ f^_ o x OS
O S CO "5 > O=CO->~> LU4-UJ LU4-LU LU4-LU
4) O
X T3 4) ฃ
O \ 0) O C 3 3
Cnc 01 O 4) -4) ซ4)E
4)1- 4) CO > L.IOC <4- <4-
O) 4) V. t) l_4-4)>0) O. ID E CL'OX
C > 13 L. 4> lOUlX L. UJEO LUEIO
IO4)O ซ1 E4)ฃ'O -!_ E
I_X 4)3 EO14)OI . 4- ป. -4-
O-ฃ ฃ 3 c m in in
4)^4)O 4-*O in - .O jC .C 14)4) 14)
ฃ O) 4) ID O *T 3 T
4- - C L. Irt4- in 4-4) E S4)
C IO 4) am X .CL. 4)OIO 4>O3
1(1 4) L, 13 ซ OOmt 4J L. > L.
o in o LU 01 3 in j: 10 *: io>
CC. 33 4- O ป COI h-C
uj f L. coma co o m 4) 4) x:
3ฃ4-O O > 03 C D L. EJCf E ^ O)
coo**- 3 om mo 010 oio~
O >-co4) 10 i_4-ซ L. 4- x:
i_O4) J3O4- in a 4- c
> o 4" 'o * >* 4)c4) in4)m m*4)
ฃ) T31O C--E < 4-4)3 13 L.4- "O C
CE 4)in Q-lOCO 4) 13 4) 4>-Q
c 4) > ^ ^ 1 1 1 c 4) L, IO ^4) 3 > 4) L_ *^
4)01 4- in IDl/14)> 3 TJ 3TJUI
>4)i_in o)34) E OJCL L. x:io L. in
34)4) TJ O > L. E 4JIOO 4)IO O
o x ID 4- i_ in a. o o -a > L. 13 > -c a.
IDO4- U! CO 4- L. L.O O
CN > Ul O L. 4) OIU) 4) X 4) 4) X 4)
4) OA 4) "O O) O Ol O) O L. O)
O 4> 4) 13 O 1 * 3 4) C IDJC4) IO _J 4) IO
OJJ= 4-m ป 4-4) 4-34- V -O4-
004-u. inn >. ID L ^ c OID c c
4-L. 3 ซ ซซ>4>IO 4) >- E 4) O 4)
4) O O O 3 CN >4- OOO ~ OCO O
> oiom 4)OiO4>
0=310 o > o ป i ป- o ~ a. <>> 0 a. - Q.
in
CN CO Ol O
d o c> o
in in
CN ป r~ fo
o in oo O*
do d do
in in in
8>ป ป r>
in in co
d o " o o o
. >. >. x
x x r> u 4-
(O ID E E 4) S
O "D IDO 1OOT3 IOO3
>X \ "O u, ฃ T3I_ OU.'O
O) 'a O) O O ID
'o- "o- "Sc- "oc- 'ocซ
tn 13 -c m-o om Om om
4)O 4) + L, 4- L, 4- L.
4-H- 4-4- C4-Q C4-0 C4-Q
cint- c 01 4- 4>ino 4) in o 13 D 01 O
OOl'a OOI3 L.OJ4- U.CD4- L.1714-
gc E C 13 4>C3 4) C 3 ฃ 4) C 3
<_o < ID Q. o a oo a. o
196
-------
(D
2
.C
*
Q.
C
O
^*
01
0)
ID
L.
Oi
0)
to
u
ID
Q.
0
c
ID
U)
O
~
4-
Q.
in
in
^
4-
O
^^
.
ป
O
C
4>
t_
03
*^_
cr
c
O
4-
10
C
a
mm
a.
X
O
ซ
i
O)
X
4>
4-
ID
in 4-
4> in
ffl ซ
x
Q
"*
41
4-
4- E
3 ID
Q. L.
C ID
a.
in
.
o:
.c
S
X
c
10
E
T
~
*
a:
Q
k
<
a.
UJ
,^
o
c
O
4-
IO
3
ซw
a
^
4)
C
T3
41
in
3
in
X
^^
O
^B
O
3
a
O
~
0
0
0
o
_
"o
in
ซ^
0
4-
C
1
13
Q.
U
C
3
S
L.
o
4-
C
g
m
in
4)
<
X
O
4)
O)
3
^_
m
4>
D)
IO
X
41
m
a.
o
c
3
S
t^
0
reuse
ง
*4_
o
4)
4-
in
0)
. l
O
t/i
*-
^4_ >~
(U 1-
c O
0) 4-
m itj
.^
- 3
cn
in tr oo
o o^
a. 41
in >
ซ >. ~
Q 4-
10 in
4) c c
O! l. O
T3 41
34-4-
Q.
co < O
^^
^
ID
l_
4-
ซ
jQ
c_
ID
4)
3
IO
>
ฃ
Ol
o
c
a
tj
c
4)
1
CJ)
O X3
3 C
IO
in
^
ao
.
.
E
D
10
_C
U
CO
..
X
o
O5
c
4-
[Q
e
4-
in
UJ
c
.V
a
4-
10
g
4-
m
4)
a
o
4)
m
3
4)
3
a
5
in
O
in
^^
O
S
O
o
4-
a.
10
o
^
.*_
._
-Q
T3
ia
O
CD
..
JZ
O)
^
ป
4-
lO
a>
CD
4)
4-
ra
E
4-
in
4)
4-
m
4)
^j
2
.ซ
t-
i
^D
ซ
r*
^
fM
O
4-
in
i.
3
in
UJ
t _
O
in
0
4)
ID
L.
o
C
a --
^^
S u
CJ XJ
i r
00
r*^ (i)
* o
rO C
* >t?
CM 4-
"O JO
4) 3
4- CO
m
4) -0
O) 4)
C 4-
10
~O QL
X
4)
.Q 1)
T3
41 C
4- O
in
4) T3
O) 41
o> in
3 a
in j3
o
O) &
C 3
ID 0
0>
O L.
I- O
41 4-
3
a ~
> JD
ID
Ol
~ ID
\\
a as
.
<^
a
i_
o
o
CO
c
^
.
u
.ป.
o
E
0
in
4>
o
3
4-
in
0
in
co
^.
M
>.
0)
X
03
II
8
c
M
ง
^y
u
u
-ง
c
c
O
4-
in
co
fe%
S
^^
in
ฃฃ
O
U
C
3
t^
in
ID
in
i_
8
a
c
8
T3
4-
3
O
1
4-
0
L.
8
^3
4-
3
O
o
L. 4-
SI
V C
3 4)
O U
*- O
0 U
197
-------
^
c
O
4-
CO
U
*~
^
a.
^
^
c
)Q
1
^B
K)
3
o
^
^
C
T3
O
V)
^
^J
V
4A
0
5
^L
5
4"
O
m
4)
3
ID
^^
4)
t_
ID
a.
T>
a
in
S
4-
^t
u>
0)
-Tป
*
^
CM
4)
Z
ID
^^
r"
4>
U
C
4)
L,
4>
*ซ
4>
ce.
c
o
4-
IO
c
a
a.
X
V
V
in
4)
4-
i
^
O)
X
4>
4-
ID
E
in 4-
4} in
m ซ
i_
4>
4-
Input
parame
i_
0
4)
t u r
c
O 10 m
in TJ 4)
oo ป-
Oป 3 "3 \-
o o: 4) o>
4) 4-00
> C l_ O>
a o O
4- C Q.
O 4- ซ
ซ u. m c r-
L. 0) O (N
E O) 4-
O c in >ป
1. C L
Qฃ 4) 10
LU 4- 2 3
2 C C
-)->
l_
o
ป^
c
O
4-
in
01
c
"o
in
L.
O
*"
ซ
3
*a
>
c
4- 41
(ft 1.
TJ
ฃ
Oป
X U
OO
o
OO
*
0
>.
o
TJ
s.
O)
8 TJ-
4)
4- V
c in TJ
34)
Q O)
E C ฃ
4-
U
4)
L.
a
ae.
LU
2
l/>
0
L.
O
c
O
4-
in
ซ
O)
c
^
"o
vn
u
0
"*"
ซ
3
ซ
>
4-
in
ซ
ฃ
O)
X
00
C5
00
O
"o
in
Amount
L.
O
0)
u :
c
ซ3 I/)
TJ 4)
-*- ป
3 ID l_
CD CC. O
4- 00
C L. O^
ID O O
c ~ a.
4-
u. in c r*ป
4) O CM
E O) 4-
c in >.
L. C (_
4) 10
4- 2 3
c c
O 10
: i/> ~ป ~a
ซ
c
4>
L.
TJ
, j
ฃ
U
A
>ป
ID
TJ
x
O) TJ
TJ ฃ
4> U
4-
m L.
4) 4)
O) TJ
S o
^r
0
l/*t
0}
&
4)
>
4-
U
4)
L.
Q
tr
LU
2
V)
O
a
in
4-
3
TJ
a
u
o
0
3
4-
m
4>
.C
O)
ฃ
in
a
e
ซ
>
Ol
_
0
o
o
o
*
o
'o
m
4-
c
3
^
l_
0
4)
u =
c
10 m
TJ 4)
4_ &
3 ra i_
o tr 4) a>
4- OO
C L. Ol
ID O O
_c a.
u. in c rซ
.
L. C L.
4> 10
4- 2 3
C C
O 10
= !/)>>
>.
ID
TJ
\
CTI
A
TJ
4>
4-
in 4-
4)
O) 3
C TJ
ID
in
4)
O
l_
V)
(_
1
4-
o
ง
I.
ป-
o
in
m~
ซ
ซ
in
I
3
in
in
ซซ
8
vซ.
8
>.
TJ
ID O
T> ซ- ฃ
*- U
"5 c -
O m
4- L.
e 4- O
4> Ul O
U 4> TJ
t_ O) 4-
& 2 8
in
4)
o
l_
3
O
in
c
8
TJ
4-
o
ง
l_
'o
in
_
^_
ซ
m
i
3
in
ID
M
8
M
8
>.
ID
TJ
"o
4-
Percen
i_
SO)
TJ
i o
C
O m
in o TJ
4) TJ
0) 4-
C 3 .C
0 0
^
I/)
4)
U
t-
3
O
in
i_
1
4-
3
O
ง
l_
tfc.
0
l/>
i
^
^_
c
in
i
3
in
V)
M
8
ป*
8
>.
ซ
ID
TJ
"o
4-
Percen
4-
ง 3
l_ TJ
*- ID
C -
O in
in O
4) TJ
O) 4-
C 3
O
198
-------
>ป
(Q
2
f-
t-
CO
CL
ง
4w
in
ป
en
c
[
O
v>
o
4-
ซJ
(J
^_
a
ex
o
c
tj
_i
i
ซ
(O
3
^
>
^
^
^
D
i
X
lu
4.
VI
o
i
o
4-
b
***
u>
4>
ID
>
U
U
4-
l.
ฃ
o
c
ID
ซn
0
4-
Q.
3
in
in
-^
ป
fj
v
J3
ฃ
O
u
0>
u
0)
OJ
cr
/
c
0
4-
O
e
ซ
0.
X
o
N.
in
ซ
4-
3
JZ
0)
X
4>
^
ID
in 4-
ซ in
ffi ซ
OJ
0)
S 1
Q. V.
C ID
O.
ID
0)
l_
ID
TJ
4)
o
w
O)
*o
3
^
V)
ง
L.
'o
in
a
in
3
in
in
O
0
-o
E ซ
o 5 -a
ป- E
O TJ ID
O 1
4- 4- 4>
c in o>
3 ซ T> T>
O O) 3 C
E C ID
< in
4-
-4- T) ซ- <
O c a Q
ID l_ U.
>- T)
4- Q
a= i-
O 0)
J Ul >.
r> i D o
ID CO U ffl
c
f^ ro in >
ID ป J3
O E
CO TJ O
r o -a L.
4- D O
O> E
U C
ซ
>
> o
J3 I.
TJ C
aj o
^B
VI TJ
a ซ
en in
oi a
3 ^3
in
a a.
en 3
c o
ID L.
L. Ol
Jf
Sl_
O
U X
^ป
in 4-
ซ
3
o a in
> ซj ซ
4- T>
in o 3
a > 4-
j= a in
en o
I ฃ O
O
r-
O
O
r-
O
,
li
4- in
a.
L. 1-
g 5
jO
ID 0
O
O l-
tn
CO
Ot
>.
a>
z
ID
X
II
1_
8 1
TJ ป* 4-
c in 3
co o
H
8 1
TJ ป* 4-
C O 3
~ CO O
C
c o
0 _
4- 0
a c
C 3 C
4- t- O
C
4) in i_ 4-
O ID O ID
C 5 L.
O in TJ 4-
O L. 4- C
83 0)
O 0
T) C
O c i- O
O
-------
Data Sources and Model Inputs for Soil Concentrations
The methods for deriving the soil concentrations for each land application site are presented
in Appendix A. The resulting average soil concentrations over 70 years are displayed in Table 2.4.A.
Data Sources and Model Inputs for Soil Ingestion Rates
To be consistent with other EPA program offices, the typical and MEI analyses uses the soil
ingestion rates-found in the memo "Interim Final Guidance for Soil Ingestion Rates," (EPA, 1989a).
This interim final guidance memorandum gives a suggested range of soil ingestion rates for children
of 0.1 to 0.2 grams per day, with a maximum of 0.8 grams per day. The guidance memorandum based
the suggested soil ingestion range for children on studies by Binder et al. (1986) and Clausing (1987);
both of these groups of researchers studied soil ingestion in children with the use of tracer elements,
such as aluminum, silicon, and titanium. The guidance memorandum suggests the use of 0.2 g/day
as a best estimate of daily soil ingestion for children. The typical exposure analysis follows the
guidance memorandum and uses the range given for young children, but uses as a best estimate of
0.1 grams/day for older children. The MEI analysis uses an ingestion rate of 0.8 g/day, as suggested
by the guidance memorandum.
The OSW guidance memorandum gives a range for adult soil ingestion of 0.001 to 0.1 grams
per day, based on work by Calabrese et al. (1987), as cited in the OHEA Draft Exposure Factors
Handbook (May 1988). In accordance with OSW policy, the typical exposure analysis uses 0.02
grams per day as the best estimate of daily soil ingestion by adults, while using 0.001 and 0.1 to
represent the low and high estimates. The MEI adult is assumed to ingest 0.1 grams per day.
Data Sources and Model Inputs for Indoor Dust Contaminant Concentration as a Function of Outdoor
Soil Contaminant Concentration
This analysis assumes that the concentration of TCDD and TCDF in indoor dust is related to
the concentration of TCDD and TCDF in outdoor soil. Values for this parameter are derived from
Hawley (1985), who assumed that indoor contaminant concentrations in dust were 80% of the
contaminant concentrations in outdoor soil. The typical exposure analysis uses a value of 80% as a
best estimate, and applies a range of 75% to 85% as the low and high estimates, respectively. The
MEI "best" estimate calculation uses a value of 80% for this parameter, while the "high risk" MEI
estimate uses a value of 85%.
200
-------
Data Sources and Model Inputs for Fraction of Soil Ingested from Outdoor and Indoor Sources
The fraction of soil ingested indoors and outdoors is multiplied by the total daily ingestion
rates used by EPA (1989a) to obtain the quantities of soil ingested indoors and outdoors each day.
To obtain an estimate for this input parameter, the typical exposure analysis relies on information
from EPA (I988a). EPA (1988a) presents a summary of the work of Hawley (1985), who estimated
the dust/soil quantities ingested indoors and outdoors from the dermal soil contact rate and from the
area of skin that comes in contact with food, cigarettes, or objects mouthed by children. Although
the Hawley (1985) estimates of the total grams per day ingested differ slightly from those used in this
analysis, the summary of these values presented in EPA (1988a) is used to estimate the relative
fraction of total ingested soil contributed from indoor and outdoor sources. For all age groups, a
larger fraction of the total daily quantity of soil ingested comes from outdoor sources, due to the
larger dermal soil contact rates outdoors that lead to larger ingestion rates. Young children have the
largest proportion of total soil/dust ingestion from indoor sources. As a person ages, the relative
proportion of soil from outdoor sources increases, due to the decline in the amount of dust ingested
while indoors.
The typical exposure assessment uses the data for each age group from Hawley (1985), as
presented in EPA (1988a), for calculating the "low risk" and "best" estimates of typical exposure.
For the "high risk" estimate of typical exposure, the fraction of soil ingested outdoors for young
children is assumed to be the same as the "best" estimate for older children; similarly, the value used
in the "high risk" estimate for older children is the same as the value used in the "best" estimate for
adults. For adults, the "high risk" estimate of typical exposure assumes that all of the soil ingested
each day is from outdoor sources. In the MEI analysis, for all three age groups, the total quantity
of soil ingested by the MEI is assumed to originate from outdoor sources.
Data Sources and Model Inputs for Fraction of Ingested Soil from Contaminated Area
The sludge-amended land application site is only one among may potential sources of ingested
soil. Soil may be ingested at locations removed from the contaminated site, such as a playground or
an outdoor workplace. The fraction of the total quantity of ingested soil originating from the portion
of the yard or farm treated with TCDD- or TCDF-contaminated sludge may be quite small.
However, there is no information available regarding what this fraction may be. In the absence of
data, the typical exposure assessment assumes that 10% of the total soil ingestion daily is from sludge-
amended land, and uses an arbitrary range of 1% to 100% to represent the low and high estimates for
this model parameter. The MEI analysis assumes that all of the soil ingested by the MEI originates
201
-------
from the sludge-amended area. These assumptions are the same as those used in the analysis of risks
from direct soil ingestion from land treated with municipal sewage sludge (EPA, 1989b).
Data Sources and Model Inputs for Absorption Through GI Tract
Absorption of TCDD and TCDF through the gastrointestinal tract has been studied using a
variety of media. Absorption will be influenced by how tightly TCDD or TCDF binds to the matrix
in which it is ingested. Poiger and Schlater (1980), as cited in Schaum (1984) reported that the
gastrointestinal bioavailability from soil in their studies was 20% to 26%. In a recent review of the
literature, FDA (1989) discussed an experiment by Bonaccorsi et al. (1984), who found that G.I.
bioavailability of TCDD from freshly "spiked" soil was 56-74%. Umbreit et al. (1988) found lower
bioavailability from an environmentally contaminated site, demonstrating that aging of the soil affects
bioavailability. Furthermore, McConnell et al. (1984) found that environmentally contaminated soil
samples were 24-32% as bioavailable as TCDD in a corn oil matrix or a freshly "spiked" soil matrix.
As a result, FDA (1989) recently concluded that a reasonable estimate for absorption from ingested
soil is in the range of 45-55%. This range is used for the "best" and "high risk" estimates in the
typical exposure assessment, while a value of 20% is adopted for the low estimate. For the MEI
analysis, a gastrointestinal absorption rate of 70% is assumed, which is in the high end of the range
cited in the FDA literature review.
Data Sources and Model Incuts for Estimating the Population Exposed
In this analysis, the population exposed to TCDD and TCDF through dermal contact is limited
to the population residing on the agricultural land application sites. The number of sites applying
kraft mill sludge to land is equal to the total number of acres applied with sludge in the state divided
by the average number of acres per site. Values for both the total acres and the acres per site were
obtained through conversations with state officials in Mississippi and Pennsylvania, the states where
agricultural land application is currently practiced. In Mississippi, 1000 acres are applied with the
sludge from one mill, with an estimated 100 acres per site, yielding an estimate of 10 sites in
Mississippi. Pennsylvania has 75 acres covered with sludge from one mill, with an average of 15
acres per site, giving a total of 5 sites. The total number of sites in each state is multiplied by the
number of people living on each site to obtain the exposed population. According to the 1980 U.S.
Census, the average number of persons per household is 2.7. In Pennsylvania, the exposed population
is approximately 14 persons, while in Mississippi, the exposed population is approximately 27. The
total exposed population is about 40 persons.
202
-------
2.4.4 Estimates of Exposures and Risks from Inhalation of Sludge-Contaminated Particulates
TCDD and TCDF adhering to soil particles can become suspended in the air near a land
application site. Transport downwind will dilute the concentration of particles from a land
application site; these particles will also redeposit on surfaces. Residents living on or near the land
application sites may be exposed to TCDD or TCDF by inhaling these particles. This section
describes the methods used to estimate the emissions of particles from a land application site and
the subsequent human exposure to these emissions. This analysis only considers exposure to inhaled
particulates for residents onsite.
To estimate the suspended particulate concentration at land application sites, the methodology
presented in Estimating Exposures to 2.3.7.8-TCDD (EPA, 1988a) is used for estimating emissions
due to wind erosion. Although other models for emissions from intermittent, short-term sources,
such as spreading operations and vehicular traffic, were also presented, the model for emissions from
wind erosion was chosen, since the analysis focusses on average exposures over the long-term. EPA
(1988a) describes the assumptions underlying the model as follows: "This method assumes that the
uncrusted contaminated surface is exposed to the wind and consists of finely divided particles. This
creates a condition defined ... as an "unlimited reservoir" and results in maximum wind-caused dust
emissions." (p.66). The model incorporates information on wind speed and percent vegetation cover
to estimate the flux of small particles (i.e., less that 10 um) from an area of land. Soil amended with
paper mill sludge may not have the characteristics assumed by the model; to the extent that the
surface of a sludge-amended site consists of crusted, coarser particles, the model is likely to
overestimate emissions.
To obtain particulate concentration, the calculated emission rate is used as input to a box
model of atmospheric mixing. The box model ignores any atmospheric dispersion downwind, and
is only appropriate for estimating onsite concentrations. The model uses wind speed, size of the site
and the mixing height to yield an onsite particulate concentration (EPA (1988a), Equation 4-7).
As an alternative approach to estimating onsite particulate concentration, the model described
by Hawley (1985) is applied; this model uses measured values of total suspended particles adjusted
by the fraction of particles assumed to be derived from local (contaminated) soils to derive the onsite
concentration of contaminated suspended particulates.
The calculation of risks from inhalation of particles requires several steps. First, the emissions
of particles from the treated area is estimated. Next, the indoor and outdoor concentrations of
particles onsite are calculated. The concentrations are combined with information about the length
203
-------
of time spent indoors and outdoors, respiratory rate, and the cancer slope factors of TCDD and TCDF
to yield the estimated cancer risks.
Description of Calculations Used to Estimate the Concentration of TCDD and TCDF in Particulates
The first step in this calculation is to estimate the emissions of particulates from the treated
area as follows:
E = 0.036 (l-V)(Um/Ut)3 F(x)
where:
E = emission rate, g/m2 hr
V = fraction vegetative cover
Um = windspeed, m/s
Ut = threshold wind speed (wind velocity at height of 7 meters above the ground
needed to initiate erosion)'
F(x) = function specific to the model, described in EPA (1988a), where F(x) is estimated
by first calculating x = 0.886 (Ut/Um)
This equation gives the flux of dust particles from the surface as a function of 1) the vegetative
covering of the surface and 2) the cube of the ratio of the windspeed to the threshold wind velocity
(the velocity required to initiate erosion). F(x) is a function that is specific to this model. The value
of x is calculated as a function of the ratio of threshold wind speed to the wind speed.
Once the value of x is calculated, F(x) can be determined by reading the value from the graph
of the function presented in EPA, OAQPS (1985, as cited in EPA, 1988a).
To convert the dust flux to a contaminant emission rate, the following formula is used:
Q = CSEA (1 hour/3,600 seconds)
where:
Q a contaminant emission rate in mg/s
Cs - contaminant concentration in the soil, mg/g
E = flux, g/m2 hr
A = area of the treated site, m2
204
-------
The next step is to estimate the concentrations of particulates on the land-treatment site. Both
outdoor concentrations and indoor concentrations must be calculated. The outdoor concentration
is derived as follows:
C0 = Q/(L MH V)
where:
C0 = contaminant concentration in suspended particles onsite outdoors, mg/m3
Q = emissions in mg/sec
L = length of one side of the treated area, m
MH = mixing height (assumed to be 1.5 meters)
V = wind speed at mixing height, m/s, assumed to be 2.2 m/s
An alternative method of calculating outdoor contaminant paniculate concentrations is to
adjust the measured TSP concentration at the site by the fraction believed to originate from local
(contaminated) soils (Hawley, 1985). This method is described by the following equation:
C0 = TSP FL Cs
where:
C0 = concentration of suspended particles outdoors originating from sludge-amended
land, mg/m3
Cs * concentration of contaminant in soil, mg/kg
FL - fraction of total suspended particles assumed to originate from local
(contaminated) sources
TSP = measured total suspended particle concentration, mg/m3
Regardless of the method used to estimate outdoor contaminant particle concentrations, the
indoor concentrations are derived using the following equation:
C,-n - CoRF
where:
Cjn = indoor contaminant concentration, ng/m3
C0 = outdoor particulate concentration, ng/m3
R = ratio of indoor particulate concentration to outdoor particulate concentration
F = ratio of the concentration in indoor dust to the concentration in outdoor soil
205
-------
First, the indoor suspended particle concentration is derived by applying the ratio of suspended
paniculate concentration indoors to the suspended particulate concentration outdoors. Next, since
only a portion of indoor suspended dust is assumed to originate from outdoor sources (the rest is
derived from smoking, cooking, etc.) the contaminant concentration in indoor dust is adjusted by
a fraction representing the ratio of indoor dust contaminant concentration to the outdoor soil
contaminant concentration.
Description of Calculations Used to Estimate Human Exposure to Particulates
Once the concentration of contaminants in particulates is estimated, human exposure to
contaminated particulates can be estimated. In the "high risk" estimate of typical exposure, and in
the "high risk" MEI analysis, the particulate concentration was estimated based on total particulates
using Hawley (1985). When calculating exposure from this method of estimating particulate
concentration, the first step is to determine the concentration of particles that are respirable. The
respirable concentration is estimated as:
RC0 = C0 FR
- Cin FR
where:
C0 ป concentration of TCDD or TCDF in suspended particles outdoors, mg/m3
Cjn ป concentration of TCDD or TCDF in suspended particles indoors, mg/m3
FR ป fraction of suspended particles that are respirable
RCQ = respirable particulate concentration outdoors, mg/m3
RCin = respirable particulate concentration indoors, mg/m3
In the "low risk" and "best" estimate calculations for typical exposure, the EPA (1988a) method is
used to estimate emissions of contaminant adhering to respirable particles (that is, all of the emissions
are assumed to be respirable). Therefore, these estimates need no further adjustment.
The next step in the calculation of human exposure to TCDD and TCDF through the inhalation
of particulates is the estimation of the daily dose. The daily dose is calculated for three age groups:
young children (ages 1-6), older children (ages 7-11), and adults (ages 12 and older).
DOSE0ffl- [(RC0 Dt ABL H0ง9) + (RCQ Dgj ABgi H0งg)] Vg / BWg
g= [g) + (RCin Dgi ABgj H,^)] Vg / BWg
206
-------
where:
ABL = systemic absorption rate through the lung
AB - = systemic absorption rate through the gastrointestinal tract
BW = body weight of individual in age group g
Dt = fraction of respired particles retained by the lung
D . ป fraction of respired particles swallowed (fraction of particles to gastrointestinal
tract)
DOSEQ g= dose to individual in age group g, outdoors, mg/kg/day
DOSEj g= dose to individual in age group g, indoors, mg/kg/day
Hl- = hours spent indoors for individual in age group g
H = hours spent outdoors for individual in age group g
,3,
V = weighted average ventilation rate for individual in age group g, m /day
In this equation, the concentration of the contaminant adhering to particles is multiplied by
the volume of air inhaled each day and by the fraction of the day spent outdoors. Similarly, the
quantity of particulates inhaled indoors each day is the product of the indoor respirable
concentration, the volume of air inhaled each day, and the fraction of the day spent indoors. The
total quantity of particles inhaled each day is then partitioned between the lung and the
gastrointestinal tract. A gastrointestinal absorption fraction is then applied to the portion swallowed,
while a respiratory absorption fraction is applied to the portion remaining in the'lung.
A weighted average dose for an individual over the entire lifetime can be derived by weighting
the daily dose received during each age interval by the fraction of the individual's lifespan spent in
that age group. This calculation is described in the following equation:
DOSEayg= E (DOSE0 *9
Fg ป fraction of lifespan spent in age group g
Description of Cancer Risk Calculations
Once the daily dose estimate is obtained, it is combined with the human cancer slope factors
of^TCD&^nd TCDF to obtain an estimate of lifetime risk from particulate inhalation to these
contaminants. The calculation of individual risk is:
207
-------
1C = DOSEayg q,
where:
DOSEavg = weighted average daily dose for an individual, mg/kg/day
1C = individual cancer risk over lifetime from DOSEavg of TCDD or TCDF
q1 = incremental lifetime risk per mg/kg/day dose of TCDD or TCDF
Individual cancer risk for a typical exposed individual is converted to annual total population risk
(in cases per year) by multiplying the number of persons exposed by the typical individual risk and
dividing by the average person's lifespan, as described in the following equation:
PC = 1C POP / LS
where:
LS = average lifespan of an individual = 70 years
PC = population risk, cancer cases per year
POP - population exposed to DOSEavg
Data Sources and Model Inputs
The values used for each model input for "low risk," "best" and "high risk" typical exposure
estimates are summarized in Table 2.4.K.. The values used to derive the MEI "besi" and "high risk"
exposures are found in Table 2.4.L. The best MEI exposure estimate is derived by combining
estimates of behavioral input parameters with the best estimates of physical/chemical properties of
TCDD and TCDF. The "high risk" estimate of MEI exposure uses the same behavioral inputs, but
combines them with the high estimates of physical and chemical parameters of TCDD and TCDF.
The following sections describe each input and documents the data sources used to derive the
values for the parameters for both the typical and MEI analyses. Where parameter input values differ
for the "best" and "high risk" MEI exposure estimates, these differences are discussed. For those
behavioral input parameters that do not vary between the "best" and "high risk" MEI calculations, a
single value for the MEI analysis is discussed.
Data Sources and Model Inputs for the Wind Erosion Flux Calculation
The present analysis follows the calculations of EPA (1988a) to calculate wind erosion, and
uses many of the input parameters used in the sample calculations in that document. These inputs
are briefly described below.
208
-------
m
w
3
X
ฃ
4-
ID
O.
c
O
4-
ซ3
^~
(D
ฃ
C
ซ
4-
a
3
o
^_
l_
ID
a.
a
c
ID
i_
O
a.
ID
aป
4-
a
o
Q.
a.
<
o
c
ID
_J
1
a
3
T>
ฐ>
5
a
o
Q.
>.
t-
a
b
*^
V)
a
3
lo
>.
t.
4>
4-
I
a
^
a
a.
o
c
a
ง
4-
a.
E
3
in
in
<
x
V
.N
0
J3
ID
4>
O
c
4)
(_
4)
1
C
o
4-
0
C
ID
a.
3
v>
o
i
c.
O)
X
4)
4-
ID
6
1 ^-
o> 4-
Sin
o
X
o
_l
I.
4)
4-
4- i
3 ID
a i.
c a
a
ป
ao
CM
vO
T> -* ""
c "u r*ซ ' "-* *-ป ป 4)
10 c in oo oo Q c
ID CJ> >O VO CC 3
O! CT* CM LU ~" 5
c I- EX
r- ป O
O> - l_ L.
m in i. in ^ fc * & ป O 0
Ot Oป +- Oป O> O O\O El-X
OO ซ ซo>m - ปo> ซoi c
>. >.CID>.>ซ EC EC (OQ
4) 4) >~ -O 0> 3 3 l_
4-ID4- ID4- IO4- O'O
cinm ui4) o
in.C CO (D4-
LUlOOป-3 l_ 4- 3
O in.c4)o u m 4ป o x: ex c o O
(.UI4-O Eฃ 4)3U4)O) in U3 U 4)
ซ. 0 in ซ o x 4- . __ <3>a,>
cnoi i- u ID j: ฃ m in 4- .* a.
C0OO. 1_ปซJ >O 4> UC I.IDID
^ C > E S Q. H- 4- T3 4) 3 >.4-
n -o in o ID 03 -oo i- 0 O4-0
0-4-4) O 4- 00- 4- ID-*- l_O.C~3.C
UJ04-O)ปO 10 O ซ c/> *- a. O o
O) IO C > E< 0T3t L. >> ซ^ O.4-V ID ID
O) 4) 0 Q.V.C4) ID in -4-CM U)l.(M X>>0
C > 3 E 4-UJ*'4-l-OCiCC O\ 0IO ID l_
O3X m*ป 4-00 L.O.C in
ui vt m ID 0 ซ- ex m ._._+.
3OiDinxO Om>ซ inin Vin roซ-i: ID^ID
OtUlD I/I 4- ID ซ ID 0 in U Nป O 4- > t ฃ
o o> oo
336 C UJ . O <0 O .CL. 0IX EO>U
u in ซ a m in 4- i_ 0 .000 o
Ul 4- < O E E0C0.000-04- C 1.4-
O ID in Q. ป 0 XO4-OTJQ.CT4-4- C *- T3 O -^ ซ L,
o 0 uj u oi-io 0OL.0IO aa c o c ซj
- ซ*. -^ e in E o E E c x 0 ID ex
"004-Eซ 1030 ID 4-" .Cm 3C
Saomocna. i_ 0 ป- m <- t.4- inx xo> 00
o> Q U. to 3i_in6ino-^ซjin 00 O in ซ0c
_l CO H- X h- OO00ซ9ฃOO.0 CD 1 _IJ= > ^ 4-
a. ID in in
co o a 0 a 0
t O 0 0
0-ซ- -^ I-. O -^ l_ O
eo o o
3 ID E m O O.4-O.4-
ai *ซ 4- o oo o> Kimi_iomi-
moOi-O \0iox0iD
iDin4->^in o O (- a. in t- a.
0
> in m
ซ ~ -0 0 T3 0
<- 0 0
ID ->- I- O I- O
14-1. O O
r* Kt 01 u KI 01 u o
CT O 4) O < X0ID>x0IDซ
uj in > u o "Z i. a. CM L. a.
ป*
O
O> 0
> 0) 01
Ol *O 0 ^ 0
C 4- 0 0
ID <*-UUป-l-O
IE4-1- O O
>O 3 4) 4) O O.4-O.4-
inoi> r- O O Z I- Q. (N L. Q.
0)
c in
c
4- O t-
(0 - 0
in O c
3 01 Ul TO O ป
O (04- Q.Q.0
E 34- tO CO 4-
ซJ 4) C O ID f h-ID
0 O <- 1.
0 ~^4-Q- -4- *- +-
(-4- Q.4-OCC00 O OO C
O "3 to ID 4) > IDO
>*- K-1-CUJ3 C C l_
3 4-OC4-ID OOIO4-4-
TJU ' (. C O C I. 4-C O.
O O04-O0 S-3 4- 1.
ฃ4- QUO U a u UO O 0)
4-1. T3 c c O- i- en ID a * in c
00 C O 3 tO 4) 4> I-O UO JD 3
20 U ซ t- Q- (. U.4- U-4- <
209
-------
TO
1
ฃ
"io
Q_
ซ
O
-t
It)
C
|T
(Q
3
g^_
4-
in co oo co
* O co CT* (^ O^
f\ GD C7\
O U
Cue - -
* O c . >- > >-
O' (O > Q QJ Q) Q)
14-
jป ao in I z z z
4-*-OS=fl3 fO TO IT3
r-. 3 O X X X X
S lo "*ป TJ "*" o o o o
a\ 10 CM 4) o t- i- i- i-
* 4 E "^^ ** ^- ^
^ TJ 10 4)
ป 10 4) E TJ TJ TJ TJ
E>4-41 0) 41 - .
ซ 3Z 3J3E 1. CM
I..X O LOEIU CMOD1O)-*-
>. u a. in 4- a. 3
c >ซ >ปa.o >. Q. a) mv_>> 4-
TO mio>> 10 om io4>i/) TJฃ a. in
>.TJIO in TJ >. l/l 1 C 41
i_ ฃt IOXTJ a.>>4-ia tti>- 4) mi.
4> IO> TJVI IO 3 TJ >. TO O.41CCฃ
01 in uin TJTJO TJIOZ ini-(j~ 4-i_
~4> CMฃ r~ Z O<0 CO
O o ป inin ซ 4-o.in>-O1^
0. >TJ maoin ฃ >.-ฃ>. to in > ID E
IO3 ^ ฃ O ซ IO l_ O >
Ol O't- l_-4- t_ 41 IOTJ TJฃOITJXCU
co in Otnc in tu a. JD inTJ^ 10 c in4>io
oo CD zi_O 4i^i O 4ixm CM m i_ E Q.
4-CTt **- 06 EOTJ4- Einu 4-*- >\Oฃ6in
EO 4-O 34- O 3Uฃ IO to 3
o-^ O TJvo ino O inฃ ฃu^a CMUIOI
I_I 3+- I/I O ฃ 1 mCM 4-O C
E>--ป-iu TJ3 10 i u > loin ป...>. c
3 TJ IO O >. IO TJ O IO >
IO3 ฃ> ฃ "lOinZ "ป-in 41 >. TJ ฃ
ฃ4- oi4) ininoi 4-Z41 4-O4) ฃ104-^ oiin
om ~i_ 4) ฃ in eซ in E 3 TJ 4- in >-
I/I ฃ e 4- ฃ > 4! ป 3 >-. 4>4>3 Ul IO 1- ฃ IO C
L. A 3C IQ AJ^iniO AOlin in*- ฃ^T3
ง41 TJ 3 m o TJ TJ o m TJ am a c TJ
4- CO ingCX TJTJ1OX TJI.IO U CM C O >
U4- ot_ a an ex in ca o to 10 10
ป- 10 01 in t. ซj m i- ซj > o ฃ -ซ -a
4-^ 4-ฃ U ฃ ฃ Oฃ ~4-m-4- X
Sฃ mi. ป i_ in * ฃ 01 x 01 4- >ปsinu aaz -j -^ CD _j as x _i o x < TJ _i CD - ฃ
IO IO IO ^ * IO IO
TJtn Tjm Tjm >-oTjinT3in
mio mio mio tn4-mioinio
l_TJ l_TJ V_TJ l_4-l_TJl_TJ
in ฃ ฃ ฃ ฃIOฃinฃ
in O CM CM o 3 m
CMIO CMOO CMin CM c CM to <ซr to
O , .- a CM CM
.
IO >- > l_ - TO
Tjm ioin ioซ >-OTJUI ซin
x>- TJ>ป TJ>^ \ x>-Tr>.>.
into x. >- >. t_ - o
10 10 in ioin >- o TJ m
Tjm TJ3~ T3>. X X >
x>- xo xio m4-mซJ
o i*^ m CMcio^o* a.
O co v ao in to m
4- 4-4-4-
C C C C 4)
O 4) 4> 41 U
TJQ. TJQ. TJQ. TJQ.IO
-. cin cin cm cina.
O IO4- TO T3 IO l_ TO l/l
4- in in in4) in4-
IO 3^3K3 >. >. _ >>>.TJ >. >^ Ol
U. IOIOTJ ID (O ฃ IOIO IOIO3C
TJTJIO TJTJO TJTJO TJTJTJ-
C ^ ป. ป H.. ^'5ซr o'oin 4>om
-------
IO
X
c
o
4-
10
IO
.e
c
"~
4)
IO
ซ
0.
._
^L
L,
IO
a.
a
10
ฃ5
Q^
U
C
a
u.
o
<ฃ
o
4-
a
c
a
Q.
X
N^
in
o
i
O)
X
a
a
E
in 4-
4) m
m ซ
3
i
L.
4)
^>
4- E
3 a
a. L.
c
L. ID ฃ X ฃ
in *o 4- ID in 4- 4>
<_ - x -a i-
.cuininx xox ^
4) i- i- mo
CNAfOT? U'O'O I/J
O o 4) ji c 4) in
+- If) T3 ._ o
uo ~>.3 .cm 3 m
81 >- in ID O 4- o coo
>- ID i- ซ ~o u me u O> 3
C Z O 41 -^ 4) O 6 4J *~IO
i- -D n o 4- o 4- ->
cnoco ie T3 in .c a >
in c a. 4- L. E c u E 4) E
>.~ 04)^ L. ป-. 3
iOL.mmO'04-" omซ4-ป XE
^ 3i.4)icin4) m *4ป 4) ^^ in 4) 10
Q,T3^:E>. 4)4- e ID 4) 4- XX
3IOID IO >~ 3 T3 a IO
x (N m 4)O \inmo O
ซJC3)L.O4- L.IOL.O4- L.m
-CL.rn c 'om fฃ f^in L^*~
O 4) I- L.C4) U C 4)
O. O .C I- O tฃซO ..ซ
m O O) 3 *- i_ mfNO) L. 4-4-
fm -a o x o o x O ma
>-Cฃ > O O 6 ซ JC ซ O O 4) E
C ~ ^ > _l 13 3J< O) _J T3 A
m T3 a x a 4- m x T3 o 4- 4-
in L.CVX 3 mxei. 3 TS m
ior^aamm->o ซj m o ซ o c ฎ
ซ>(.ซ >>>ซ a
">>-4-IOฃ4-f 10 4- 0 4- ฃ ฃ
X > V) 'O OOI X T3 in O D> BO)
O ซ3 ป- ป ซZ~ O bc2 O
ID ID 10
o m "o m *o M
x >ป x >. x in u
in ID in 10 mio t- o
i. -o i- T3 i_ro OO
ฃ in jz fin O *ป T?
O 3 in L. > OQ4-
IOfO O
ID a ID
T> m o in a u
X >- X >. x in u
in (D in ID mio L.Q
L.'OL.TS L.fO OO
j= in c. j: m o TJ
O 3 in u.>> x> ปซ 4-
to 10 ป to lOOffl cos
Q. CM CM ซ-ป--O OO O
>. X >. >ป
a ซ a a 10
Qin'om omT> M
in ID in ID in 10 in i_ o
(.OL.-O L.TJL.CM OO
finf f m f m o T3
(N3(N CN 3 L.X-Oซซ4-
CN oo T oo (Nin ^-Qio ciA3
o. ป- o rป-o
4- 4- TJ
C C 4-
4) . 0) 4- C
on. o a. f c ID
cm emu ID in c
a 10 c ซJ
in -o IB L, E
>.> > >~ y> E c ID c
IOIO 1010*0 IOO4-O
0 "5 C 4^ O O 4^
i.*4- L.-4- OIOUIO
OO* OO* UL.CL.
a. in a. i/> 4-OL.4-
(-L, L.L. U C O C
moo moo 004-00
L. n O . L..OO 0 o o -a u
O 3 C O3C CO33O
Xc~ I e Ox-Ou
211
-------
X
ID
2
.C
ID
Q.
C
o
4-
^
C
ฃ
Q
4>
4-
10
_
O
4.
L.
^
*
0
Z
ซ
u
c
a
*?
ir
o
4ป
a
c
a
a
x
^^
tf)
i
x:
at
x
4
4-
a
in 4-
a> tn
ffi ซ
^
o
4-
+- E
3 a
a. i-
c a
O.
in
00
O>
00
00
^m ^,
4>
< X
Q. 10
LU X
_
-
1 C
^3 O
ซ.
* +
LU +-
ซ
O) 01
- >
in
3 ^%
f^
^y in
4)
4- 01
10 C
"3 ~s
O 3
U ID
.. 00
4) OO
+- 0ป
ID
?s
in o
o *-
E O
m M
in o
ID in
" *O
o- O
uj in
-
O)
c in
^ A
ID
in
o
- E
10 4>
U
4>
1_ 4-
O ID
>*-
3
1 ฃ
4- l_
4> ID
Z a.
10
O E
+ป 'V,
01
*- 3
O
^f
s*
i&
3 1-
(A
.
4>
ป
10
in
o>
c
~
S 8
5 ง
f 1
Q. ^0
tn o
1- O
B S r
O t- O
4- ป- Ifl
9
^
4-
a
ซ 0
ซ O
> u
o
c
ro
O)
c
_
in u
00 4>
^- ^)
ป O)
4> _
X
ฃ OJ
Ifl
(Q ^
> JD
.O
2 i
8 =
O O
3 4-
O <0
x- 4-
O c
c u
0 c
- 0
4- O
U
c a.
3
^
o
>
_e
X
IQ
E
m
4)
4-
J
4-
in
^}
01
X
I-
8
I 1
- 0
u
8
o
oo o
c
10
in c
(0
E
c to c
O 4- O
ป- c
+- O O *-
L. C L.
4- O I- 4-
C 0 C
4) 4- O 4)
O O "O O
C C 4- C
O 3 3 O
U ป- O O
in
00
o\
^,
4)
X
ID
dT
(/^
t
O
4>
>.
C
O
m
4)
MB
^
^
a.
a
u
*
c
>
"01
1
Q
ซ
o
o
-~
5
a.
(^
^
4-
ซ
U
ฃ
i
*~ >-
+- i_
v- c O
ID O ซ-
CT
w in
"5 o
e a. ซ
(/> a. ^>
CD (A (U
T9 * 3
C L. 6 " "
a L. 4i
ID 4- O
0 m *
4-4-
4- 4- 4) -a
m 4>
in en 3
8 *. 2
y _ X
u in o u
i ^ ฃ i
4)
ID
l_
a.
in
4>
u
**-
E
-j
o
c
(O
l_
g
^
o
c
(D
0.
o
L.
O) O) O)
c c _*
331-
O
4) .
^- r- _
at 01
3 3 JD
O O ID
t- I-
4- 4- ID
>
O Tป ID
4) 4) O
J3 .0 -
fe >fe "
in in E
A A O
10 10 L.
tf) U)
_ _ m
8
o
8
*
_ o
o
ID
Q- Ol Q- -t- 4)
V> C t/1 4-
H- 3 h- ID
- "0 u
**- ป^
O O O O c
V 4- O
C C
O 'O O "O 4-
._ ft) ._ fl) Q.
V 4- 1-
U ID U ID O
U C I-C JD
U. U- <
ao
_
ซ
>0
-"
4)
Q c
rf 3
LU ~~y
X
* uT
o
a.
(O a
tn 4>
-^ ^z
4^
4)
3 ฃ
O
10 10
> 4)
in
~ 4-
1 ฃ
4-
in
Ot 4)
03
Ol U
- U
C 10
ID a.
4) 4)
O)
C
3
212
-------
(O
i
I
(0
Q_
ง
ฃ
J5
(Q
ฃ
C
O
^
3
U
*T
^
ID
o.
XI
ID
O
ID
^*
I
.
c
t"
^^
^
^
^
^_
c
ZJ
1
1
_ _
(Q
n
t^
^
t^
w
c
4)
O
X
UJ
ce
c
O
4-
a
c
a
a.
X
9
in
^_
Q
%
j;
CT
X
4)
4-
ID
E
in 4-
m o
t.
4)
4-
4)
4- E
3 10
a. i_
C ID
a.
__
ID O)
C
o
c
in .a
- XI
4)
UJ Q
Z I.
o
O m
4- .0
ID
4>
4)
X> U
ID ID
U
in
4>
a
ex u
a
4-
4- U
O ID
"i. a.
^
2
z.
*
.
a
XI
in
i-
in ID
I- XI
O
CM Kl
^
a
xi in
^ >
a> a
L. -a
[^
f^
JVJK,
+.
C
4)
xi a.
c in
a
in
^> ^
ID 10
XI XI
X -
ซ O
a.
in 4)
L. JD
3 E
^_
3
^5
ID
ซ
in
L.
8
XI
4-
8
in
CO
Oi
-
41
f
ID
X
s
XI
4)
4-
a.
O
XI
^
.
>^
ID
X"
-V
in
CM
in
o
Q
XI
4-
O
in
ID
0.
XI
Mป
<^
u
in
4)
3
in
<
10
XI
in
i_
CM
.,.
ID
XI
in
i_
JS
CM
XI
C
ID
^b
10
XI
l_
4)
Q.
in
t.
3
5
^
*
l_
OJ
Q.
U)
(O
*o
^^
ซ
L.
o
4-
$
>.
z
m
ID
XI
CM
CO
in
^ป
10
XI
fg
CO
+.
c
4)
a.
m
XI
in
x *~~
ID C
XI U
*^ ^
O in
L.
^
in
L.
JC
in ID
L. XI
^^
CM
CM in
4-
C
4)
xi a.
c in
ID 1.
in 4>
>. >> xi
ID 10
xi xi o
i. - ซ
4) o in
a. i_
i_ o'
in 4> o xi
3 E 4- ~
O 3 3 ฃ
X C O U
in
CO
O
-
41
2
ID
X
1
L.
XI
4)
+
a.
O
XI
<
>^
in CM
i_
iz
CM O
in L.
XI f
4)
Q.
in u
O
4-
in
3 1.
xi j:
ID
CM
4-
ID
4- O
i^
XI
4) >
E "0
3 XI
in
in
a
L.
u c.
4-
4- ~
< -^
I- ซ
>- U
in 4-
l- 4-
-C ID
CM C
,
L.
> U
in 4-
L. 4-
f a
CM C
^_
c
4)
a a.
c m
ID
in
^* ^^
ID ID
XI XI
1. >-
4) O
a.
3 E
X C
c
X
ra
XI
O X
Kป m
^
o *r
4) ซ- (N
U
a ซ in
a. u 3
in ID
a. ex
O) in
c in
O) .C
> C 4-
_ _ c
-J > O
E
O U
C . 3
a in
C XI
N. c
in
in ฃ in
ID CM ID
XI XI
.
ID
o in
in ID
1- XI
ฃ in
O 3
CM ro
a.
X.
a
TJ in
X >.
in ID
i_ XI
ฃ in
ex
4)
u
ID
a.
in
4-
: O)
3' C
XI
a >
ซ _
in
O u
XI 4-
C 4-
ID
1.
ID
4)
O
4-
in
4>
L.
t_
O
4)
U
ID
a.
in
OI
c
^
.^
ID O)
xi m c
m ID >
l_ XI
in
V .
in ID >
l_ XI
in
CM CM
4)
U
10
a.
in
4)
u
ID
a.
in
213
-------
I/)
10
ฃ
IO
Q.
0
4-
IO
-m
.C
C
4)
^
a
u
.
,
IO
0.
T3
C
10
1
(0
*
c
o
V
u
,y
^
fi_
ex
-^
^
10
*^
1
<^
IO
3
^
w
^_
2
^
ซ
1
X
LU
i
4)
^^
^
^
O
**-
U)
3
_.
^*
^^
>-
1
u
Q.
C
10
1
"^
ex.
~
u5
(fl
^
1
*
*
CM
K
0
u
c
41
41
41
Ct
C
O
4-
0
C
a
^m
a.
X
X
i
ฃ
O)
X
41
4-
10
E
I _
ซ 4-
41
4-
4- E
3 a
a. <.
c a
a.
in
co
2
^^1
41
Z
ia
T.
ง
it.
o
41
a.
O
T3
1.
41
Q.
IA
1.
CM
in
Q
Q
T3
C
^
in
>.
IO
ซ*.
a.
o
ฃ
U
Wl
3
m
.
a
TO
in
i.
ฃ
CM
^
a
o
X.
m
i_
ฃ
CM
o
c
ID
^^
a
o
*
IO
^y
0)
c
ซ^
(.
3
TJ
^
X
1.
41
a.
x
a
D
r-
a
in
10
o
PM
ao
in
>.
a
^3
PM
CO
4-
C
41
a.
m
in
IO
O
E
O
a
u
41
a.
in
L.
e*
^
(N
iO
1
e
^
s
'o
PM 10
o
m x.
3 m
i_
0. ฃ
^ 7*
PM IO
o
in x.
3 m
L.
a. ฃ
^
^
ป
r-
o
,
in
8
a
c
in
10
"O
CM
co
ui
^*
IO
T3
PM
ao
in
ao
2
^
41
I
IO
X
ง
L.
o
41
4-
O.
O
^
CM
41
O)
a
u
>
a
c
o
ซ
^
o
Q
^
C
in
o
JJ
U
y}
s
3
in
in
<
a
o
m
i_
CM
^
a
o
x.
in
u
ฃ
CM
o
c
a
^b
a
L.
0
a.
ซ
u
3
X
m
8
o
c
in
o
c
ID
ซ
l_
41
^
E
41
4-
a.
41
i
>.
10
ง
U
ซ
^
o
in
(_
ฃ
in
IO
o
PM
in
-
m
>b
fc
10
o
ซ^
O
41
E
^
10
in
L.
ฃ
^f
CM
U
IO
41
>ซ,
41
ฃ
4-
^,.
O
in
41
V.
a
ฃ
4-
o
10
TJ
in
i- PM
in ฃ
3 l_
ป o
Q. CM ซ-
.
-------
Vegetative cover is assumed to be 50% for the land application sites, since these sites are used
for agricultural application; the sites are assumed to be bare for about half of the year, and covered
with crops for one-half of the year. This assumption is used for the "best" estimate of exposure and
for the "best" estimate of MEI exposure. For the emissions estimate in the "low risk" typical exposure
analysis, 90% vegetative cover is assumed.
Mean annual windsoeed is 4 meters per second. This is the average windspeed at a height of
10 meters for 60 major cities in the United States (EPA, 1988a).
The threshold wind speed can be derived if the roughness height and the threshold friction
velocity of the surface are known. EPA (1985, as cited in EPA (1988a)) describes the ratio of
threshold wind speed to friction velocity as a function of roughness height. The threshold friction
velocity for unlimited reservoir surfaces is less than 75 cm/s (EPA, 1988a). Based on this
information, EPA (1988a) adopted a value of 50 cm/s for this type of surface. The present analysis
also uses this value for threshold friction velocity at sludge land application sites. For the roughness
height of an agricultural land application site, a value of 2 cm is used, which corresponds to a field
with grass cover. The ratio of threshold wind speed to threshold friction velocity for a roughness
height is derived from tables provided in EPA (1985), as cited in EPA (1988a). For a roughness
height of 2 cm, the ratio is 15. To obtain the threshold wind speed, this ratio is multiplied by 50
cm/s, the assumed value for the threshold friction velocity. The resulting value is 7.5 m/s.
The value for the function F(x) can be obtained from a graph of the function found in EPA
(1985), as cited in EPA (1988a). First, the value of x must be calculated. For a site with a threshold
wind speed of 7.5 m/s, the estimate is 0.886 x [(7.5 m/s)/(4 m/s)], or 1.66. From the graph provided
in EPA (1985), cited by EPA (1988a), the value of F(x) for x = 1.66 is 0.65.
Data Sources and Inputs for TSP Concentrations
As an alternative to the wind erosion dust flux calculation, the method used by Hawley (1985)
is used in calculating the "high risk" estimate of typical exposure, and in calculating the "high risk"
estimate of MEI exposure, since the result derived by this method is higher than the result using the
wind erosion dust flux equation. This method bases the estimate of paniculate concentrations on
measured values of total suspended particles (TSP). The annual geometric mean values for suspended
particle concentrations for SMSAs of between 500,000 and 1 million people were obtained from
National Air Quality Trends Report, 1982 (EPA, 1984). The average value for these areas is 64
215
-------
ug/m . This value is similar to the value of 70 ug/m used by Hawley (1985). This value is then
adjusted by the percent of suspended particles derived from local soils, assumed to be 50 percent
(Hawley, 1985).
Data Sources and Model Inputs for Soil Concentrations of TCDD and TCDF
The estimated soil concentrations for each land application site are displayed in Table 2.4.A.
The methods used to derive these concentrations is discussed in detail in Appendix A.
Data Sources and Model Inputs for Deriving Indoor Airborne Particle Concentration as a Function
of Outdoor Particulate Concentration
Hawley (1985) compared several studies that investigated the relationship between indoor
particle concentrations and outdoor particle concentrations. Whitby et al. (1957), as cited by Hawley
(1985), found that, for the City of Niagara Falls, New York, the indoor suspended particulate
concentration was 65 ug/m3 and the outdoor particulate matter concentration was 93 ug/m3. This
yields an indoor to outdoor ratio of approximately 0.70. Sterling and Kobayashi (1977), also cited
in Hawley (1985), found that this ratio ranged from 0.77 to 0.85. In the typical exposure assessment,
the value 0.75 is used as the best estimate of this ratio, while 0.70 and 0.85 are used as the low and
high estimates, respectively. For the "best" MEI exposure estimate, a value of 0.75 ;s used, while the
"high risk" MEI exposure assessment uses a value of O.S5 for this parameter.
Data Sources and Model Inputs for Deriving Indoor Dust Contaminant Concentration as a Function
of Outdoor Soil Contaminant Concentration
Roberts et al. (1977), as discussed by Hawley (1985), studied the relationship between lead
concentrations indoors and outdoors near a lead smelter, and found that the mean concentration of
the lead in household dust was 75% of the concentration of lead in the outdoor soil. For his own
analysis, Hawley (1985) assumed that indoor contaminant concentrations in dust were 80% of the
contaminant concentrations in outdoor soil. The typical exposure analysis uses a value of 80% for
the "best estimate", and applies a range of 75% to 100% for the "low" and "high" estimates,
respectively. For the "best" MEI exposure estimate, a value of 80% is used, while the "high risk"
MEI exposure assessment uses a value of 100% for this parameter.
216
-------
Data Sources and Model Inputs for Estimating the Percent of Particulates that are Respirable
Deposition in the lung depends on the size of the particle. The typical exposure analysis uses
a method of estimating emissions of particles less than 10 urn in diameter to derive "best" and "low"
estimates of the concentration of suspended particles at a land application site. The "best" MEI
exposure estimate uses this value as well. Schaum (1984), presenting data from ICRP (1968), states
that almost all particles less than 10 um are respirable. Therefore, for the "low" and "best" typical
exposure estimates, all of the particles were assumed to be respirable. In the "high risk" typical
exposure assessment, and in the "high risk" MEI exposure assessment, the concentration of suspended
particles is derived from measured values of total suspended particles, a measurement which includes
particles of various sizes. Therefore, the fraction of total suspended particles that will be deposited
in the lung must be estimated. Hawley (1985) and Schaum (1984) both assumed that 75% of inhaled
particles are retained in the body. For particles 0.2 to 20 um, ICRP (1979), as cited in Hawley (1985),
indicates that the fraction deposited in the respiratory tract ranges from 60 to 90%. For the "high
risk" typical and "high risk" MEI scenarios, it is assumed that 90% of TSP is respirable.
Data Sources and Model Inputs for Estimating the Fraction of Inhaled Particles Deposited in the
Lung and in the Gastrointestinal Tract
Schaum (1984), citing ICRP (1968), discusses the distribution of inhaled paj-ticles within the
body. Of the particles initially retained by the body, one-third remains in the lower sections of the
lung, and two-thirds remain in the upper respiratory tract. Eventually, some particles in the
respiratory tract are swallowed. After twenty-four hours, approximately one-half of the amount
originally retained in the lower sections of the lung is swallowed. For the "best estimate" and "high
risk" typical exposure scenarios, it is assumed that one-third of particles are retained in the lung for
a sufficient length of time for systemic absorption of TCDD or TCDF to occur through the lung. For
the "low risk" typical exposure scenario, this analysis assumes that only one-sixth of the particles are
retained in the lung for a sufficient length of time for absorption through the lung to take place. The
rest is swallowed and absorbed through the GI tract (which has a lower systemic absorption rate).
In the MEI analysis, it is assumed that all respirable particles are retained long enough in the lung
for systemic absorption through the lung to occur (that is, all of the contaminants adhering to respired
particles that are absorbed are absorbed through the lung).
Data Sources and Model Inputs for Respiration Rate
Respiration rate is used in the model to assess the total daily volume of particles inhaled. For
adults, the average respiration rate was calculated by EPA (1985b) to be 23 m3 per day. This value
217
-------
was calculated using data on the ventilation rates during different levels of activity, and the amount
of time spent per day engaging in these levels of activity, to obtain a daily total. For children,
Hawley states that the ventilation rate of young children engaged in light activity is 7.6 1/min, while
the ventilation rate during rest is 2.8 1/min; assuming children spend roughly one-third of their day
engaged in light activity and two-thirds at rest, the total ventilation rate is 6.3 m3 per day. For older
children, the ventilation rate is 11.6 1/min during light activity and 4.3 1/min at rest (Hawley, 1985),
with a total ventilation rate of-8.4 m3 per day for both the typical exposure and the MEI exposure
analyses.
Data Sources and Model Inputs for Estimating Absorption Through the Lung
Little data are available to estimate the systemic absorption of TCDD or TCDF through
inhalation. Faced with a lack of information, U.S. EPA (1989d) assumed that TCDD is almost
completely absorbed from respirable particles (i.e., those less than 10 um in diameter). This analysis
follows this assumption, and assumes that 100% of the TCDD and TCDF adhering to respired
particles is absorbed through the lungs for both the typical and MEI exposure analyses.
Data Sources and Model Inputs for Estimating Absorption through GI Tract
Absorption of TCDD and TCDF through the gastrointestinal tract has been studied using a
variety of media. Absorption will be influenced by how tightly TCDD or TCDF binds to the matrix
in which it is ingested. Poiger and Schlater (1980), as cited in Schaum (1984) reported that the
bioavailability of TCDD from soil was 20-26%. In a recent review of the literature, FDA (1989)
discussed an experiment by Bonaccorsi et al. (1984), who found that availability of TCDD from
freshly "spiked" soil was 56-74%. The memorandum also cites a study by Umbreit et al. (1988), who
found lower bioavailability from soil from an environmentally contaminated site, demonstrating that
aging of the soil affects bioavailability. Also discussed was a study by McConnell et al. (1984), who
found that environmentally contaminated soil samples were 24-32% as bioavailable as TCDD in a
corn oil matrix or a freshly "spiked" soil matrix. As a result, FDA (1989) recently concluded that a
reasonable estimate for absorption from ingested soil is in the range of 45-55%. This range is used
for the "best" and "high risk" typical exposure estimates, while a value of 20% is adopted for the "low
risk" typical exposure estimate. No gastrointestinal absorption of inhaled particles occurs for the
MEI, since all absorption is assumed to occur through the lungs.
Data Sources and Model Inputs for Determining the Fraction of the Day Spent Indoors and Outdoors
In order to estimate exposure duration in indoor and outdoor settings, the methodology
developed by Hawley (1985) to estimate the time spent outdoors and indoors by different age groups
218
-------
is adapted. For adults, Hawley (1985) also presented values for estimating exposure to dust while
cleaning infrequently used spaces, such as attics, that have been incorporated into this analysis.
Young children have the most exposure to outdoor particulate concentrations. In the typical
exposure analysis, these children are assumed to be outdoors 8 hours per day, five days per week
from May to October. The remaining time is spent indoors on the site. For the "high risk" estimate
of typical exposure, it is assumed that young children are outdoors for an average of six hours a day
for the entire year, which is the equivalent of twelve hours per day, seven days a week for the six
months out of the year. This assumption is also used for the most exposed young child.
Typical older children are assumed to spend an average of five hours per day from May to
September outdoors. This value is the average of time spent outdoors after school on school days
and time spent outdoors on weekends and on school vacation days. Older children are assumed to
be indoors on the site for sixteen hours per day for the entire year. The remainder of the time is
spent at another indoor location, such as school. For the "high risk" estimate of typical exposure,
older children are assumed to be outdoors an average of 5 hours per day for the entire year, which
is the equivalent of 12 hours per day for five months, and to be indoors on the site for the remainder
of the time. This assumption is used for the most exposed older child analysis as well.
This analysis assumes that an adult living on an agricultural land application site (i.e., a farmer)
works outdoors 5 days per week, 12 hours per day, for six months, and spends the rest of the time
indoors on the site. This value is used in the typical exposure analysis for all but the "low risk"
scenario, and is used in the MEI analysis. For the "low risk" typical exposure scenario, it is assumed
that the adult resides on the farm, but works elsewhere; in this case, the adult is assumed to engage
in outdoor activities onsite for 8 hours per day, two days a week from May to September, and to
spend 16 hours per day indoors on the site all year long.
While indoors, adults may spent a limited amount of time in an extremely dusty area, such
as an attic, where exposure to inhaled dust would be higher than in normal living spaces. Hawley
(1985) estimated exposure to dust to be 20 mg during one hour in the attic, and assumed that adults
are exposed at this level for twelve hours each year (either one day for 12 hours or one hour for
twelve days). The current analysis incorporates these assumptions into both the typical and MEI
exposure assessments.
219
-------
Data Sources and Model Inputs for Estimating the Population Exposed
In this analysis, the population exposed to TCDD and TCDF through dermal contact is limited
to the population residing on the agricultural land application sites. The number of sites applying
kraft mill sludge to land is equal to the total number of acres applied with sludge in the state divided
by the average number of acres per site. Values for both the total acres and the acres per site were
obtained through conversations with state officials in Mississippi and Pennsylvania, the states where
agricultural land application is currently practiced. In Mississippi, 1000 acres are applied with the
sludge from one mill, with an estimated 100 acres per site, yielding an estimate of 10 sites in
Mississippi. Pennsylvania has 75 acres covered with sludge from one mill, with an average of 15
acres per site, giving a total of 5 sites. The total number of sites in each state is multiplied by the
number of people living on each site to obtain the exposed population. According the 1980 U.S.
Census, the average number of persons per household is 2.7. In Pennsylvania, the exposed population
is approximately 14 persons, while in Mississippi, the exposed population is approximately 27. The
total exposed population is about 40 persons.
2.4.5 Estimates of Exposure and Risks from Inhalation of Vapors
Residents of land applications sites may incur risk from the inhalation of volatilized TCDD
and TCDF. The methodology for estimating the emissions of TCDD and TCD,F vapor at land
application sites generally follows methods for estimating volatilization described in EPA (1988a).
Because actual locations of the land application sites are not known, the ISCLT model could not be
used to estimate downwind concentrations. As a result, this analysis estimates only exposures to
onsite residents, using a box model to obtain the onsite concentrations from the emissions estimates.
The calculation of risks from the inhalation of vaporized TCDD or TCDF requires first the
estimation of emissions, then the calculation of indoor and outdoor onsite concentrations. The
concentrations are combined with data on time spent indoors and outdoors, respiration rate and the
cancer slope factors of TCDD and TCDF to obtain the estimated cancer risk from this pathway of
exposure.
Methods for Estimating Vapor Emissions
This analysis uses a set of equations from U.S. EPA (1986), and Hwang and Falco (1986) as
described in U.S. EPA (1988a), to predict emissions from a land application site. It assumes that
emissions from land application sites (in g/m2/s) are described by:
220
-------
[* a T]1'2
where:
D,-E4/3
a =
ps(l-E)/Kas
Kas
and:
DJ = the molecular diffusivity of contaminant vapor in air (cm/second)
Cs = the contaminant concentration in the soil (g/g),
E = effective porosity of soil, assumed to be 0.25 (unitless)
HC = Henry's law constant (atm m3/mol)
ps = true density of soil, assumed to be 2.65 g/cm3
KQ = the soil/water partition coefficient (cm3/g) = (organic carbon/water partition
coefficient)(fraction of organic carbon in soil)
air/soil partition coefficient (mg/cm3 in air per mg/g in spil)
Na =ป rate of emissions from the soil surface (g/m2/second)
T ป duration of exposure (seconds), assumed to be 2.2 x 109 seconds (70 years)
This equation uses information on the partitioning of TCDD and TCDF between soil and air and
between water and soil to estimate emissions of TCDD and TCDF vapor per m2 area. The emissions
estimate is then multiplied by the area of the site, in m2, to obtain the total emissions of vapor from
the land application site:
Q = 1000 Na A
where:
Q = emissions rate for contaminant vapor, mg/s
A = area of the land application site, m2
1000 = conversion from grams to milligrams
The emission rate is then coupled with a box model to obtain the onsite concentrations of vapor.
Outdoor vapor concentrations are estimated as:
C0 = Q/(L MH V)
221
-------
where:
C0 = concentration of vapor outdoors, mg/m3
Q = emissions, mg/sec
L = length of one side of the treated area, m
MH = mixing height (assumed to be 1.5 meters)
V = wind speed at mixing height, m/s, assumed to be 2.2 m/s
The indoor vapor concentration is derived by applying the ratio of vapor concentration indoors
to the vapor concentration outdoors, as described in the following equation:
Cin - CoR
where:
Cin = indoor vapor concentration, ng/m3
C0 = outdoor vapor concentration, ng/m3
R = ratio of indoor vapor concentration to outdoor vapor concentration
It is assumed that the relationship between vapor concentrations indoors and outdoors is similar to
the ratio between indoor and outdoor paniculate concentrations (that is, indoor concentrations are
approximately 75% of outdoor concentrations with a range from 70-85%).
Once the concentration of contaminant in the air is estimated, the calculation of exposure
and risks from the inhalation of vapor then proceeds in the same manner as the exposure and risk
from the inhalation of particles, described in section 2.4.4. Table 2.4.K. summarizes key assumptions
and input parameters for estimating typical individual exposure through the vapor and particulate
inhalation pathways, while Table 2.4.L. summarizes the inputs used in the MEI analysis. In some
cases, the data inputs used for the estimation of exposure and risk are different than those used in
section 2.4.4. The data inputs unique to the calculation of risk from the inhalation of vapor are
described in following sections.
Data Sources and Model Inputs for Estimating Volatile Emissions
The emissions model requires the soil/water partition coefficient as one input. This partition
coefficient is, in turn, based on the fraction of organic carbon in the soil. The higher the fc value,
the lower the emissions, since more of the contaminant will partition to soil. For land application
sites where sludge is soil-incorporated, it is assumed that the fraction of organic carbon in the sludge-
soil mixture is approximately equal to the fraction of organic carbon in the soil alone. A reasonable
best estimate for fc for soils is 1%, with a range from 0.1% to 4%. This range of values is used in the
222
-------
typical exposure analysis. In the MEI analysis, the "best" estimate of MEI exposure is calculated
assuming an fc value of 1%, while the "high risk" MEI exposure estimates is calculated using an fc
value of 0.1%. For those land application sites where top-dressing is practiced, the average value of
fc for sludge is used. NCASI (1984) reports that the organic carbon content of sludge ranges from
14% to 40%. The typical exposure analysis uses 25% for a best estimate of f for sludge; the "low
risk" analysis uses a value of 40%, while the high risk analysis uses a value of 14%. Values of 25%
and 14% are used in the calculation of the "best" and "high risk" MEI exposure estimates, respectively.
Data Sources and Model Inputs for Estimating Exposure to TCDD and TCDF Vapor
The data inputs and model sources for the vapor exposure estimate are the same as those
described in section 2.4.4., with two exceptions. First, 100 percent of vapor emissions are assumed
to be respired. Second, all of the vapor is absorbed through the lung; none is absorbed through the
GI tract.
2.4.6 Estimates of Exposure and Risks from Ingestion of Drinking Water from Ground Water
Sources
Land application of sludge is not expected to present significant risks to human health through
contamination of ground water. As explained in Section 2.1, conservative, "high risk" estimates of
ground water contamination from sludge in industrial landfills yielded risk estimates on the order of
10"6 for a "most exposed individual"; "best estimate" assumptions yielded lower risk estimates.
Consideration of some major differences between landfilling and land application /of pulp and paper
sludge suggests that risks of groundwater contamination and human health risks from land application
should be lower than those estimated for landfills.
Land application and landfill sites differ in at least five important respects. First, land
application sites may be larger than landfills. Second, local geo-hydrological or weather conditions
may differ between landfill and land application sites. Third, sludge may be placed in landfills to
a significant depth below ground level; land-applied sludge is generally applied to the ground surface,
or incorporated into a relatively shallow surface soil layer. Fourth, the quantities of sludge applied
to a hectare of treated land tend to be much lower than the quantities deposited in a hectare of
landfill. Finally, the maximum concentration of TCDF reported in the 104-Mill Study for land-
applied sludge is lower than the maximum reported for sludge placed in landfills. These differences,
and their implications for human health risks estimates, will now be examined.
223
-------
Land application sites can be larger than sludge landfills. A known forest application site in
Wisconsin, for example, reportedly covers approximately 1,000 hectares of land (Wisconsin
environmental department). Landfills are generally assumed to be much smaller; Section 2.1 of this
analysis assumes that they generally cover less than about 25 hectares. If the land application site is
significantly larger than a sludge landfill, then the area of underlying aquifer receiving recharge from
the site may be larger as well. Section 2.1.2 assumed that the vertical migration of contaminants
through a sludge landfill and underlying soil layers is not affected by the area of a site. Once
contaminants enter the ground water, however, the extent of their dilution in transport to a nearby
drinking water well is affected by the area of the site.
Repeated execution of the AT123D model (discussed in Section 2.1.2) shows that estimated
concentrations at the receptor well increase approximately four-fold as the site size is increased by
a factor of 100, and all other inputs for the model runs (except site size) are held constant. A landfill
with characteristics described in Section 2.1.2, but with a size equivalent to the area covered by the
Wisconsin land application site would therefore be expected to result in health risks no more than
four times higher than those estimated in Section 2.1.2.
The second difference between the two sludge management practices is that land application
sites may be located in areas with different topography, geology, and meteorology than those areas
in which pulp and paper sludge landfills are sited. If these local conditions are more conducive to
ground water contamination than those assumed for landfills, then the risks associated with land
application could conceivably be higher than those estimated for landfills. For example, strip-mined
land receiving sludge applications may include fractured rock layers that are especially vulnerable
to the migration of sludge contaminants to underlying ground water. The "high risk" scenario
considered in Section 2.1.2, however, assumed a zero distance between the bottom of the landfill and
the water table, assumed that leachate concentrations are .limited only by the concentration of
contaminants in the sludge and their estimated partition coefficients, and used reasonable worst case
parameter values for most of the environmental transport calculations. It is unlikely that an aquifer
located beneath a land application site would be significantly more vulnerable to contamination than
an aquifer of characteristics assumed for the "high risk" scenario in Section 2.1.2.
A third difference between the two practices is that for most land application sites, sludge is
applied to the soil surface, or incorporated into a soil layer of only 10 to 20 centimeters in thickness.
Landfills, by contrast, can extend several meters beneath the soil surface. Section 2.1.2 examined a
"high risk" landfill scenario that included a sludge layer that extended six ^neters,downward4o the
water table. The vertical distance between the sludge layer and the water table is likely to be greater
for land application sites than the zero distance assumed for a "high risk" landfill scenario. TCDD
224
-------
and TCDF concentrations in water that has percolated downward through an initially clean soil layer
are therefore likely to be lower than those predicted for the "high risk" landfill scenario. Even if the
sludge could be incorporated into the entire soil layer above the water table, expected contaminant
concentrations in the soil-sludge mixture (and hence maximum expected concentrations in ground
water) would be significantly lower than those modeled in Section 2.1.2.
A fourth difference between land application and landfilling of sludge is that the
concentrations of TCDF in land-applied sludge tend to be lower than those in sludge that is
landfilled. Maximum reported concentrations in landfilled sludge are 520 ppt for TCDD and 6,740
ppt for TCDF. The highest reported concentrations for land-applied sludge are 681 ppt and 1,300
ppt for TCDD and TCDF, respectively. Although maximum TCDD concentrations are comparable
for the two practices, maximum TCDF concentration in land-applied sludge is about five times lower
than that in landfilled sludge. Since TCDF dominates the risks estimated in Section 2.1.2, one would
expect that the lower TCDF concentrations in land-applied sludge would result in lower risks from
this sludge management practice.
A fifth difference is that the quantity of sludge applied to a hectare of treated land is generally
lower than the quantity deposited in a hectare of landfill. If so, and if the concentrations of TCDD
and TCDF in land-applied sludge are lower than the concentrations in landfilled sludge, then it
would follow that loadings of these contaminants to ground water beneath a land application site
would probably be lower than those beneath a landfill, all else equal. Based on estimated cumulative
application rates and sludge quality for land application sites (described by Table 2.4.A), the average
loading of TCDD per hectare of treated land will range from less than 0.01 percent (Maine) to about
40 percent (Mississippi) of that assumed for the "high risk" landfill scenario. The corresponding
range for TCDF is from less than 0.01 percent (Pennsylvania and Maryland) to 1.3 percent (Georgia).
Since almost all of the human health risk associated with ground water contamination from TCDD
and TCDF is caused by TCDF, these loadings suggest that estimated ground water contamination
from land application is likely to be lower than that from landfills, if all conditions other than
contaminant loading are held constant.
Although a comparison of contaminant loading per hectare for landfills and land application
sites would suggest lower risks through ground water than those estimated for landfills, the "high risk"
methodology used for estimating landfill risks actually ignores sludge quantity when estimating
ground water contaminations. Instead, it predicts maximum water concentrations based solely on
contaminant concentrations in the sludge, organic carbon content, and equilibrium partition
coefficients for TCDD and TCDF. Still, the theoretical basis for that methodology would suggest
lower risk estimates from land application sites for two reasons. First, if sludge concentrations of
225
-------
TCDF are lower for land application sites then for landfills, then the maximum expected
concentration of dissolved contaminant in leachate beneath a land application site would be
correspondingly lower than that expected beneath a landfill. Second, if sludge is soil incorporated
(as in Mississippi, Ohio, Pennsylvania and Wisconsin), then contaminant concentrations in the mixed
sludge-soil layer will be significantly lower than the concentrations reported for pure sludge. It
follows that maximum concentrations of dissolved contaminant in that layer will also be lower. For
land application sites in which sludge is not soil incorporated, the maximum concentrations predicted
on the basis of sludge quality alone are unlikely to apply to water that has percolated through an
initially uncontaminated soil layer beneath a land application site.
The remainder of this section will argue that risks to human health through ground water
contamination beneath land application sites are lower than those estimated for the "high risk"
scenario examined in Section 2.1.2, and are therefore too low to justify more detailed evaluation.
Each of the seven states in which land-application is known to be practiced will be considered
individually.
In Georgia, sludge is applied to forest land at a relatively high rate of about 2,000 dry metric
tons per hectare. Since TCDD and TCDF concentrations in the sludge are significantly lower than
those assumed for the "high risk" landfill scenario, and since application areas are assumed to be
comparable to those assumed for landfills, there is no reason to believe that risks from these sites
could be as high as those estimated in Section 2.1.2. In Maine, sludge is applied to forest land at
relatively low rates of application. If, as a worst case, the entire 450 hectares of treated land were
modeled as a single site, then ground water contamination at a well 200 meters from the site would
be expected to be about three times as high as those estimated for a similar site of only about 25
hectares. But the concentrations of TCDD and TCDF in the Maine sludge are more than one and two
orders of magnitude lower, respectively, than those evaluated in Section 2.1.2. Hence risks from this
site should be significantly lower than those estimated for the "high risk" landfill scenario.
In Maryland, sludge is applied to strip mined land, sometimes at depths as great as 25 meters
for filling open cuts. This depth exceeds the maximum of six meters assumed for landfills, and might
conceivably result in direct contact between sludge and ground water. It should be noted, however,
that the "high risk" landfill scenario discussed in Section 2.1.2 assumes that leachate from the landfill
enters directly into the aquifer (without passing through a thickness of intervening soil), and that
contaminant concentrations in that leachate are limited only by sludge concentrations and partition
coefficients for TCDD and TCDF. Although sludge may be applied at great depths in localized areas
maximum predicted leachate concentrations should not exceed those estimated on the basis of sludge
concentrations and partition coefficients, as explained in Section 2.1.2. Since the concentration of
226
-------
TCDD in the Maryland sludge "is more than an order of magnitude lower than that assumed for
landfills, and the concentration of TCDF is lower by a factor of more than eighty, risks of ground
water contamination from the Maryland sites should be significantly lower than risk estimates derived
for landfills. Sludge is also applied to strip-mined land in Ohio. As in Maryland, the relatively low
concentrations of TCDD and TCDF in the land-applied sludge more than offset possible increases
in risk associated with a large application area.
In Mississippi, sludge is applied yearly at moderate rates to a relatively large total area of
agricultural land. The Mississippi plant reported a zero concentration of TCDF in its sludge,
however, and since TCDF tends to dominate risk estimates through the ground* water pathway,
estimated risks for these sites will be low. In addition, sludge is soil incorporated at this site,
reducing sludge concentrations by a factor of forty or more. Maximum dissolved concentrations in
water percolating through the treated soil should therefore be significantly lower than those estimated
for landfill leachate. In Pennsylvania, sludge is applied yearly to agricultural land at relatively low
application rates. Incorporation of the sludge into soil probably reduces TCDD and TCDF
concentrations in the sludge-soil mixture by a factor of 200 or more, resulting in relatively low soil
concentrations, and hence lower risks than the estimates derived for landfills.
Finally, sludge is applied to a relatively large area of forest land in Wisconsin (about 1000
hectares). The sludge is applied only once to each treated area, at a relatively low, rate of about 40
metric tons per hectare. Since the large application area is expected to increase risks by no more than
a factor of four (relative to the 25 hectare site considered for landfills), since the sludge
concentrations reported for the Wisconsin plant are roughly five and four times lower than maximum
concentrations of TCDD and TCDF respectively in landfilled sludge, risks from treated areas in
Wisconsin can be expected to be lower than the estimates derived for fandfills.
For these reasons, the risks estimated for the "high risk" landfill scenario provide an adequate
upper bound of possible risks through the ground water pathway of exposure for land application of
sludge. In fact, there are several reasons to believe that ground water risks from land application
would be significantly lower than those estimates for landfills. Since health risks from landfills were
estimated to be 10"6 or less for the ground water pathway of exposure, risks through ground water
contamination from land application do not appear to justify more detailed evaluation.
227
-------
2.4.7 Estimates for Exposure and Risk from Ingestion of Drinking Water from Surface Water
Sources
Where pulp and paper sludge is applied to land, particles of sludge or soil from the surface
can be transported by erosion to nearby lakes or streams. If humans consume water from these lakes
or streams, they may be exposed to TCDD and TCDF from the land applied sludge. This Section
discusses methods used to estimate the extent of this potential exposure, and its associated risks to
human health.
The methodology to estimate exposure consists of three general steps. First, based on sludge
concentrations of TCDD and TCDF, local topography, land use and other factors, it estimates
contaminant concentrations in sediments and surface water. Second, it uses these estimated
concentrations, assumptions about individual ingestion of drinking water, and assumptions about the
bioavailability and the cancer slope factors of TCDD and TCDF, to estimate individual health risks
for humans potentially exposed. Third, it combines these results with estimates of the size of exposed
populations to derive estimates of total human health risks to the U.S. population. Each of these steps
will now be discussed.
Methods for Estimating TCDD and TCDF Concentrations in Surface Water
4
Details of the methods used for these calculations are presented in Appendix B. In general,
the methods are adapted from U.S. EPA (1985a), and use the Universal Soil Loss Equation, together
with estimates of sediment delivery ratios, to estimate the fraction of a lake or stream's sediment that
originates from the SMA. By multiplying this fraction by the original concentration of TCDD and
TCDF in sludge or soil particles on the SMA surface, the methodology derives estimates of the
concentration of contaminants in the sediment. This contaminant load is then partitioned between
adsorbed and dissolved phases, based on the assumption of equilibrium partitioning between the two
phases.
Methods for Estimating Human Dose of TCDD and TCDF from Ingestion of Drinking Water from
Surface Water Sources
Water concentrations are multiplied by human consumption of water and a bioavailability
factor to yield the estimated human dose of TCDD or TCDF. This estimated dose is divided by
body weight to derive an estimated dose of contaminant per unit of body weight per day:
Cw Qw BAH
Dose,
w
BW
228
-------
where:
Cw = Concentration of contaminant in water (mg/liter),
BAH = Unavailability of TCDD or TCDF from ingested water (unitless),
BW = Human body weight (assumed to be 70 kg),
QH = Individual's consumption of water (liters/day),
DoseH = Dose of contaminant from consumption of water (mg/kg/day).
Methods for Estimating the Size of Exposed Populations
The size of the population exposed to contaminated water is estimated by first multiplying
the area of the drainage basin above each SMA drainage point by the estimated population density
of that area. This estimated population is then multiplied by the fraction of the population that
takes its drinking water from surface supplies:
PEU = AB PD PSW
W D
where:
PEU = Population exposed to contaminated water;
AB ป Area of the drainage basin (ha);
PD * Population density for region of land application site (persons/ha);
PSW ป Percent of population served by surface water.
Data Sources and Model Inputs
The values used for each model input are summarized in Table 2.4.M and Table 2.4.N for a
typical and the most exposed individual, respectively. The following sections describe each input
and document the data sources used to obtain values for each model input.
Data Sources and Model Inputs for Soil Concentration
Eight mills in seven states are currently land applying sludge and are considered in the surface
water pathway. Risk estimates are calculated separately for each group of land application sites
associated with a particular state. State-specific data includes estimates of: sludge concentrations; soil
incorporation depth; area of the application sites; type of application (e.g., forest, agriculture, mine
reclamation); and distance to the nearest stream.
229
-------
m
>.
ID
X
J=
4-
ID
0.
4)
4-
<0
X
4)
O
ID
L,
3
C
O
4-
ID
u
w
a
a
<
o
c
o
_j
i
^B
ID
3
o
.
>
o
C
1e
u
"a
h-
o
*
m
o
3
a
9>
u
41
4-
i
L.
IO
a.
o
c
ซJ
in
ง
4-
o.
3
in
in
<
X
*r
CM
O
JO
i
c
0
4-
Q
c
a
a
X
5
(ft
4)
i
j=
O)
x
4)
4.
ID
E
4-
in 4-
4) in
00 O
|
4)
V 1
3 ID
a. L.
C ID
a.
~
O) -
c Q
O
+- o 4-
IO ( C
E i o
co 6
+- in
in in
LU r-ป D ~^
= in 4-
* in *-
- 1*1 < (0
" L.
00 4) O
CO CM L.
at 3
o m co
ซ- 4- o co
a. at
< 4) X
0. L. LU
UJ 3
m a.
O < 3
uj O O
O
c
f*
%
ป*
3
ซซ
8
L.
4>
>
3
4)
>
4.
0
ซ
O)
O
>
<
2ฃ
O)
C 3 C
+- r 4-
O ป 0
in
O 4) "O
4) in 4)
L. in L.
g. o Q-
L.
3
O
4-
C
O
u
ซ
m
3
m
E
L.
ID
m
4)
E
3
in
in
a
4- JC
X W
5 z
ซ X
ซ o
M -1
\o -
ป - -
c
o
4-
ID
4) E
L. ID
3 ^
4- L)
0 - 4> 0
~ 3 4- L.
4- U 10 4-
a 4> 4> a
L. L. L. C L,
O) O
s < u. x =
O a.
: =
r-
r*^
in
Ifc
j;
8
J3
T3
C
IO
X
<
Q
V)
3
* ^
in
u
J=
.ซ
en
c
u
ง
4-
ง
U
w
-
4-
/)
4)
O
u.
in
4)
u
4-
U
a
L.
&
_
c
O
4-
IO
ID
U
4)
L.
4)
C
Z
4)
E
3
m
m >
a L.
L. ID
mo 4-
O > m 3
~ a : 4> xs
L. E JkC 4- *
a m L.
c c . in 4-
4) L.
DO 4> L.
m 4- x .c 4>
C O 4-
j= : in ID
en c a E
4) E m
.c o .c 3
: L. t in ID
o in
e a o
c in 4-
ซ ซ > O e
4- a
: x L.
4- m L. a c
in 4) c
ป ซ 4- ซ ID
(O J= ID L) L.
: 4- x m -o
S
*%
in
at
CM
JZ
_
O
~*
in
at
01 co e
c CJ> in
m
4> ^~ 4>
LU O
0 X L.
3 4- O CO
i
d
^
o
d
o
o
c
O
JO
L.
ID
U
u
c
ID
O)
L.
O
4-
ฃ o
u
3
CM
CM
CM
0
4-
d
E
3
in
1
L.
4)
4-
ID
3
K
^^
CO
co
0\
^x
2
1
CO
4h
pป.
IO
CM
O
-t-
'
^*
>.
ID
o
in
4)
**
4-
C
4)
in
in
l>
in
in
<
4-
m ID
O L.
0. 0
X
UJ
a.
f 3
uj O
O O
230
-------
o
e
^
Q
U
5,
(i*
fc
*^
ซ
c
3
a
ง
a
i-
a
a.
ID
tf)
8
f\
ซ
4-
s
*
z
15
0
3
a
^ .
0
U
C
t_
o
0
IT
c
0
4-
a
c
a
o.
x
x
4-
ai
^_
x
4-
E
in 4-
eo
O
c
4-
0
3" 1
o. L.
c a
a.
O O in
E u. 0)
O 3 4-
aur^oooo ซin^
O > ซ oo > O. C T3 C
*- jQ O O 0 C
O a. o in x a
T> O U
4- 0 'O in O 3
c E 0 4- a u.
0 3 in o
u in O 4- f i i
L. c a. 10 in 4- i i
0 o x JE a
a. o 0 4- u. t.
^
a.
LU
CO
""1
+-"
tn
ฃ
o
c
IO
0
o
c
a
L.
0
4-
10
t
"5
0
V.
in
0
3
U
C
= .
T3 OO
C OO
IO C^
1
a
Z *
ง |
"S ซ
c c
10
E 0)
IO
4- ฃ
ii
in c
i ~ 0>
x o in ao
o c x
._ a) ._ ,ซ,
Qr x (N
O X
S . 5 *
in
-4- 3 >-
~ ฐ ฐ
01- 4-
E O >- U
0 4-3
E ^ T3
L. O
< O I.
a. z jo a.
LU ID
L. L.
. 4) ._ J)
10
=} a. 10 a.
8
O in
O O>
8
8
8
in
oo
.c
in
T3
0
C +-
O ID
0 ฃ
j: 10 E
>> Q. C
H- jo o o
O a. u
o
+- 0 is in
c E 0
0 3 in
u
-------
(/>
>*,
10
ฃ
4-
IO
CL
0
4-
IO
^B
U
(O
*^
L.
Z
o
4-
10
u
Q.
Q.
o
C
a
i
a
2
^
o
c
~
ซ.
o .
u
a.
H"
a
L
Q
*
in
0
3
a
0
i
o
^
IO
a.
c
IO
v>
o
4-
Q.
3
m
1
U
***
z
^
^S
0
^
a
O
u
c
0)
X
0
a
E
in 4-
0 in
(3 0
X
o
u
0
a. u
c LL.
1 L. Q
ฐ ฃ ฃ
m u
S 10 T3
U. C
- IO
O c
E 4~ O O*
O t-
< L. 1
a. 4- oo ซ
LLI C 4-
0 t^ m
0 3
C K> Ol
O 3
3 U ^
t^
C
c O
V
C ID
O t-
4- C .C
e 0 in
u o
4- C *-
C O
0 U 0
^ . 0
O in ฃ
-
D -
L. O tU
m ra >
EL. ^
E 4- 3
0 *n (^
c3 _Q
^
SI*' 3
in
0
4-
IO
U)
4-
c
10
0
0
L.
t.
O
s.
a
0
<
.
Kซ
1- vO
- 9
Kl
f* iO
*
0
C 4-
^ O 0
0 X
X 4-
4- ซ3 0
E 0
in aa
C 0 Q. *-
01. O 1.
O IO Q. 3
3 in
O in O >
v >
o . c -o
*CL 00
3 O O >
a. 0 i. L.
O a. 00
a. -^ CL m
T3
OO 4- 0 1
O^ CO ^*
0 u. r.
E Oป
>. V. ซ "~
0 IO C
ฃ ฃ IB 0"
a. oo a 4- >
E O> 10 I.
3 10 Z 3
X ซ- 3 =
-------
in
?ป>
ID
2
ฃJ
+
. o
+- t. c
m flj (O
0 at 4-
in
4ป *^ *~
in o ^
.0 0 -o
4- 0
O ID 4-
C 4- ID
a in E
S0 4-
ฃ in
_J 4- 0
O
oo
00 09
(S (M
ซ ซ
m m
00 CO
f* <>4
K> fO
E
(Q
J_
V)
o
a
^
i
^
X
0
0.
a.
ID
ซ^
O
L.
0
ID
0
U
a
u
in
.
m
i_
.
0)
+-
c
X
tft
E
^
ao
(\4
*
00
ซ
m
,^
a.
a.
ซ
in
in
in
m
Z
T3
0
10
E
4_
I/)
0
C
ID
^
a.
o
5
0)
-*-
_>
c
0
.._
O)
c
0
X
0
u
c
ID
in
o
*^
in
>. c
in o
c u
c m
0) ._
Q. -X.
.
0
^
_
3
U
a
o
4-
-ฃ
II
U
2T
in
0
i
233
-------
m
ซ
X
T
0
4-
ซ
U
Q.
Q.
e
_j
"io
o
c
o
1
X
i
s
4-
$
3
O
t.
i
a
o
a.
_
a
(A
e
O
1
3
in
in
z
w
-
4- o 4- in in in ir>
ID K- C O ^ O "*
E i a) u i_
CO E LU ^ LU -*
4- m o o
m in o o
LU ^ป Q) .Q O
I in ID T3 ID T3
* ro ^ c ID c (O
_ _ ._ -^- r^ -^
03 41 a >a
oor\ii--~ cr< o;<
oป 3 4- a o
- O "> ป O)CO O)CO
< 4-OO C3 C3
a. L.
< ซ X Q 4-= 4- =
O_ L. LU O- 0ซ
uj 3 in in
in * o. -o ID o u
O<3 um oin
COO.LUO L. 1/1 L. I/)
x X i. Q. 0 Q. O
3UJOO - -i - I
4-
X
4-
o
JJ
c
2 o
^
a
> c c
S .2 ฐ
4- 4-
O .. ID ID
> > (.Cue (-(.1.C
O) O O) O
< : : :
1988), "Estimating
*^
^
0.
UJ
3
a
o
1
k
CM
O
4-
Q)
L.
3
in
O
Q.
X
LU
sure Assessment
a
X
LU
LU
5
4-
10
L.
a
a.
O
0
0)
c
4-
IO
E
4-
m
LU
03
Ol
*^
^
0.
LU
CO
Q
8
03
CM
O
4-
8
d
o
o
ซ3
<0
a>
en
a
ID
01
O
Q.
E ~
3 >
in ID
c ~o
O in
L. ซ
I) 4-
4-
ID
234
-------
10
&
4-
3
Q.
C
0
U
c
0
u
0
0
a
0
4-
a
9
a.
x
0
10
0
i
a>
X
4-
a
E
10
L.
0
4-
to
L.
ซo
O.
L.
O
LL.
T
~ ^ E ^
L. -t- Q m ^ c
"*" E O ซ r> E c
ฃ t in Oi in O
o> 4- i 10 in
~moo< ~04-r
X LU in 10
s r~ ซ >. in c c
^ ซ L. ^^ ID ^C *^ IO
01 ro3-t- L. e 01
aoao >in>^ j=^io
- ao *- 3
o 8
E a
0
E Jt
< o
a x
LU
L.
in a.
IO
3 a.
8 in
Ol
S3
..
o
4-
Q.
O
10 "
O t- '
< 0 C.
4- in
C ID
<0 2 U.
T
C
in
c
X
0
5
^K
0
4-
^_
Q
ID
^K
^
ID
10
C^
00
ro
CM
X
~
*
in
u
3
o
O
L.
Q.
L.
4>
Q.
a.
i
o
en
o"
E
11
E
f^.
00
0>
^
^
a.
LU
cn
in
O
in
0
4-
tm_
*
C
ง
4-
a
t,
4-
c
U
J
L
O
U)
L.
O
4-
u
ID
LL.
C
ป_
4-
ia
i_
4-
C
O
u
o
u
c
c
O
4-
a
L.
4-
C
0
o
I
in
>
;
ป
U.
a
CJ
i
TJ
C
^
ซ
CN
C.
in
0
"o
4_
in
3
Ol
3
235
-------
in
ID
z
jC
4~
ID
Q-
4)
^
IO
2
4}
U
IO
ซ^
U
3
CO
, ,
I
4-
10
u
Q.
&
<
o
c
ID
_)
1
a
a
>
o
c
o
4)
in
O
Q.
X
tAj
2
5
4-
fc
ป m
a. 4) a>
i:
4) 4- C
m
C.
4-
in ~
a.
O Q.
+- -o
ID cm
4> > ID in
o in o> 4>
c i. t. c >ป m
ID 4) O !- m
t) 4> O Z Z Z
S
0 ~
(D
.C C
4-
>- L.
n a)
4-
C 4)
4) E
Ol
L.
c O
4)
ฃ ซ
Z "
4. i
o *~
c
ID 4)
4- 4-
in 4-
J3 ID
ID 4-
m
4)
ฃ L.
+- O
O
f\
S
a
c
ID
>
>.
m
c
1
2
-
-
C
v>
o
u
in
2
236
-------
The method for calculating soil concentration from sludge concentration is discussed in
Appendix A. The application rates and soil incorporation depths for the sites receiving the sludge
of the seven land applying mills are displayed in Table 2.4.A along with soil and sludge
concentrations. Sludge concentrations are taken from the U.S. EPA 104-Mill study. The remaining
data were obtained from discussions with state officials familiar with the site land application
practices except for the following data. Application rates for one mine site and one forest site were
obtained from Keenan (1989) and Martin et al (1987), respectively. Sludge applied to forest land,
though surface-applied, is assumed to be soil-incorporated to a level of 2.5 centimeters to reflect the
accumulation of duff and the effects of activity on the forest floor over the 70 year period of
analysis.
Data Sources and Model Inputs for Sediment Concentration
As explained in Appendix B, the parameters necessary to calculate sediment concentration
(before partitioning) from soil concentration are: site area, site sediment delivery ratio, drainage
area, drainage area sediment delivery ratio, and the Universal Soil Loss Equation parameters.
The soil concentrations are discussed above. SMA areas are listed in Table 2.4.A and Table
2.4.B for each of the states. These areas were obtained from discussions with state officials familiar
with the SMA characteristics. In some instances, the area was calculated by dividing total land
applied sludge quantities by the application rate per unit area. Maine's SMA area was available only
as a range. The midpoint is taken as the "best estimate". At least 25 hectares are estimated to be
receiving sludge in the State of Maryland, with an upper limit of 40 hectares. The "best estimate"
assumes that 25 hectares are sludge-amended.
The calculation for the SMA sediment delivery ratio, as shown above, depends on the overland
distance between the SMA and the water body. In the majority of cases, this information was
unavailable. The State of Maryland, however, gave this figure as approximately 2 miles (or 3218
meters). This distance was used as the "best estimate" of typical risk for all states. Three states had
permit requirements for the minimum distance between an SMA and a waterway. When available,
these limits were used for the "high risk" estimate for typical individuals and the "best" and "high" risk
MEI estimates. Where no permit limits exist and actual distances are not known, the "high risk"
estimate for typical individuals and "best" and "high" risk MEI estimates assume that the SMA is
located 1 meter from the stream. The values used to estimate the distance between the SMA and a
-Stream are_preseJited in Table 2.4.M. for typical risk and 2.4.N for MEI risk.
237
-------
The watershed sediment delivery ratio is dependant on the area of the watershed. The "best
estimate" and "high risk" typical individual estimate assume the water body receiving SMA runoff
is a major stream with a watershed area of 5,000 square miles. This estimate is based on U.S. EPA
Geographic Exposure Modeling System (GEMS) estimates of drainage area for major streams (U.S.
EPA, 1989c). The "low risk" typical individual estimate assumes that the receiving water body is a
smaller tributary with a watershed area of 500 square miles. The watershed in the MEI scenarios is
assumed to be 10,000 acres (approximately 40 square miles). This area corresponds to a relatively
small stream (U.S. EPA 1988a).
The cover management and support practice variables from the USLE equation are determined
as a ratio of SMA to watershed. This analysis assumes that forest and mine reclamation SMAs will
be surrounded by land with similar cover and will not have any support practices. The "C" and "P"
ratios are therefore 1:1. However, agricultural land may differ in its cover from the surrounding land
and practices may be in place to slow the runoff water and thus reduce the amount of soil it can
carry. "C" values for cropland range from approximately 1% to 85%, averaging approximately 40%
(Science and Education Administration, 1978). In other words, the approximate average soil loss
from crop land under specified conditions is 40% of the corresponding loss from clean-tilled,
continuous fallow , "C" values on permanent pasture, range, and idle land range from approximately
0.3% to 45%, with an average of approximately 10%. "C" values on undisturbed woodland range from
0.01% to 9% with an average of approximately 0.3%. On woodland that has been grazed, burned,
harvested, or re-established after harvest, the approximate average "C" value is 7% (Science and
Education Administration, 1978).
In this analysis, the "low risk" typical individual estimate assumes that the watershed area for
streams receiving agricultural runoff is largely agricultural land; the "best estimate" for typical
individuals assumes that the watershed area is largely pasture land; and the "high risk" typical
individual estimate assumes that the watershed is largely woodland that has been grazed, burned,
harvested, or re-established after harvest. The MEI risk estimates assume that the watershed area
is largely disturbed woodland. The resultant "C" ratios are displayed in Table 2.4.M. for typical
individuals and 2.4.N. for the MEI.
The "P" factor in the USLE is the ratio of soil loss with a specific support practice to the
corresponding loss with up-and-down-slope culture. Tillage and planting on the contour result in
a "P" factor of 0.75. Stripcropping, a practice in which contoured strips of sod are alternated with
equal-width strips of row crops or small grain, results in a "P" factor of 0.6 (Science and Education
Administration, 1978). This analysis assumes Stripcropping for the "low risk" typical individual
238
-------
scenario, contouring for the typical individual "best estimate", and no control practices for the "high
risk" typical individual scenario. The MEI analyses assume that no control practices are in place.
Data Sources and Model Inputs for Deriving the Partition Coefficient
Koc, the partition coefficient between water and organic carbon, is multiplied by the fraction
of organic carbon in the sediment to obtain Kd, the partition coefficient between sediment and water.
The Koc value for TCDD used in this analysis is 1 x 107 (Jackson, 1985 in EPA, 1987c). The Koc
value for TCDF is 3.5 x 104 (CHEMEST procedure in GEMS, U.S. EPA, I989c). The organic carbon
content is assumed to be 0.04 in the "low risk" typical individual estimate, 0.01 in the typical
individual and MEI "best estimate", and 0.001 in the "high risk" typical individual and MEI estimate
(U.S. EPA, 1988a).
Data Sources and Model Inputs for Estimating Human Dose
Individual water consumption is assumed to be 2 liters per day (U.S. EPA, 1988a). The
bioavailability of ingested water is assumed to be 100 percent (FDA, 1989).
Data Sources and Model Inputs for Estimating the Size of the Exposed Population
*
The size of the exposed population is estimated by multiplying the watershed area of the
contaminated stream by the population density of the regions in which the SMA's are located. To
accurately estimate the size of the population exposed on a site-specific basis, it is necessary to know
the stream into which the SMA runoff drains, the dilution and dispersion pattern of the contaminant,
and the patterns of water withdrawal from the water body. In the absence of this site-specific
information, the size of the exposed population is estimated as follows.
The typical individual "best estimate" assumes that the receiving stream for each SMA is a
major stream with a 5,000 square mile watershed area. It is further assumed that the size of the
exposed population increases with stream size, and that stream size increases with drainage area i.e.,
the larger the drainage area, the more people are likely to use the stream for a drinking water supply.
To capture this relationship, the size of the exposed population is approximated by the number of
persons expected to live in the area of the drainage basin containing each landfill. The area of each
basin is multiplied by the average population density in the regions through which the waterway
flows. Population density is determined by averaging populations for the regions of the United States
where sludge is land applied and dividing by the area for these regions. Regional populations were
considered rather than state populations because the contaminated waterways are not limited by state
239
-------
boundaries. The average population density for regions where land application sites are located is 173
people per square mile (U.S. Department of Commerce, 1987).
Only a portion of this population will rely on surface water for their drinking water. The
estimated population exposed is therefore reduced by multiplying by the average percentage of
population served by surface water. Forty-six percent of the population is assumed to be served
by surface water (U.S. Geological Survey, 1985).
This analysis assumes that the entire exposed population ingests water at concentrations
estimated at the "point" of entry of the SMA runoff into the stream. Since the population exposed
will inhabit an area of approximately 70 by 70 miles, this assumption is conservative. In reality,
dilution and dispersion of the contaminant will probably have occurred before much of the
population is exposed. However, in the absence of additional site-specific information, it was not
possible to calculate exposure more exactly.
One test of the reasonableness of the analysis' assumption about the size of the exposed
population is to compare the amount of drinking water assumed to be withdrawn from the
contaminated stream with the stream's flow rate. If the stream can not supply the quantity of water
that the analysis assumes is ingested then the assumptions must be reexamined.
>
The "best" risk estimate for the population assumes that a water body with a drainage area of
5000 square miles is contaminated. A water body with this drainage area would be a relatively
major stream, receiving runoff from an area approximately 70 by 70 miles. Several streams located
near paper mills have a drainage area of a few thousand square miles. For example, Raccoon Creek
at Granville, OH has a drainage area of 8,270 square miles (GAGE data base, U.S. EPA).
The U.S. Geological Survey (USGS) has studied the relationship between a stream's drainage
area and mean annual flow rate using regression analysis. On a national scale, USGS found drainage
area to be the most significant variable influencing flow rate; other variables, particularly average
annual precipitation, were also sometimes significant (personal communication, USGS). Since annual
precipitation varies substantially throughout the United States, drainage area alone is not a precise
predictor of stream flow. However, the general relationship between drainage area and stream flow
can be estimated for regions of the country.
For regions east of the Mississippi, the range of stream flows associated with a drainage area
can be approximated from the following equations (personal communication, USGS).
240
-------
Low Estimate
MAP = DRA
High Estimate
MAP - 1.5 DRA
where:
DRA = Drainage area (square miles), and
MAP = Mean annual flow rate (cubic feet per second or cfs).
Regions west of the Mississippi will exhibit greater variability in stream flows with relationship to
drainage area, with extremely arid areas approaching zero cubic feet per second per square mile of
drainage area. However, moderately arid regions are likely to exhibit a relationship between stream
flow and drainage area that can be approximately represented by the following equation:
MAP = 0.5 DRA
Using these formulas to predict mean annual flow rates for a stream with a 5,000 square mile
drainage area yields 5,000 to 7,500 cfs in relatively humid (generally eastern) states and about 2,500
t
cfs in relatively arid (generally western) states. Converted to liters per year, the mean flow rates are
4.5 x 1012 to 6.7 x 1012 liters/year for humid states. Arid states have a mean annual stream flow rate
of approximately 2.2 x 1012 liters per year associated with a stream from a 5,000 square mile drainage
area.
The flow rate for a stream with a 5,000 square mile drainage area can be compared with the
quantity of drinking water assumed to be withdrawn from the stream to evaluate the feasibility of
assumptions about the size of the exposed population. To estimate drinking water withdrawals, the
exposed population is multiplied by the quantity of water consumed per person. The analysis assumes
that a person consumes 2 liters of water per day. The population exposed varies between different
types of sludge management, based on the population density of the regions of the country in which
the facilities are located. The exposed population also varies according to the percent of the
population that receives its drinking water from surface water. For surface impoundments, the
population exposed to contaminated water is estimated to be 80 persons per square mile of drainage
area.
241
-------
A comparison of water withdrawals and stream flows for land application sites suggests that
a relatively large population might use the affected water. The water withdrawals will be from a
stream with a relatively high stream flow:drainage area ratio, since land application sites are located
in the eastern United States. Water withdrawal for drinking is calculated by multiplying a 5,000
square mile drainage area by 80 exposed people per square mile. This yields an exposed population
of 400,000 people. Total surface water withdrawal for drinking, at 2 liters per person per day, is 2.9
x 108 liters/year. Comparing the water withdrawals to the stream flow shows that between 0.004 and
0.006 percent of the stream flow is withdrawn for human consumption.
It is likely that the population receiving its drinking water from contaminated surface water
will receive the remainder of water for domestic uses from this same source. Therefore, it is also
informative to compare total water withdrawn for domestic use with the stream flow. Average
domestic water use is 78 gallons, or 295 liters, per person per day (USGS, 1985). This quantity
multiplied by a population of 400,000 people equals annual domestic use of 4.3 x 1010 liters per year.
Total domestic use as a percentage of stream flow is between 0.6 and 0.9 percent.
In evaluating the plausibility of this result, one should note that 77% of water used
domestically is return flow (USGS, 1985). This means that the water reaches a ground- or surface-
water source after release from the point of use and thus becomes available for further use. The
above discussion suggests that the assumptions in this analysis about the size of populations exposed
to surface water area are plausible.
2.4.8 Estimates of Exposure and Risks from Ingestion of Fish from Surface Water Sources
Where pulp and paper sludge is land applied, particles of sludge or soil from the SMA surface
can be transported by erosion to nearby lakes or streams. If the sludge contains TCDD or TCDF,
then those particles can carry these contaminants to the surface water bodies. Fish living in the lakes
or streams can take up sludge contaminants into their tissue; if humans then consume those fish, they
can be exposed to TCDD and TCDF.
This Section discusses methods used to estimate the extent of this potential exposure, and its
associated risks to human health. The methodology is quite similar to that discussed in Section 2.4.7,
in that both methodologies begin by estimating sediment concentrations of TCDD and TCDF in water
bodies as a result of runoff from land application sites. Once sediment concentrations have been
estimated, however^ the methodology departs from that described in Section 2.4.7, and uses fish to
sediment bioconcentration factors and estimates of human consumption of fish to calculate
contaminant doses to humans. The last step in the methodology involves estimating the size of the
242
-------
exposed population, combining these results with estimates of individual dose and health risk to
derive total health risks to the entire exposed population. Each of these steps will now be discussed.
Methods for Estimating TCDD and TCDF Concentrations in Surface Water
Methods used for these calculations are described in detail in Appendix B. In summary, the
methods are adapted from U.S. EPA (1985a), and use the Universal Soil Loss Equation, together with
estimates of sediment delivery ratios, to estimate the fraction of a lake or stream's sediment that
originates from, the SMA. By multiplying this fraction by the original concentration of TCDD and
TCDF in soil particles from the land application site, the methodology derives estimates of the
concentration of contaminants in the sediment. These sediment concentrations can then be used to
estimate contaminant concentrations in the tissues of fish.
Methods for Estimating the Concentration of TCDD and TCDF in Fish Tissues, as a Function of
Sediment Concentrations
Methods for estimating contaminant concentrations in fish are discussed in detail in Appendix
B. Based on the assumption that sediment concentrations are the best predictor of fish concentrations
of hydrophobic compounds like TCDD and TCDF, the methodology uses empirical fish to sediment
bioconcentration factors to estimate concentrations of contaminant in freshwater fish as a function
*
of concentrations in stream or lake sediment. As explained in Appendix B, the concentrations of
TCDD and TCDF in the muscle tissues of fish (consumed by humans) are considered to be fifty
percent lower than the whole body concentrations of these contaminants.
Methods for Estimating Human Dose of TCDD and TCDF from Ingestion of Fish
Estimated contaminant concentrations in fish tissue are multiplied by an estimated amount
of fish consumed daily and a bioavailability factor to yield human dose of TCDD or TCDF. This
estimated dose is divided by body weight to derive an estimated dose of contaminant per unit of
body weight per day:
Cf Qf BAF
Dose,
'F
BW
where:
BAF = . Bioavailability of TCDD or TCDF from fish (unitless)
BW = Human body weight (assumed to be 70 kg)
243
-------
CF = Concentration of contaminant in fish tissue (mg/g)
QF = Individual's daily fish consumption (g/day)
DoseF = Dose of contaminant from consumption of fish (mg/kg/day)
Methods for Estimating the Size of Populations Exposed to TCDD and TCDF through Ingestion of
Fish
The size of the population exposed to fish containing TCDD and TCDF is estimated by
multiplying the area of the drainage basin containing each land application site by an estimated
population density of the regions containing the sites.
PEF = ABPD
where:
PEF = Population exposed to contaminated fish
AB = Area of the drainage basin (ha)
PD = Population density for region of land application site (persons/ha)
Data Sources and Model Inputs for Estimating Human Dose
A contaminant dose from ingestion of contaminated fish is estimated for three populations:
typical individuals, an MEI, and sport fishers. As an input to the dose calculation, consumption of
freshwater and estuarine fin fish and shellfish is estimated for these three populations. Numerous
studies report quantities of fish consumed by humans but most include marine fish. U.S. EPA has
cited 6.5 grams/day as the average freshwater fish consumption (U.S. EPA, 1980, in U.S. EPA
1988b). The Food and Drug Administration (FDA) has estimated 16 grams per day as the upper 90th
percentile ingestion rate of freshwater fish in the Great Lakes area (U.S. EPA, 1988c). This analysis
assumes fish consumption of 6.5 grams per day for a typical individual in the "low risk" scenario and
"best estimate" and a consumption rate of 16 grams per day for the typical individual "high risk"
exposure estimate.
Since these overall averages include a large proportion of individuals who eat no freshwater
fish at all, particular populations may consume larger quantities. In particular, sport fishers are
likely to consume fish at a higher rate than a typical individual. The "best estimate" and "high risk"
sport fisher scenarios incorporate the assumptions used in the population "best estimate" and "high
risk" scenarios, respectively except they use different rates of fish consumption and a smaller exposed
population to reflect the typical behavior of sport fishers. This analysis assumes sport fishers
consume 48 grams of fish per day in both the "best estimate" and the "high risk" scenarios. This is
244
-------
the median consumption rate for sport fishers in Michigan reported by Humphrey (1983). The MEI
analysis uses the 90th percentile consumption rates of active sport fishers, 100 grams per day, to
represent MEI consumption rates in the "best" and "high" risk estimates (Humphrey, 1976).
Bioavailability rates for the contaminants consumed in fish are taken from U.S. EPA (1989c).
TCDD and TCDF are assumed to be 85, 90, and 95 percent bioavailable for "low", "best", and "high"
calculations, respectively. Contaminants consumed with fish are assumed to be 95 percent
bioavailable in both the "best" and "high" MEI risk estimates.
Data Sources and Model Inputs for Estimating the Size of Exposed Populations
This analysis assumes that all fish are consumed regionally. An alternative would be to assume
that the fish are distributed nationally. In that case, the percent of the freshwater fish each person
consumes from the contaminated stream might be calculated by calculating the ratio of the drainage
area of the contaminated stream to the drainage area of the entire United States. This percentage
could then be used as the percent of contaminated freshwater fish consumed by the entire U.S.
population. Instead, the current methodology uses drainage area to determine the proportion of U.S.
citizens who are exposed to the contaminated water and fish. This population is then assumed to
consume 100% of their freshwater fish from the contaminated stream. These two methods will yield
similar estimates of population risk since the analysis assumes a linear dose-response relationship.
*
As previously discussed, sizes of exposed populations are estimated by multiplying estimated
watershed area by estimated population density. To accurately estimate site-specific populations
exposed it is necessary to know the stream into which the SMA runoff drains, the downstream uses
of the surface water (e.g., fishing, drinking), and the distribution of the fish that are caught. In the
absence of this site-specific information, the size of the exposed population is estimated as follows.
It is assumed that the size of the population exposed will be positively correlated with stream
size and that stream size will be positively correlated with drainage area; the larger the drainage area,
the more people are likely to receive their drinking water supply from the stream. To quantify this
relationship, population exposed is modeled as a function of drainage area. Each unit area of the
watershed is multiplied by the average population density for the regions through which the
waterways flow to yield population exposed. For land application sites, the population density is 173
people per square mile.
In the "best estimate" and the "high risk" scenarios for typical individuals, it is assumed that
the receiving stream for each SMA is a major stream with a 5,000 square mile watershed area. In
245
-------
the "low risk" estimate for typical individuals, the drainage area is assumed to be one-tenth of this
size, or 500 square miles. The drainage area in both MEI scenarios is assumed to be about 40 square
miles. This corresponds to a relatively small stream (U.S. EPA, 1988a).
To calculate the sportfisher population exposed to contaminated fish, the total population
exposed is first calculated as described above. The percentage of U.S. non-metropolitan households
consuming "home produced" fish and poultry is estimated to be 7.1% (U.S. Department of Agriculture
1978). The percentage of these households consuming "home produced" poultry is estimated at 3.4%
(U.S. Department of Agriculture 1978). If the populations home-producing poultry and fish do not
overlap then 3.7% of the population home-produces fish. To estimate the population consuming fish
at the median sportfisher level of 48 grams per day, the total population exposed is multiplied by
3.7%. This calculation assumes that the family of a sportfisher consumes fish at the same rate as a
sportfisher.
This analysis assumes that the entire exposed population ingests fish at concentrations
appropriate for the "point" of entry of the SMA runoff into the stream. Since the population exposed
will inhabit an area of approximately 70 by 70 miles in the "best" and "high" estimates, this
assumption is conservative, and will tend to overstate exposure and risk. In reality, dilution and
dispersion of the contaminant would have occurred before much of the population was exposed.
t
2.4.9 Summary of Results
Human exposures associated with the land application of sludge are summarized in Table
2.4.O. The risks resulting from these exposures are summarized in Table 2.4.P. Typical risks to
individuals are low through all pathways analyzed for this disposal/use practice. Highest typical
individual risks are estimated for persons living on the land application site who are exposed through
direct ingestion vapor inhalation, and dermal contact with contaminated soil. The highest typical
individual risk is associated with the vapor inhalation pathway: the incremental lifetime risk from
this pathway is estimated to be 7 x 10"6. However, because the size of the population exposed
through this pathway is small, the total annual cancer risk resulting from this exposure is estimated
to be only 4 x 10"6 cases per year.
Lowest typical individual risks from the land application of sludge are for individuals in the
general population that consume produce, meat or dairy products grown in sludge-amended land.
The typical incremental lifetime risk from dietary exposure is estimated to be 2 x 10"8. Although the
typical individual risk is low, the populations exposed through these pathways are large. The food
from sludge-amended land is assumed to be nationally distributed, while contaminated surface water
246
-------
a ^T
ป|l
HI
- S f
o
X
u
u ra
(0
o co
o
Q) *-ป
e_ ra
3 C
O E
Q. CO
X * 0)
0) C O)
o -6
o 3
ra
* o
in
41
S?
(U flj
E"S
0) O -*-
O l/l O
i a
o ra
Q. <-* u
c a.
o ra
*-ป
4-
>ป- c
o
3 a
w a
a
o
*-*
0) -Q
3 T3
U) D
a. ra
x c
D>
c a
o ซ->
ซ o
to o _>
a u)
CA o
247
-------
g
~ u.
n 3
U t-
oil
*rf
01 H-
3 00
a.1*""
x i*T
'"fC
1 ฃ
x '3
"S "8
si
ii
M c
|JJ 0
(A tft
CD jjp
~s
f U
OJ
S 1
U 0.
^p
O CO
* a
f^J 3
01
1 ฐ
ง
^*
X
(0 ^
3 3I
'ill
^ x ;s
"- u |>
.0^
g
h-
H
."ง!
* i5.
S E
"^
x
1
8.
41
UJ
-O
ro
'o
X
fO
^
o
1
o
X
i
u -
o ซ->
a a
- u
!.ซ->-
3 ซ-
-
I *Z
E a)
O T5
U X 01
ra
01 -0 C
U 01
3 **
(A (Q CO
x 'i o
o> m u
Sc oi
00)
o o -6
4-1 U) 3
W) L. >
Oi Oi E vi
O> *^ o
C a L. -t-
3 *- 0
o
<\l
'o
X
f\J
V
(M
O
PJ
O
X
<\J
V
1
u ro
o C
Is
o u
0)
O) O
c w
i S 01
COO)
o> a)
t- ฃ 0
ซ- o
ra c
0) 01 O
3 *^
U) X fl)
8.-ฐ"
X TJ
* ฃ &
C <0 ID
.2.5-0
~ E C
w ro ro
OI f
?g X
0 J3
O
*^
O
X
t\l
N.
O
CO
'o
X
01
u
a o>
-*- O)
in ^ O
.C 0
ป Sa
<4- 1- S
'o x"
-JJ-g
0) tQ >
a.H-ฐ
X E TJ
ซD 01 Q>
-M 4-t
ii S
v- CJ **
w 6
u> i- m
01 01 4->
?ซ i
3 0
^-'
O
X
P
O
op
'o
X
ro
it
4-* C
10 3 c
01 L. O
o x *ป
o> JL a
caught by r
ntaminated
land applic
o
ฃ o X
U) -Q
-<::2"8
o a ซ->
4-> 3 a
c
01 01
<- u E
3 (0 a
(A H- 4-ป
aL. C
3 0
X V) U
" c . ..
ง m oi
4) O)
0)
01 f E ซ)
O) (A O
C - l- <*-
*~ *4- ซซ- O
o-
o
A
'o
X
h-
5>
A
op
'o
X
01
0 01
u> T2
.sป
f*^
o
".ง
01 w
o a
1"
la
CO
2 "g
H- a
(U ~"
V)
8. "8
X +*
01 eg
^"i
ro CD
4-* 4ปป
i S
0 0
u.
a
u
ฐฐ. '
w*
f\T
o
0>
3
1
01
?
i
4^
41 S
O
* u
^~
O o
^~ *rf
ป~ 01
+ 3
<= 9
8 |
00 """
C\T
o
01
(A >
a
O
u
h
0
1.
V)
a
X
UJ
o
1^
^
o
Q
Exposure to
<
5
X O
01 <-
1
m en
~ -^
3
01 4^
2 "3
ซ tr co
O UJ (J
Z CO JD
248
-------
sS
w CO
<9
." '*.
Erg*
ll
* CO
M Pป"
CO
~ K!
ฃ N"
4^ ฃ.
a i~
Oi >
\l
O 4->
25
M
'Z a
ซ in
I -
a.
^ a
cu
3
""S
.a
L.
-1 .w
2 w ^
o T co
H- O
CO
CO
u
7
ซi
*^ i
a
1
^O(O ^
(J ^ 4J
'a. ,2 '^
ฃ u
u
u.
i
<9 '**
*ป=
U
**
X
I
i
1
LU
-o "* r^. ^^
i ^i i O*
r- A ซ A
X X
(\J CM
O O
-sT -^3-
* ~ N. ~
O <> O C>
* A A
X ^ X '"'
Kt vf
^ ^
'o & 'o c?
A ป- A
X X
(\J ^ป
0 '5
CO H-
*- O
j: a ป-
4-1 u o c
'5 '*" ' c"
&o ซ
a
o re *- o
re co
4J -g ai -
c c o> O.
o re c a
o - re
c- oi re
01 4J o "D X O
r c Q E a
co o o ~o a
O E u c. cu o
O. re i t- ^ >
x 4-ป oi re
Oi C 0) w V C 4->
^CJjOi DEOICU
re o co re en u
E 'IOC. Q4-I-DC-
i. "- ti a. c 3 0)
OIOH-OL x o a.
QCOOv> UJOCOv^
-O '-
'o c>
ซ A
X
^
0
"^
>O '^
o <>
* A
X
r>k.
OC?
/*" A
X
(M
V*
0 1
C
E
E a
a w
Si-
o u o>
o
a co co
o E *-
4J O O
cu *- c
L. 0
10, O ซ
Q IDU
S. 4J 0 ~
x re Q
oi N a
Q. CJ
ง a i-
re
*3 ro "g C
a c cu
' o a u
re > c_
~* A & *s
*o -^
'o o
ซ A
X
CM
o
^
f -
o <>
A
X
K(
>o ~
'o c?
* A
X
in
ง
^ ง
CO -
01 ~
4J ID
re u
11
4-* (D
a?
re
o
Oi -Q
c_
3 T3
CO 0)
Q 4J CJ
Q. re -*
x C a
oi a
E u
c re i-
O 4-> 01
ซ if S
re u 3 oi
' O
a ' CA c.
.c cu
C 0 *- 0.
CA O ^
249
-------
g
~ 14.
*rf O
19 O
U 1
"S.00
^ *
TH ป
ij
*" 8
(A I
5 ป.
*ฐ"
S ***
X J=
ffl x
II
S 2
si
IS
U (A
4-1 at
f TJ
41 3
09
<^ L.
J 8.
~
s ซ "8 1
C 0 CO C
CO
O t X T3
3 J3
u X U l.
**ป i 4-ป **-
CO
4> T3 C 41
u 41 i_
3 w E 3
eft co co o cy)
p C -M ^i o
IE ง ง x-
4) CD U CJ 41
งC 4) *~ C
O 0) 4J O
- 0 O "S C
4^ (A 3 4) 4-1
co t- ป o in
CD 4) E C/J L. CU
C CO i- H- Q. C
a H- o ^ ~
T3
z
o
z
tv.
o
X
o
V
fซta **
CM
X
o
V
1
CO
c
contami
'o
U)
ง41
en
-3
ro -
CO CD
CD CD O
8 X2;
ro '-
i OJ
0 0
ซ ^-^
X
o
o
o
o
o
NO
eo ซ
x
in
to '^
x
?
CO
d) ^
*-. eu
S|
I'l
4J CO
Sฃ
o o
o
t- a. cu
c 3 ao.
~ IA CO ป^
fO .
i CNJ
O O
r- **s
X
o
o
o
Od
C\J
N. **
i OJ
0 O
X
^j
Kl ^^
oS
X
CA
CO
01 41
L. 0)
a "O
*-
H- IA
3 "S
4J l_
f C
01 41 O
30
O CD w
0 ซ- CO
<- O
^ 3
IA IA
ฃ C |
*J O) ^
41 CO
l. f >
3 IA
m xo
O ป- -Q *ป
a. a
X ' ^ O
4) CO 4) CJ
C *J (-
C O ID
O C <-
4J C
w CO E 41
CA 4) CO O
Ol u C 41
C 4i o a.
(M ~
'o o
ซ~ A
X
N.
o
o
o
o"
o
o
0
CO <-
08:
*^ A
X
(M
CM *"ป
08:
*- A
X
f\J
^rf
. 4(
gf
- w
3 'S
I- C
at o
4) 4^
0 CO
3 O
O '
a &
ID
O "O
*- CO
4( ~*
X V >
V CD
til
CD CO O
41 C 41
- o a.
O O v^
'
,^
>
S
0
1
LU
41
'Z
y\ *"*
0
^. "ez
o
O W "
T CO Cj
.< a
Q. 0
,_ 4^
X "S 4.
- 1 1
Q) ^^ tf)
3 UJ 9-
W) *^
8. x "
X -*
LU IA
"S ซ
4-* '
E u
.^- .^ >
w Q.
ซ X C
UJ h- C
iป_t h.^ ^
UJ (/I 0
oj ro (i
"S "2 1
*-* *^ ป
(0 (Q C
3 "3 1
(A U U I
M (D
f
^^
o
u
H-
0
u
3
(0
1
UJ
<
>
3
4)
D
9 A
CD
1 0
! a
3 CO
J
[1 O
3 -oz
250
-------
is assumed to be used by nearly 3 million persons. As a result, the dietary pathway yields the largest
number of cancer cases per year of any of the land application pathways analyzed (7 x 10 cases per
year). Exposure through drinking of contaminated surface water leads to 2 x 10"3 cases per year.
The "most exposed individual" risk is typically one to two orders of magnitude higher than
typical risks for the first four pathways shown on this table. The MEI and typical individual risk
estimates are close for these pathways because both the "typical" and "MEI" risks from these pathways
are evaluated for onsite residents of two agricultural land application sites. In contrast, MEI risks
are five to six times higher than typical individual risks from the dietary, drinking water and fish
ingestion pathways. For these three pathways, typical risks are estimated for an offsite population,
while the MEI risk is assessed for an individual living on the land application site. The highest MEI
risks are from the ingestion of produce, meat and dairy products grown on sludge-amended land and
from the ingestion of fish caught in contaminated surface water bodies. For the produce and meat
ingestion pathway, the MEI is assumed to be a farmer who raises his/her own meat and dairy
products on sludge-amended agricultural land. For the fish ingestion pathway, the MEI is assumed
to take fish from a relatively small stream from the location in the stream with the maximum
dissolved concentrations of TCDD and TCDF. The MEI risk associated with the ingestion of
contaminated produce and meat is 2 x 10"2, while the MEI risk from the fish ingestion pathway is
4 x 10"3.
*
For the dermal contact, direct ingestion, vapor inhalation, paniculate inhalation, and produce
ingestion pathways, TCDD dominates the MEI risk. For these pathways, the MEI is assumed to
reside at an agricultural site. The sludge applied at the two agricultural sites assessed in this analysis
has lower TCDF concentrations than TCDD concentrations. In Mississippi, a value of zero is
reported for the concentration of TCDF in its sludge, while in Pennsylvania, the reported
concentration of TCDF is 3.4 times lower than the reported TCDD concentrations. As a result, all
of the risks for pathways that assume that the MEI lives on an agricultural site are dominated by
TCDD.
The fish ingestion pathway MEI risk is also dominated by TCDD. This result is due to the
higher sediment partitioning coefficient assumed for TCDD. In contrast, the surface water and
groundwater ingestion pathway MEI risks are dominated by TCDF. This is due to the assumption
that the solubility of TCDF is higher than the solubility of TCDD, resulting in higher dissolved
concentrations of TCDF in water.
251
-------
REFERENCES FOR SECTION 2.4
Bovard, K.P., J.P. Fontenot, and B.M. Priode (1971). Accumulation and dissipation of heptachlor
residues in fattening steers. J. Anim. Sci. 33:127-132.
Consumer Product Safety Commission (1989). "Common Assumptions for the Assessment of
Human Dermal Exposure to Polychlorinated Di-benzo-p-dioxins and Dibenzofurans,"
memorandum dated July 6 from M. Babich.
Delta Western Feed Mills, Mississippi (1989). Personal Communication. July.
Food and Drug Administration (1989). "Bioavailability of Ingested 2,3,7,8-TCDD and Related
Substances," draft memo dated June 22 from Ivan Boyer.
Fries, George. (1982). Potential poiychlorinated biphenyl residues in animal products from
application of contaminated sewage sludge to land. Journal of Environmental Quality.,
Vol. 11, no. 1.
Geyer, H. J. Scheunert, I., Filser, J. G. and F. Korte (1986). Bioconcentration potential (BCP) of
2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) in terrestrial organisms including
humans. Chemosphere 15:(9-12) 1495-1502.
Hawley, J.K.. (1985). Assessment of health risk from exposure to contaminated soil. Risk Analysis
5(4):289-302.
Keenan, R.E., Sauer, M., Lawrence, F., Rand, E., and D. Crawford (1989). "Examination of
potential risks from exposure to dioxin in sludge used to reclaim abandoned strip mines."
In: The Risk Assessment of Environmental and Human Health Hazards: A Textbook of
Case Studies. D.J. Paustenbach, ed. J. Wiley and Sons, New York, pp. 935-998.
Kimbrough, R., Falk, H. and P. Stehr (1984). Health implications of 2,3,7,8-TCDD
contamination of residential soil. J. Tox. Envir. Health 14:47-93.
Martin, S.G., Thiel, D.A., Duncan, J.W., and W.R. Lance (1987). "Effects of a paper industry
sludge containing dioxin on wildlife in red pine plantations." In: Proceedings of 1987
TAPPI Environmental Conference.
National Council of the Paper Industry for Air and Stream Improvement (NCASI) (1984). The
Land Application and Related Utilization of Pulp and Paper Mill Sludges. Technical
Bulletin Number 439, August.
National Council of the Paper Industry for Air and Stream Improvement, Inc (NCASI) (1987).
Assessment of Human health Risks Related to Exposure to Dioxin from Land Application
of Wastewater Sludge in Maine. June.
National Research Council Canada (1981). Polvchlorinated Dibenzo-p-Dioxins: Criteria for Their
Effects on the Environment. NRCC Document Number 18574.
Pocchiari, F., F. Cattabein, G.D. Porta, U. Fortuniat, V. Silarno, and G. Zapponi (1986).
"Assessment of exposure to 2,3,7,8-TCDD in the Seveso area." Chemosphere. 15(9-12):
1851-1865. Cited in Wisconsin Department of Health and Social Services (1989).
"Human Exposure Assessment for Dioxin and Furan Contaminated Papermill Sludge
Applied to Soils". Final Draft.
252
-------
Rumsey, T.S., and J. Bond. (1974). The effect of urea, diethylstilbestrol and type of diet on
distribution of aldrin and dieldrin residues in finished beef heifers. J. Agric. Food
Chem. 22:664-667.
Sacchi, G.A., P. Vigano, G. Fortunati, and S.M. Cocucci. (1986). "Accumulation of 2,3,7,8-
Tetrachlorodibenzo-p-dioxin from soil and nutrient solution by bean and maize plants".
Experientia 42:586-588.
Schaum, J. (1984). Risk analysis of TCDD contaminated soils. U.S. EPA, Office of Health and
Environmental Assessment, EPA 600//84-4/031, November.
Science and Education Administration and the U.S. Department of Agriculture in Cooperation
with Perdue Agricultural Experimental Station (1978). "Predicting Rainfall Erosion Losses:
A Guide to Conservation Planning." December.
U.S. Department of Agriculture (1978). "National Food Consumption Survey 1977-1978." Report
Number H-6. Human Nutrition Information Service. NFCS-1977-78.
U.S. Department of Agriculture (1982). Food and Nutrition Service, "Commodity Maps".
Prepared by Schnittker Associates. Washington, D.C. December.
U.S. Department of Agriculture (1985). Agricultural Statistics 1985. United States Government
Printing Office. Washington, D.C.
U.S. Department of Commerce (1987). Bureau of the Census. Statistical Abstract of the United
States.
U.S. EPA, Office of Air Quality, Planning and Standards (1984). National Air Quality and
Emissions Trends Report. 1982. EPA-450/4-84-002, March.
U.S. EPA (1985a). "TCDD Transport from Contaminated Sites to Exposed URE Loactions: A
Methodology for Calculating Conversion Factors." Final Report. G.W. Dawson, et al.,
Batelle Project Management Division, Richland, WA, June.
U.S. EPA (1985b). Development of Statistical Distributions or Ranges of Standard Factors Used in
Exposure Assessments. EPA/600/8-85/010, August. Prepared by GCA Corp., Chapel
Hill, N.C.
U.S. EPA (1987a). "Comparison of Food Consumption Data". Tolerance Assessment Program,
Office of Pesticides and Toxic Substances. Washington, D.C.
U.S. EPA (1987b). Office of Health and Environmental Assessment, Evaluation and Criteria
Assessment Office. Development of Risk Assessment Methodology for the Land
Application and Distribution and Marketing of Municipal Sludge. August, final draft
version.
U.S. EPA (1987c). The National TCDD Study: Tiers 3, 5, 6, and 7. Office of Water Regulations
and Standards. EPA 440/4-87-003.
U.S. EPA (1988a). Office of Health and Environmental Assessment, Exposure Assessment Group.
Estimating Exposure to 2.3.7.8-TCDD. Draft Report March.
U.S. EPA (1988b). Office of Water Regulations and Standards, Technical Support Document for
the Land Application and Distribution and Marketing of Sewage Sludge. Draft Report,
August.
253
-------
U.S. EPA (1988c). Risk Assessment for Dioxin Contamination, Midland, Michigan. EPA-905/4-
88-005. April.
U.S. EPA (1989a). "Interim Final Guidance for Soil Ingestion Rates." Office of Solid Waste
Emergency Response Directive Number 9850.4, January 27 from J. Winston Porter.
U.S. EPA, Office of Water Regulations and Standards (1989b). Human Health Risk Assessment
for Municipal Sludge Disposal: Benefits of Alternative Regulatory Options. February.
U.S. EPA (1989c). Office of Toxic Substances. Graphic Exposure Modelling System: User Guide.
March.
U.S. EPA (1989d). Health and Environmental Review Division memorandum to Greg Schweer,
Office of Toxic Substances, U.S. EPA, on dioxins in paper products: bioavailability by
inhalation. Memorandum dated June 16 from F.J. DiCarlo.
U.S. EPA (1989e). Memorandum to Dioxin-in-Paper Workgroup, on the bioavailability of dioxins
in paper products, dated June 23 from C. Cinalli and Conrad Flessner.
U.S. EPA (1989f). 104-Mill Data Base. Office of Water Regulations and Standards, July 17
version.
U.S. EPA (1989g). Memorandum to Dioxin-in-Paper Workgroup, dated July 21 from C. Cinalli.
U. S. EPA (1989h). "Memorandum: OTS/EEB Aquatic Life Hazard Assessment (Including BCF
Values) for 'Dioxin in Paper'." Office of Pesticides and Toxic Substances. Washington,
D.C. Memorandum dated August 8.
United States Geological Survey (1985). "National Water Summary - 1985". Washington, D.C.
Wipf, H.K., E. Homberger, N. Neuner, U.B. Ranalder, W. Vetter, and J.P. Vuilleumier (1982).
TCDD levels in soil and plant samples from the Seveso area." In: Huntiziger, O., R.W.
Frei, E. Merian, and F. Pocchiari, editors. Chlorinated Dioxins and Related Compounds:
Impact on the Environment. Pergamon Press, New York.
Young, A.L. (1983). "Long-term studies on the persistence and movement of TCDD in a natural
ecosystem." In: Tucker, R.E. A.L. Young, and A.G. Gray, editors. Human and
Environmental Risks of Chlorinated Dioxins and Related Compounds. Plenum Press, New
York.
254
-------
2.5 Exposure and Risks from Distribution and Marketing of Pulp and Paper Sludge
Sludge that is composted and marketed can be used as a soil amendment in residential settings
as well as for agricultural and commercial purposes. According to the 104-Mill Study, seven mills
in five states distribute and market at least a portion of their sludge. Based on data from the 104-
Mill Study, the total volume of sludge distributed and marketed by these plants is estimated to be
208,000 dry metric tons per year. In some cases, the plants in this study reported two methods of
sludge disposal, but did not provide a break-down of the quantities of sludge disposed by each
method. In these cases, it is assumed the entire quantity of sludge by the plant produced is
distributed and marketed. To the extent that this assumption overestimates the quantity of sludge
distributed and marketed every year, the population risk estimates derived from this analysis will
overestimate the true population risk.
This analysis estimates risks to members of households using composted sludge from the
following routes of exposure:
Home gardeners incorporate distributed and marketed sludge to home gardens. The home-
grown crops incorporate small amounts of contaminant into their tissues. Household residents
then consume the home-grown crops.
*
Home gardeners incorporate distributed and marketed sludge into their home gardens, or
use it for other home gardening purposes, such as lawns or flower beds. Children and adults
in the gardening household come into direct dermal contact with the sludge. TCDD and
TCDF from the sludge is absorbed through the skin.
Children ingest small amounts of the sludge/soil mixture through normal mouthing
behavior. Adults also inadvertently ingest small quantities of sludge/soil.
TCDD and TCDF in distributed and marketed sludge volatilizes from the sludge into the
air. Residents of the household inhale the volatilized TCDD and TCDF.
Distributed and marketed sludge is applied to home gardens or other home uses. Particles
of the sludge/soil mixture become suspended in the air. Members of the household inhale
the contaminated particles.
255
-------
Since the actual users of distributed and marketed sludge are not known, a generic scenario
is used to estimate risks from the distribution and marketing of sludge. In this scenario, a household
uses composted sludge as a soil amendment for ornamental or vegetable gardening. "Low risk,"
"best" and "high risk" estimates are derived for typical exposure, while "best" and "high risk" estimates
are derived for MEI exposure. The following discussion briefly describes the scenario considered
to obtain these estimates, as well as the data sources used to construct the generic scenario. The
estimated soil concentrations used in the "low risk," "best estimate," and "high risk" scenarios for
typical exposures are shown in Table 2.5.A. The highest soil concentration presented in this Table
is used in the MEI analysis. Other parameters that describe the generic scenario, as well as the
physical chemical parameters of TCDD and TCDF, used in the typical exposure scenario are
presented in Table 2.5.B. Parameter values used in the MEI exposure assessment are shown in Table
2.5.C.
Description of Distribution and Marketing Scenario
According the National Gardening Survey (1987), 34 million of the 69 million households
involved in gardening activity in 1986 grew vegetables. Based on these data, the best estimate of
typical exposure assumes that approximately one-half of the distributed and marketed sludge goes
to households with vegetable gardens, while the other half goes to households that use the sludge for
ornamental gardening. Furthermore, best estimate typical analysis assumes that households using
sludge for ornamental gardening apply sludge at the same rate as households with vegetable gardens.
For the "high risk" typical exposure scenario, it is assumed that all of the distributed and marketed
sludge goes to households with vegetable gardens. This assumption is also made for the MEI analysis.
The concentrations of TCDD and TCDF in sludge from plants that distribute and market
sludge were obtained from the 104-Mill Study. The methods for calculating the soil concentration
of TCDD and TCDF in home garden soils amended with the composted sludge are described in
Appendix A. The soil concentration model requires inputs for the initial TCDD and TCDF
concentrations, length of the application period and the depth of incorporation with background soils.
Decay of TCDD and TCDF during the composting process is assumed to be negligible. The analysis
assumes that composted sludge is applied to a home garden for 20 years. The home gardener
continues to use the garden for an additional fifty years.
In the "low risk" and "best estimate" typical exposure scenarios, the analysis assumes that
sludge is soil incorporated. For the "high risk" typical exposure scenario, it is assumed that sludge
is applied only to the top layer of soil. USDA (1979) recommends depths of incorporation for the
maintenance of crops ranging from approximately 6 to 10 inches. Sludge may also be applied without
256
-------
m
0
10
L.
4-
C
1)
U
c
IO
o
01
ซ
^
l_
O
O)
c:
e
3
in
in
o
in
4}
r-
U
C
vO
O>
c
3
m
a
in
4)
ฃ
u
c
Mป
0
O
c
1
3
m
ซ
a
c
o
4-
ID
O
a.
i_
0
u
c
_
..
O
in
c
O
4-
IO
i_
O
a.
b
o
c
.*
4.
a.
4)
o
O
4-
ซJ
O
a.
i_
O
u
c
_
^
O
.c
4-
Q.
o
o
IO
L.
4-
C
U
c
0
u
4)
0)
o
3
C
ซJ h-
I ^
D CL
LL.
c
x h
O Q-
a.
a
c
a t
i_ a.
3 a.
a.
c
5 1
a
c
IO ป-
i n
I tL
3 a.
u.
c
O t
w ^^
c
IO (
L. a.
3 a.
LL.
c
X (
o a.
a.
a
ro "^ r** ^ ^h v i^
V ^ ^
tN T (N 00
in TT
p^ CM O* r~ป ^r r^ ao
O T CM
O CM 03 in
O\ in in
ซ3 O ""ป CM o
CO K1 _ ป - fS,
*r\ O O O O
CM
O O ป >O CM
O f O T O
CM
in v o> in
O O 10 KI KI o
O O O O O O
*
*
^ ^ ^
CM f CM OO
in ซr
* *
* ซ
r^ CM Oป r* ^ rป ao
<3 *r "~ "~ "" CM
in
4)
10 c
4- a
10
c
b
ซ_
^
IO
U
c
(O
m
c
c
a!
o
0)
ซ
in
10
2
O
ฃ
0
c
10
in
c
c
4)
CL
in
.
o
o
c
ID
TJ
4)
Q.
a
O)
o
O)
c
i i
3 Q
in
in o>
10 c
- o
a> ~
i. +-
3 O
O 10
in u
a. Q.
x
4) in
4-
4- C.
O a
a i_
4) O
O c
P> O
1. 4-
4) 10
> l_
O 4-
c
in 4)
c o
2 ง
4- O
10
L, 4)
4- O)
C IO
4) L.
U 4>
C >
O 10
IO 3
257
-------
O)
c
4-
4)
jฃ
U
i
c
4-
E
4-
m 4-
41
5^
a>
>
t_
3
I/)
C
4>
o
L.
ID
CD
ID
c
0
4-
ID
z
^
o
Jf
tf)
ID
to
c
o
a
4)
V)
ID
J3
B>
X
O
o
c
ID
4)
4-
a
E
4-
V)
4)
in
CD
^^
8
*ซ
o
in
O
in
4)
O)
o
3
in
O
4-
C
4)
U
L.
B
fซk
oo
o^
^
0)
t.
3
I/?
c
4)
o
L.
ID
U
ID
C
0
4-
ID
Z
c
4)
o
L.
ID
4>
s
ฃ
O
B
4)
JD
in
O
a.
4)
3
ซ.
ID
^
E
E
X
a
E
O)
X
m
4)
m
^
41
C.
O
in
0)
O)
a
^
to
0)
ai
(0
X
(U
0
41
m
p
S
a
in
4)
u
c
o
in
4>
u
c
10
m
4)
.c
u
c
0
'o
to
4-
C
CD
r
C7*
^
4-
ul
3
O)
3
ป
1
LLJ
Z
i
2
^
GO
CTi
^
.^
^
t
S
*^
258
-------
O)
^_
~J
^"
t)
^
o
ซ
_
o
4-
3
J3
U
m
Q
"ซ
o
ฃ
Q
s^
^~
o
4)
5
"In
>L
^^
^z
c
u
3
U)
R
X
LLJ
U
1i
.^
.
^
in
3
Q.
C
o
*
^
IO
in
L.
4)
4-
i
. a
^^
^^
*
A
8
CO
ID
CM
4)
^
IO
^
4)
U
C
4>
4)
1
O
4-
10
C
ID
0.
X
4>
in
4)
1
j=
O)
I
4)
ID
E
in 4-
4> in
CD 4>
X
o
L
4)
4)
4- E
3 IO
Q. I.
C 10
Q.
^-^
CM
CO
ซ r*.
^ co '
r- co
co o>
o^ ^ *^
IO
^^
SO. 4>
UJ
I C
IO
3 E
T> >
3
TJ
IO TJ
X 0
ฑ i
4)
TJ
4> TJ
4- 3
ID IO
E -J
4- T3
m c
LU ID
<0
\
o
X
in
O m
O in
>o
i
O
X
in
O m
O in
*o
1
O
X
in
O in
0* in
-,
>* ^^ >^
4- O 4-
> in >
.^ ^^ ซv
inCM in
3 E 3
>*. o **- *^
- " *- 1. O
_ ~ 4) 4)
TJ I. TJ 4- in
10 X.
O C U C U
l~ ^. t . ^^
^^
vo
co
._
> '
._
^
^
n
3
in
i
O
CM
4-
a
c
4-
E
__
4-
3
lm
O
u
4)
U
U
4>
4-
10
X
c
*-
1 at ions
3
E
in
4)
o
01
O)
^^
C
o
c
V
4-
^
ซ
**
in
co
~* P
oo
z ^
o
T3
0)
i 4)
in 4-
-4-
O
o
4- 4)
O
C. -
Q.
ง^ C
10 E
1- J=
j=
- O)
<
Q. J=
UJ
in -o
ID c
1/5 IO
*-^
in o 4)
CO 4-
O\ C IO
E
TJ 4-
c 4) in
O 4- -^ 4)
in i_ rป
^ O oo 4-
o Q. o> in
10 4) CO
-5 l_ -^ CD
in
1
O
""0 "~
X
CM X \O
CM
in
i
O
ro "~
X
CM X >O
CM
VO
1
O
O X
CM
CM X
CM
p"l .
4) 1.
O O ~ 4)
E ฃ * 0, X
1 X t- < t
U
in
4j
3
IO
>
4>
4-
IO
E
4-
m
4)
4-
C
a
4-
in
8*
X
_J
4)
IT)
4-
m
4)
X
o
ซ
^^
VO
CO
*
__
O
o
-
__
O
E
^
E
E
IO
.
^^
p^.
co
**>
<
a.
UJ
"\
3
g
1
259
-------
CD
c
4-
4)
I.
ft]
c
e
j but ion i
i.
4-
5
4)
o
ฃ
o
4*
TS
4)
in
in
in
^
Q
^
^^
L.
in
O
Q.
X
UJ
IB
U
a
H
.
O
in
4-
Q.
C
"ffl
O
^
a
in
4)
41
1
u
a
a.
~
+-
0
o
m
in
o
10
"
4)
U
C
4)
^
10 T3
X O
.C 4-
4- 4)
e
x
41
^y *
0) T3
f- 3
E i
* T3
"/) C
UJ T3
*o
0
X
to
in
o
o
x
to
in
to
o
x
tO
in
c
._
>*
4-
^^
> 0
_ 4)
m in
3 t- X
S"" ^ *%
c
^
~-
ฃ
^H
>.
^
>*
^
^B
JO
3
^B
O
in
E
u
IO
4)
L,
4)
4-
IO
X
c
in
c
O
4-
a
3
E
in
__
o
o
O
E
*^
O)
X
en
3
^*
C
O
in
c
4-
<
ป
*ป
in
^3
Ot
^^ ^**^
r*.
ao
ON
Q
2
(_
o
ฃ5 *
O "O 4)
t OJ +-
in 3 E
a in
Ifl 4-
4> a in
E 4)
a u)
in j*
in in
o >.
E O 1
3 4- J=
in o o>
in j=
< ex j=
4-
C
C
O
* ^
O -^
O c
4- 4)
0 * 5
a
L.
ง O
a.
10 10
^ JC, >
* to
*-* 'S.
in o uj
LLJ C. O
E J^ JT IO
4) U O) JO C
JZ (O
O ซ_ J= C
in (o 4) 4)
O
10 3 - 4- 4-
> a 10 -o 10 T> m
4) > O E O LU
.c u j: ~ J= E
4- O X 4- 4- 4- 4)
O L. o 4) in 4> x:
m a. _j E 4) E U
'
in
0 0
X X
to in to
10 to ao
in
's 'ฐ
x x
to m \o
ro 10 ao
in
*0 0
x x
10 to
fO ซ ao
e
ID
in 4-
3 C
U >. ID
41 L. 4-
u c ซ
E +- ^ T C\
t- ฃ5
L^ O) LL. LL.
Q Q Q X
O 4) O O IO
1- X 1- 1- _I
_
^^
E
E
-------
O)
g
^_
0
1
^J
c
ID
C
o
4-
L.
in
O
in
"in
15
c
<
$
o
ซ
o
1
o
^
0
in
in
3
n
C
"
i
i
o
c
ID
in
0
i
i
i_
o
in
0
a
Ifl <
CD
Ol
^"
iซr
<
Q
to
Z)
o
CM
o
0
ID
a
C ID
O 0
^ "S^
ID 1
O Z
'Q.
a.
261
-------
c
4-
ซ
jฃ
i
c
15
OL
O
4-
ID
C
ID
^_
Q.
X
4)
X
in
0)
4-
*
C
<3>
X
4>
ID
E
4-
in 4-
4> in
m ซ
^
4-
4- E
Q. 1_
c a
a
ซป **. ^-
-.00
i co
ง 2 _J
^ 10
** < 4-
4) OB 4)
4> UJ
> C
3V) E
13 >.
LU =3 -I
V
3
o
a TJ
X O
j: 4-
4~ Q)
w E
f
4)
-0 _
4> T3
4- 3
ID ID
E -1
4- T3
m c
Ul ID
ซo
1
0
""
X
in
O m
O in
m >
incvi in
3 E 3
ป. u * "-
<*. *~ *- I. O
_ 4) 4)
o L. a 4- in
ID X
O O ฃ
O c O c o
l_ _ (- ~
^v
co
0>
*r
ID
4>
4>
a.
ID
4-
o
c
o
o
m
c
O
4-
10
L.
C
4)
U
C
O
u
8
o
fs^
o
o
*.
4-
.^
_
.
o
3
"5
in
t
,
o
fM
4-
ID
C
4-
,^
e
*
^
*-
^_
HK
J3
3
^
O
in
r*
O
ID
0)
L.
U
4)
4-
10
I
C
in
c
0
4-
IO
3
E
m
_
.
CO
0>
ii ^
^
0
"~
o
3 0)
m 4-
m ID
a E
m 4-
~ m
m 4>
^
ซ. ^
O in
+ .^
O u
a. c.
a>
i Z
i
o
CM
9"
O
a.
U> 10
>
in
^ป *^
O
O
4- 4)
O -^
*,
ง^ C
^ _~
10 E
|_ jC ^
f>
co
< Oi
Q.
UJ ~~*-
in
ID
^ O CO 3
u ex o
ID 4) * ID
in
i
^ 0
^ ""
- X
CM X >O
10 ~ ~
in
i
O
O ~~
X
-
4) V.
O C
O O -~ 4)
E 4- X 0) X
งO) O fO ^3
_ Q ฃ Q
Q) O O O
t * H ^ 1
X E
262
-------
O)
e
*~
0
^
L.
z
o
C
IO
0
L.
4-
m
Q
t/>
MB
i
^
to
e
^
ซB
U
^
^
ฃ
O
2
1
in
+
3
O.
C
"0
*
c
IO
in
i_
0
+ป
i
L.
ID
0,
+-
o
u
0
in
X
o
"jj
E
in +-
o in
CO O
0
ซ- 1
3 a
Q. V.
c a
a.
** *-*
CM fsi
ao ao
^
ซMT -*-
_ป _t
10 10
+- +-
0 0
c c
10 10
E E
_J _l
y
0 3
Jฃ -0
ID X>
x x o
J= "O C +-
ป- O +-0
^ J3 *^ ^
X +- X
0 0
o E o
0 0 T3
+-0 +-3
10 0 to ia
E _i E -i
+- T3 *- T3
me in c
LU IO LU IO
0
X
in
O ซ
O in
0
i
O
"~
X
in
O >0
O m
c c
ซ r*, ซ
u
X. 0 >-
4- in +-
><\ > u
~ E 0
in u in m
3 ~- 3 t_ X
u. **- u. **- 0c\
Q * u. Q t- +- E
a ~ o .
._
_
.^
jQ
O 0
in +- ~
10 en
U. X X
8 c ?
I ซ %^
c
o
U)
c
*-
^
.ป
^*1
in
00
ซ
>^ *~^
E 1
z ^
o
~z
L.
o
^5 *
^5 ^3 4)
h- 0 +-
E 10
in 3 E
a in
in +-
ง10 in
0
m
m ^
m in
o >.
01.
E O 1
3 +- J=
in o O)
m ฃ
< 0. J=
0
c
ซ_
ป. 10
= %
o
CM
l*> (O
%
in
^
* ^
O ~ 3
O c u
+- 0 0
0 - E -
ฃ O
a. E +-
Lu "^ O Lu O
Q a. Q
O ID ID O0
1 C > t X
a
2
LU
o
IO
10
+- c
in in 0 0
LU E ^ L.
E 0 3
0 O E T?
O c C. u
*- O
O 0 in a.
L. L. 0
<- 3 +- +-
T3 10 13 in
0 0 E O LU
3 0 J= E
O + * 0
10 L. in 0 r:
> a. LU 6 O
in
*2 2
X X
in to
M OO
in
^ i
0 0
"~ "~
X X
in vo
ro oo
in +-
- c
>. 10
u +-
u cm
O 0 c
^ I O'
S14" R J
Cj *D
t- ( -1
_
|
o
E
E
IO
263
-------
incorporation (i.e., top dressing). In the "best estimate" scenario, this analysis used 6 inches of
incorporation. Assuming 10 inches of incorporation yields a lower exposure estimate due to greater
dilution of the sludge contaminants, while assuming the top-dressing scenario yields a higher
exposure estimate. The top-dressing scenario is used to model risks to ihe MEI.
Data Sources and Model Inputs for Application Rates
The best estimate for typical exposure assumes that the home gardener applies 10 dry metric
tons of sludge per hectare per year to his or her home garden. The recommended application rates
for home uses ranges from 5 to 20 dry metric ton per hectare (USDA, 1979). However, assuming a
higher (or lower) application rate forces a smaller (or larger) estimate of the number of households
affected, as long as there is a constant amount of sludge dedicated to home uses. Population cancer
risks are relatively insensitive to this parameter as long as cancer risks are assumed to have no
threshold. MEI risks are affected by this assumption, however. The MEI was assumed to apply 20
metric tons per year to his/her home garden.
The size of the garden assumed in the generic scenario affects the individual risks estimates
and influences the estimate of the size of the exposed population. According to the National Garden
Survey (1987) the average garden size for combined rural and urban vegetable gardens is 0.016
hectares. For the "high risk" scenario, the average rural garden size (0.022 hectares), is assumed. The
MEI is assumed to have a garden 0.022 hectares in size.
Using the generic scenario described above, this analysis estimates exposures from dermal
contact, vapor and particulate inhalation, and direct ingestion for young children (ages 1-6), older
children (ages 7-12) and for adults, while dietary exposures are estimated for young children (1-6)
and adults. The average daily exposure over a lifetime from each of these pathways is the weighted
average of the daily exposures during these stages of life. These exposures are combined with cancer
slope factors to obtain incremental lifetime risk from TCDD and TCDF exposure. The following
sections describe the methods and data used to estimate risks from home uses of distributed and
marketed sludge through each these pathways. Results are summarized in the final section.
2.5.1 Estimates of Exposure and Risks from Dermal Contact with Skin
Humans coming in direct contact with sludge contaminated soils may absorb TCDD and
TCDF through their skin. The amount of TCDD and TCDF absorbed will depend on the area of skin
exposed and on the length of time that the contaminated soil is in contact with the skin. The
264
-------
following discussion summarizes the model used to estimate exposure through dermal contact with
composted pulp and paper mill sludges used in residential settings.
To estimate exposure through direct contact with soil containing TCDD and TCDF,
methodologies presented in Schaum (1984), Hawley (1985), and EPA (1988a) were used. The model
uses empirically-derived information on the amount of soil or dust that adheres to a square
centimeter of skin, the area of skin exposed in various settings and the absorption rate of TCDD or
TCDF through skin to derive the dose of TCDD or TCDF from dermal contact with contaminated
soil or dust.
Description of Calculations
The calculation of dermal exposure proceeds in two steps. First the average daily exposure
from dermal contact is calculated as the product of area of skin affected, the contact rate, the dermal
absorption rate and the duration of contact. Second, the risk from dermal contact is calculated using
the estimate of daily exposure and the slope factors of TCDD and TCDF.
Description of Exposure Calculations
The concentration of TCDD and TCDF in the soil in home gardens or lawns is derived as
described in Appendix A. Contaminated soil may also be brought indoors through airborne dust
particles, and/or tracking of dirt on clothes and shoes. The concentration of TCDD and TCDF in
indoor dust is calculated as follows:
C = C F
*-in ^o r
where:
Cjn = concentration of TCDD or TCDF in dust indoors, mg/mg
C0 = concentration of TCDD or TCDF in soil outdoors, mg/mg
F = ratio of indoor concentrations to outdoor concentrations
The indoor dust and outdoor soil contaminant concentrations are used to estimate human
exposure and risk from dermal contact with these media. Daily doses are estimated for three age
groups: young children (ages 1-6), older children (ages 7-1 1) and adults (ages 12 and older). The dose
for each age group is calculated as:
DOSEg = [g SAin
-------
where:
ABd = systemic absorption rate through the skin
BWg = body weight = 70 kg for adult, 16 kg for young child, 35 kg for older child
C0 = concentration of TCDD or TCDF in soil outdoors, mg/mg
Cin = concentration of TCDD or TCDF in dust indoors, mg/mg
CR,_ = contact rate of soil with skin for age group g, indoors, mg/cm2
in, g
CRQ g = contact rate of soil with skin for age group g, outdoors, mg/cm2
DOSE = dose from outdoor exposure for age group g, mg/kg/day
Hi = hours spent indoors for age group g
HQ = hours spent outdoors for age group g
SA. = surface area of skin exposed to soil for age group g, indoors, cm2
i n, 9
SAQ = surface area of skin exposed to soil for age group g, outdoors, cm2
M = matrix effect on absorption rate, percent
For each age group, the soil contaminant concentration (expressed in mg/mg for ease of
calculation) is multiplied by the soil contact rate outdoors (mg/cm2) and by the area of the skin
exposed during outdoor activity (cm2) to obtain the total quantity of soil-bound TCDD or TCDF
adhering to the skin (mg). The quantity of contaminant on the skin is then adjusted by two factors:
the fraction of the contaminant that migrates from the soil matrix and comes into contact with the
skin; and the fraction of TCDD or TCDF that absorbs through the skin. Since the dermal absorption
rate is expressed as the fraction of TCDD or TCDF that is absorbed through the skin per hour of
contact, it must be multiplied by the hours that the soil is assumed to be in contact with the skin.
The same calculations are also performed for exposures in indoor settings, using the corresponding
indoor values for the model input parameters. The total daily dermal absorption of TCDD or TCDF
is the sum of indoor absorption and outdoor absorption. Dividing the total dermal absorption for each
age group by the body weight for that age yields a daily dose of TCDD or TCDF through dermal
absorption in mg/kg/day.
To obtain the weighted average dose over the lifetime of an individual, the following
calculation is used:
where:
DOSEavg = E FRACg (DOSEg)
DOSE = weighted average daily dose for an individual, mg/kg/day
DOSEg = daily dose for individual in age group g
FRAC = fraction of lifetime spent in age group g
266
-------
Description of Cancer Risk Calculations
Once the daily dose estimate is obtained, it is combined with information about the slope
factor of TCDD and TCDF to obtain an estimate of lifetime risk from dermal exposure to these
contaminants. The calculation of individual risk is:
1C = DOSEavg Q;
where:
DOSE = weighted average daily dose for an individual, mg/kg/day
1C = individual cancer risk over lifetime from DOSEayg of TCDD or TCDF
q^ = incremental lifetime risk per mg/kg/day dose of TCDD or TCDF
This calculation is performed for both a typical and the most exposed individual. Individual cancer
risk for a typical exposed individual is converted to annual total population risk (in cases per year)
by multiplying the number of persons exposed by the individual risk and dividing by the average
person's lifespan, as described in the following equation:
PC = 1C POP / LS
" where:
LS = average lifespan of an individual =ซ 70 years
PC = population risk, cancer cases per year
POP = population exposed to DOSEavg
Data Sources and Model Inputs
The values used for each model input for "low risk," "best" and "high risk" typical exposure
estimates are summarized in Table 2.5.D. The values used to derive the MEI "best" and "high risk"
exposures are found in Table 2.5.E. The best MEI exposure estimate is derived using by combining
estimates of behavioral input parameters with the best estimates of physical/chemical properties of
TCDD and TCDF. The "high risk" estimate of MEI exposure uses the same behavioral inputs, but
combines them with the high estimates of physical and chemical parameters of TCDD and TCDF.
The following sections describe each input and documents the data sources used to derive the
values for the parameters for both the typical and MEI analyses. Where parameter input values differ
267
-------
ID
X
.C
4-
IO
Q.
ID
E
L.
4>
4>
tr
c
O
V
IO
c
IO
a.
x
X
in
ซ
I
^:
O)
X
41
4-
43
E
4-
in 4-
o m
O 4)
B
0
^
a>
4-
-ฃ>> > > >.
3 3 41 4) 4) 4) 4>
a ID
.CX.CXX X X X
U IO x ui x x x x x
4)
C IO >.
>-4) in 4) Dl Ol T3C T3C O-t-
Jf~* *^ c cr "^ 4) x ^ *^
in D> ' c oo c in cm a. e -t-
X EE 4)Oi O inx O mx in*uo4-
^ *O* * O>ป * O> OI4-IO
oo >ป oo ฃ ^ ^ 4> in 4) in c4-Ec4-ซ
01 in 41 010 4) j< in o >> L. O > L. 4- Oc^
r~- 4) i/> 4) i_ j: vixajo X4>O > . in ox o g X x T3 gxx-a c 4- x
E ID x E n H3 oo x <- ce. o ซJ4- 5 104- OTJ
3 > 3 > Ol T3 O U>>X3 l_ >- X 3 C ป- OJ OJ O
ax c ID-O E i- ซ- c *- & O H-O O- ^ O i- 4-
UQ.4- U4-O* U>>4) 4)*OO%^Jฃ 4)^0^^^ inO4)~OE4-
* * 3"O l.4)L.O4) U4)C-O4) 4-X3C ID
_l E 4) U 4) Ol 4) O4-<^CN4) lD4-ซ-(N4) 4-
c 4- c - 4>cioin x x 3i_io m-^
.. +- ^4>X 3 > IO in U 4) II l/> U 4) II OQ>;_4)O
0) i/l U in 3 in O A 4) 3 O 4) 3 O IOซ- O
T334) -or- x IOTJX 3 m in x 3 in in x a ป- >. i.
4) 0) Ol ID > O' a ID l_ 4- IOIO1.4- U-- 4)3
inio4-o in > c m in 10 >O o >O o c TJ ซ o
3>inoO 3 -* "4- 00 >*- O>- > > O>- -ซ-4>4>4-X.C
4) O 4) 4) C Ol IO4>'OI. >O 4) T3 L. (.O.UIO
ฃ ^3 X4-TJ 4-DO ฃr*4-Cซ ฃrป4-C4> 4JTJOIOX
งO> O O ซ3 4) ID-O > Ol Ol a > Ol Ol 10 > 4-' 4-
uin OOCLE4- B ซJ 'O E 4) E 41 ID 4) c < i.
OlฃO Ol 4> 4)>ซ ฃ ^^ .C-* Uฃ>O(-O
ซ- J| _J 4- U 4- C U) 4) 4- Ol 4- Ol O 4) O ซ-
O E a ป 4> 4> AX CU4>ซ-~ CU4><^~ U l_ 4- Ol
in 3 mo o 'o ID ซJ o c ao c IOO*U>>E
O ID E 34-OO 4- OJ X 4-4-4-a 4-4-4-ID -t- i- in ฃO S m oi mmx xumino xt-inmo c ooumo
4IUI. 5<04> 4>aO>> O ซ3 W 3 QIOV3 QmOl 3
T3 (/> ป- Q: > J3 CD A J A _l X CD T? U -J X CD T3 U O ID TJ T>
in in in o O O
KI O O O
lO iO IO ^
in in in o ฃ
in in in o O oca x
O O in O O O ~- E
^ ^ ^
12 2 2 3 3 o
* *
o o o o o o
t
41
u
ID
- - a.
i_ i_ ป * in
o O O i- u i-
O O O O OOI4-
m 0) "O 0) -o 4)O 4) O 4)OC
4-JZ +-+.-O 4-+- 4--0 4-TJ-a 4-T3 3
IOU IO3 IO3 IOC IOC IOC>'O
|_ t-O l-O 1- 1_ U ID
ฃ JC
4-l_ 4--O 4- 4-ซ 4- O 4-ซป
O! U
IOEO ID ฃ I- IDE'*- IO ฃ "O IO E L. IDE4-OU
4-OT3 4-04) VO- 4-O VO4) 4- O O
CX4- CX-0 CX3 CX CXT3 CX3T34-
5OI3 OO) OOi"O OOIฃ OOI QO)"OC4-
EO OEO O S m OEO OEO OEO ID
268
-------
ID
X
4-
10
0.
_
IQ
&
O>
c
4-
4J
a
o
c
a
c
o
^_
3
^
[[_
Q
o
3
o
_
C
a.
1-
a
ซ*
in
o
3
U
4>
4-
i
i_
a
a.
o
c
a
in
O
4-
in
in
1
**
Q
m
*N
.g
^^
^^
o
^
4>
L.
ซ
ID
a:
c
O
4-
a
c
a
Q.
X
in
4)
I
O)
X
4)
a
E
4-
in 4-
4> in
CO 4>
g
J
L.
4-
4- 1
3 IO
a. i.
c a
a.
in
8
^
^
4>
X
10
X
in
00
0
ง
O
in
r-
O
c
ID
C
1 ~
4- c o c
c O in O
o ซ ป
U 4- O 4-
IO 4- ID
4- i_ i. in
O 4- in 4- t_
c i_ c o
O 0) 0 4) O
O O 0 TJ
4- C -0 C 4-
a o c o 3
(T 0 0 O
งin
00
1- O>
41 >
4- 4)
CL
O X
< X
-* >.
X ID
X 4) "O
in L. ^^
> O .
in x
in ** 4-
"10 C
o ^ o
4- X E
10 in **^
c *
4- O -
m (N ^
* ~ ? x
4- - .C tn 4-
5>. a a a
x ป -D 4-
in e
o .c tn r- o
C 4- U
0 B 0 ~ 0
-i - O o
"~-
ง
1- Oi
<* 00
Ot
T3
งin in a 9i E in
CO 00 4- OO O CO -
l-Ot O^Qld (_O\C
**-' O- *ป- (O
' c
4)>* >.C OX4)
4-4) q> .. 10 4-4)^
Q. 4- ฃ C O.
Ox inxo>4> Ox~o
4- 4-
in 10 in in m
> I- O ฃ 33
W ฃ 4- "O 'O
(N U-4- OO ซ L. 1-
E 4)4- 0 10 E .* 0 Qin
O >OIO4-OX OO3
ป- 4>a 4- cinmx -o -o_.
*-** 4-co in cca.m
in me o o V. ui >. _ ^
in >- u o o ot-ia ซ4-
uaoioo *- .c -o -4- x-mc
00^4- o tnx >ป >-^
4-inin3O in x ซ 10 10 o i-
3 ID O m ^ in in o) TJ OE4)
O 'X ซ- O>-4)Cin X X E
om o IDE > in in i. E
O>>4> 1- !n*O3.C!O U 1.4)3
"0 O .c t-j=cfNซio c
O>-4> COO- K1 IOCMX
l_ ^3 OO O O)--4> 4) (.
4) 4)* Einc4)i->-' ป->^inioO
4->4-4-in in (D 4- O 4- 4) -4-
$ 064- E 3 in E ^ป >- 4)' 3 O >
i I. 4- C 3 * -^ J3 (O 1/1 I- IO
4) c 4- g u>jฃinx4- m m TJ
VjOOlAE in X CO m4-f 'O4-IO4-X
C 6 O 4) IDX 04IO4- CO OM
ซJ 4> -!_ IOC ID IO ID l_
5d o>(- x'om^teoicg xcoc
ซ O O O 4)4>O O O O
-------
a
2
x:
4-
o
a.
^ป
Q
S
at
^j
0)
c
4-
Q
*
a
c
a
c
O
4-
3
^J
ซ
4-
0
ป
a
i
a
3
o
^
^3
c
^
-
a
u
a.
H
a
0
a
a
5ป
L.
4-
1
a
a.
c
a
0
0
1
*n
^^
4-
ง
q
m
CM
O
O
a
1
4)
U
C
4>
4)
i
c
o
4-
3
C
a
a.
x
in
V
o>
X
o
4-
IO
4-
in 4-
4> m
in 01 o>
4) 4) c
*: 3 ~
. ซtr
4- O > o 4) in
IO L. ^- O>
S * 10 X TT
4) in ^r 10 GO
4- C l_ CO OO XO>-
in o .
4) .-ซ o
4-4) t 4) ซ l_
^ Ol L. Oป O* 4) Ot 3 *^ 4) 3 E (Q 4)
*^ - *~ i"* ^_ (/) L. ^ (Q j* ^j
in 3 x .c 4- o
TO* "O* ป "Oซปin
ซ> 4) > > c > X co O 4) c
a. a. ^og"T34-ฃ4-
Ox ox x xxoioooxcino)
XllO *OIO (O OlO* O\ C. O 4) 4) O
. 01 0) >- - ^.
(U *O C. "^ 13 E -C (Q U) dl
t. ฃ.^-oa)O DCD'O E" "o tn >
83L.-ss.El/)aui X, L_>ป l_ 4) *
^ ^ L.IO
tot. 4- CM OซJซJ jC O> C IX-L.V)
o E 4->ป c x ซ^mtn mL.c e ซ- .cซ
>ปTO ~ ซoซi4- 4-mji 13 T3 3 10 o O mm
a c 4- >. o T3 +- in i_ m TJ 13 c - T3 c T3 +-
13 *^ m ID 4 **s ID 4) 4} ^ 4)4)IO*in 4)IO in 4 O) C
X 4> 13 c m c m>ฃXE m^Tjo) a) in c s L.
^ mi- x 4) c x 4-a, X4-O L.O IDซ
>ป 4)C4-CM >ซ Qซซ 4)tnO * 4)OCL~ (Oฃซ 4)^
^ n .0 inซ~ ID j^O i.oo)in ^xin ojtnin xu
a CM3 T3i_inmino)c'O in am c 3 oc.c cO4)L.io C4>cio* minio
E 4) L. ID 4 L. ^ C^ C O 4 ID ^ E 3 ^ *O E X fm ^ *^ 4 f" c
3>.ฃ OOOlin ฃ 313 -C D O IO C 4-4)
m XIDO 4) ฃ x i3m x: in*o^ 4-.CV)T34- >.U) 4-O. O. 4- 3Q4-
o 10 c c m 3O*>Om ป c in ฃi c 4) *"* 01 ^7 n ^j ID c jC
in > u 4- in 10 o in 4> O O4>4iciDin ID
4-4-4) 4) in c a c 4) -C "a 4- 4- a. m 13 4-cin m in in i.
m o E u 4- 4) o i_ 4- incax4>ซ m 10 ^ 4> i_ - m>4> 4>ฃO
203 OIAEE XJฃO~CM 4) ซ3 E 4) 6 in 4) E E3 4) O 6 E4-
4- in c o. 3 3 o in ป x e a. 3 o E E 3 13 E 3 3 in
em m 13 m L.
co o CT L. ซ) 4- m O > 4) m c 4> 4> a 4> m 4) >ซ- 10 4j in ซ 10 ID 10 o
a -a co cx:u m c 6 a - a m 10 4- in 4-
4- ฃ 4- ox: 4- 4- i_ -a 3 U4- x:4- 4-inx:o >-4-ox:uo)
Sin o c >i3Oixc4--^i: m x 4) i m o> o> a. x in o. o
3 O C O 4- 4) (N in O 4> J) C 4) O 4) 4) X O ฃ 4) X .CO
CM CM (N
E E E
o u o
SO CM O O O
r- pป o O T
10 m co CM ot
ป IO CM IO CM
CM CM CM
E E E
u u u
S 25 8 S3
*r m 10 rป
ป CM
(N CM CM
E E E
U O O
S S - 0 0 0
V CO O O
4> 4) 4>CM
in in in m E
I_ O O O 0
~ O a. o. a.
1. 5 X XX
.CT3 tt>t> ป<- 4)4-
^ c ซ 4)
_ 4) C C X) C3
o o x: __,-. -o
4- 4- ป ซJ ^ O Jฃ O OO OO O Q
8L. 4-CO 0-ป Q13 O
fli 4) ^ซ* lo 13 CM ID 13 ' ID 13
E-OT3 E3>V 4J4-E 4)4- 4)4-
_C 13 4- (-3U (-3.C L.3
I- O t- ID 0
-------
ID
X
ฃ
4-
Q_
ID
E
&
o
.5
4-
4>
l_
ID
z
TJ
C
ID
C
O
>
4-
3
^
4-
IA
O
1
^B
ID
3
TJ
2.
^3
C
^ป
ป
ex
>
H
o
o
4>
ID
>
^
4)
i
ID
L.
0.
C
a
in
O
4-
a,
3
in
in
^*
4-
0
U
Q
in
a
a
9
u
c
u
4)
1
c
O
4-
IO
C
ID
X
4>
in
I
0)
X
4>
4-
10
E
4-
4! in
CO 41
O
u
*
4)
4- E
3 ID
a. L.
C ID
a
in
2
4>
U
10
X .
13
C
in a
ฃ ซJ 4) 4>
L. L.
ฃ in a ID
4- 4)
O E 4) *
^ 3 o in o
IA ^O E
^ป in ^ L. -
ID
O 0 IA Lป
4- O w
U) (0 *^ *ป 41
1 * ฐ T, I
3 4- *- e in u
0 0 c TJ
L. C
<
in
CO
^
4)
X
ID
4)
L.
(0
in
TJ
C
ID
ฃ
>.
O
4)
E
in
ID
4>
ID
E
4-
in
0
4-
C
a
3
CM
E
u
in
CM
CO
CM
E
U
CM
E
U
I
TJ
4>
IA
O
a.
x
4)
C
^
IA
^>
ID
0
l_
<
TJ
k
.^
C
0
in
ซ
E
in
TJ in
$ "
O ฃ
a. 01
5 E
L.
4)
CM"
O E
U
l_
8
TJ
C ฃ
O
in
ao
o>
Ik
4)
X
10
TJ
4)
in i
O 4
A ^
X 0
in
in
c t-
ซJ ID
ฃ 4>
in
4) 4-
E
3 3
IA TJ
in ID
ID
4>
4>
4-
1 "x
^ M
m 4)
4) U
4- a.
in in
^ 01
e
TJ ~
C >
a
I =
CM
E ฃ
0 4-
8 5
(^ C
CM O>
E c
o
>
(N O>
E C
U
>
ป -
TJ
4) ~-
O E
a. u
x ^
4>
4-
C
3
Jt TJ
J1 ID
^
O u*
ID O
4) TJ
l_ C-
<
U 4- C
(O +- 3
+- ป- "^ JC
c to ^
ID L.
ttฃ 8 *
TJ 4-
.* .* C 1_
O I-
ซ O ฃ
c x o m
CO TJ
4) 4- TJ 4)
a. 4> >
O ซ m 4>
4- O 41
c a a.
ID ฃ X I/I
4) 1
ฃ L. 4-
4- O 4- l_
O
4> TJ m
> ID
4- o cn
L. in c
a 4>
in o 3 o
c in a)
TJ in x
ซ a
> in 0
0 ซ ฃ
0 O O)
IA in x x
TJ
C
ID
Ol
U C 4)
~M . (J
4- > ID
4- Q.
ID in
CM
E ~
^ O O
0
O O 4-
0 O -t-
a. r- n
in
CM
E *ป
. o o
0
O O 4-
IO O 4-
a. rป ID
Ifl ~
m
+.
*o
f
^
o
IA
4>
o
01
o
c
ป
IA
1
in
ซ
in
4-
C
ID
ex
.*
u
0
c
c
0
a.
O
4)
ฃ
4- IO
E
l- l-
ฐ ฃ
m c
C (O
O e
4- X
a.
E ซ-
3 O
in
in 4-
< e
O lA
E m
e 4)
p in
^,
Q. O
3
O C
Ol .*
* in
i_
0 -jj
>- 4-
4-
TJ
ID m
ID
ID
ID in
O CO
m 2
* >.
in 0
a
TJ X
ID
It
C
1 0
ฃ L.
TJ
.
L. TJ 3
O C O ~>
10 E
ฃ (U -
U in E CJ
O x "ซ O
0. O in
c *
o a 10 ฃ
4- 1 L. U
a. 3
0 1 *- J3
L. O O ID
3 N N ffl
in c c
O 4) 0 O Ol
X Ol
(/)
*-
3
^
(O
,^,
o
c
o
a.
8
IA
^
ID
4-
0
U
ซ
X
4-
IA
ID
C
0
TJ
ฃ
U
^*^
C
O
0 TJ
ฃ 4)
4- ID 4-
E <0
l_ l_ C
O 0
ซ- Q (-
o
lA C
C ID ฃ
O E U
i5 1
E * Q.
3 O
IA O
in 4- 4-
< c
c e i_
O
Tป
4?
c
1
u
ฃ
in
-
f*>
O
_
ป -^
O O
in
ฑ ง
L.
ID C X
XL.
ID O 4-
> ID
< TJ E
*,
E .
TJ *" ~3
c o -ป
ID E
in i o"
l=. 8
O in
O ID ฃ
1 L. U
a. 3
O o "ง
N N CQ
C C
0 0 0 Ol
- .- 01
271
-------
10
2
4-
1
ID
s.
c
IO
in
ง
Q.
in
in
LU
in
>I
'
4)
a
4)
U
C
4>
4>
^B
4)
or
c
o
4-
a
c
a
X
in
41
I
g.
X
4)
IO
E
in 4-
4> in
u
ซ
4>
4- E
3 10
a. u
C ID
Q.
v in ซr in in in in in
00 00 00 OO O3 OO OO 03
Ov O* O* O* 0^ O^ O* O^
E > E > >- >- > >
34)34)4) 4> 4) 4)
0 0
.C X .C X Z Z Z Z
O X X X X I X
0)
E -
T3 >
IOX ซ3 OO 0 O L.
~- ปป 4) t. X -UO
in O c 4- z
e e a 4- x> o xi
10 a 10 ซj >. _* o i- +-
C. ฃ. ^ I ซ| L. 104-
u u in c xiE4-
(/) OO XI 0> 4-Z3C ID
o c i ซj __._+.
c c 10 OX 3i-ioin*.
c >. o > 10 *- o
งO ft *^ IO *^ >* t.
L. > 4) 3
z gx> o c xi o
3. 3ป ป L. ซ O4> o4)4>4-ZX:
4) 4> *. . ฎ T3 O IO X
O O > O 4- 4-
03ซ COO 8 T 5 -ฐ 4) IO 4> C L.
O>> O> > 0 U O U U1 U ฃ > O U O
u a uxป aa u 4> u *
^ Q. ~- a ซ->. ซซ 4-xia
O O 4-in<-ปUU4-O>
in in 4) o> 4> 4) >o o o ป o >- E
ซ 4> 3C 3O 3rซ. 4-ป-in4)4-
9 > 9 > * coouino
Q 4) Q 4> BO ซ5 in IO ^ Qinaซ~3>-
(rxi (rx) >xi ปo > o a xi xi
in in m O O o
HI o O o
^
in in in o o o to ^.
QJ
- in 0 0 0 ซ* E
4)
U
10
t. u ซ - ซ m
XI O Q U L. L.
o o o o 004-
4) *^ 4) XI 4) ^ 4) O 4) O 4) O c ^*
4- ฃ 4-+-XI 4-4- 4-XI 4- XI XI 4- XI ~ 3
a u ซ>3 03 ioc 10 c T3
u ^ UO UO u u u n
OCMO OCM OCM .U3 QO> QO''0 O W ฃ QO1 OO1XIC4-
OEO UEO UEa O E 0 O E O O E 10 a
272
-------
>.
IO
2
4-
10
Q-
M
Q
0)
c
4-
4)
i
T3
C
0
3
A
4-
tn
5
1
15
o
*^
o
f2
~*
ซ
O
in
X
L4J
4-
tf)
^t
a
u
O
ป^
in
4)
3
a
ป
i_
4)
i
a
a
CL
"*3
a
(ft
O
4-
Q.
E
3
in
in
<
4-
o
U
LU
e
Q
L_
p
4>
U
C
4>
4)
4>
tr
c
o
4-
IO
C
IO
^^
X
4)
in
4)
x:
O)
X
4)
IO
E
in 4-
4> in
^
4)
4-
4- i
3 10
a. L.
C IO
a.
in 6
03 O
**
-o
>. 4)
4) 4-
a.
X 0
0} T3
X <
10
o
m
L.
C.
2
^
L.
X
in
.c
4-
c
Q
i
tO
^
*
in
IO
in o
00 9>
O PM
00 Ol
O CN
4-
C
10
C
j: _
4- C O C
c o in O
O 4- O 4- -1-
(O 4- 10 C
->-i- i_ in 4>
O 4- in 4- l_ Q.
c u c o in
O 4) O 4) O
O O O 'O 4)
4- C 'O C 4- E
10 O C O 3
a o o o i-
m
CD
^
^
>*
4)
X
10
X
f-
u
a
4-
o
u
ซ_
s
rf.*.
k ^_
(/) ฃ
1. 'ซ-'
8 T,
a
4-
3 .C
O U
s
1^
o
4)
a.
o
o
*
10
TJ
^
L.
4>
4)
in
t-
8
o
3
o
o
u
in
i
3
in
in
o
_
k
4)
X
ID
X
*-
U
10
4-
ง
U
'o
L.
PM
in
c
i
O
i.
c.
a
A
& ฃ
L. (J
8 u
T3 (1)
3
O .')
4-
m
4>
m
4>
o
^
o
+m
U
a
t-
c
o
u
o
in
in
t_
f-r
Iff
in
in
in
0
in
in
a.
m
1
t-
e
2
ซ*_
o
in 4> O>
00 *- 00
^S CL Oi
O
a
> c
o .. 10
X O) 4)
IO 4)
X X *
u
in
ฃ
4-
C
O
X
X
in
a
o
m
* >*
O) ^
c ^
ฃ O
in K>
IO
B
^v
ป L
m ฃ
L.
8 ฑ
4- 3
3 -a
O 10
ง
^
i^
o
4)
ป-
Q.
O
13
"*
in
3
o
u
g
c
O
.
X
IO
X
in
3
a
^
in
ฃ
O
E
U
4)
4-
C
..^
X
O
*
4-
U
IO
4-
o
(J
k
in
8
o
c
01
00
Ol
^
c
10
c
4)
4)
o
C
10
in
JC
4-
C
ง
L.
3
in
^O
1
>
IO
o
in
i.
ฃ
^*
i.
ฃ
o
JC
u
07?
-------
^
IO
X
4-
ID
Q.
_
a
E
L.
&
..
OJ
c
4
>
TJ
C
TJ
O
lfl
1
x
LU
4.
U)
^^
a
in
4)
3
a
3ป
l_
4)
4-
i
ซ
0
Q.
_
c
a
CA
ง
4-
Q.
3
I/)
in
<
ง
u
1 11
in
CM
&)
0
^
4>
U
C
4)
4)
41
a
c
O
4-
a
c
a
a.
X
4)
in
ฎ
i
O)
X
4}
4-
o
E
in 4-
o in
-
^
4)
4-
4)
4- S
3 a
a. L.
C 10
a.
ง
**ป
TJ
ซ
4-
0.
O
TJ
f.
4-
in
3
TJ
8
c
*~
**
o
^
ID
TJ
(A
ฃ
CM
in
4J
3
in
in
o
ao
*
S
3
4-
C
4)
a.
in
g
P
in
~
X
4)
X
a
4)
la
4-
u
a
4-
S
u
.
u
-C
_
.
.c
in u
8 4)
TJ TJ
~Q
L.
<*ป
T>
4>
4*
Q.
*
U
ID
4-
C
O
u
4
in
3
TJ
8
c
~
in
4)
E
3
in
<
i
u
a
a.
m
01
c
^
^
*
(A
8
TJ
C
.^
4-
C
4)
a.
in
4>
E
1
in
ao
ON
^~
>ป
4> '
X
ID
X
in c
L. *
x: j<
in
CM
c
0
in
3
ป *^
a. o
in
m ฃ
C 4-
c x
i
^
i_ m ฃ TJ
-t- 4-
c c CM
._ g _
m
4-
ซป ซJ 4-
O CM
in
vfl
O CM
in
ซ
4}
U
^ a
t- Q.
f m
^^
O)
+- c o
3 > 4-
TJ ~ 4-
ID IO
in
ao
O
f~
ซ,
>.
0)
%
(T5
X
4-
Ijj
^
TJ
C
ID
ฃ
4-
x 5
0
o m TJ
^ ซ ซ
in E m
L. 3 O
ฃ in a
in x
^
10
E
4-
m
4)
4-
m
f\
TJ
C
ID
X
O
01
ui
O
ex
X
4)
4-
4)
.^
TJ
C
a
m
en
4)
ซ
m
TJ
C
ID
f
^
^
in
g
3
in
m
CM
E
U
8
CM
"I
CM
E
u
8
CM
,->
4)
in
O
a.
x
4>
C
X.
in
^
O
ID
4>
L.
C
*O
C
^
-w IO
4) in
c i_ oo
O (0 Oi
. ^
m ex u -
ao e a >
O* 3 *- 4)
in L.
in 3 x
ซ ซj in ID
> X
4) .. ..
ฃ Oซ E
x ai co o
IO O* Lป
X X
,
a
"o.
01
*
u
3
TJ
[_
4)
^3
OCM
E
0 .
I- '
O T>
4 ซv
0 0
O)
c
^_
o
^
ao
O>
- in
E ON
3
ia
^^
o >-
oo o
X C
O 4)
I (/>
ซ
T3 m
4) 4)
> O
4) C.
4) in
in ป
i in
4- 4-
l_ C
O a
j=. a.
in
O) ^
c u
u c
10
4) C
X 4)
a.
m o
c
4- a
3 ฃ
TJ 4-
a
x
in
1 +:
3 C
in
in ฃ.
< tn
CM
e
u
o
TT
ON
(N
CM
U
O
ON
CM
TJ -^
4) CM
in E
0 o
a. "
X
4) 4-
C 3
T>
-^ a
U)
^ป I
ฐ 8
4) 4-
l_ 3
< O
in
CO
V
J^
4)
X
ID
X
41
4-
ID
.^
4-
in
4)
4-
m
m
,
4-
IO
^
in
4)
0
01
0
'-f_
^
in
^ ^~
ao
__ >^
O
e* TJ
3 C
ID 4)
t~ 1/1
U
-------
>.
X
f
+.
to
e^
^_
s
O)
C
4_
0)
<0
a
(O
c
Q
4-
13
J3
U.
4
_
Q
|
^_
to
3
^^
>
C
~
X)
<0
4_
0)
ซo
(L
x>
c
IO
in
O
4-
gi
a
in
in
<
4-
o
u
*^
uj
.
.
CM
_Z
^
IO
^_
4)
U
C
4>
4>
4>
CC.
C
O
4-
a
c
a
a.
x
4>
in
4-
2
01
X
4)
4-
a
E
1 I
l_
4)
4-
0)
4- E
3 a
a. L.
C ID
ex
in
00
Oi
^
^ป
4)
X
ID
h,
4)
X
ID
X
in
8
TJ
C
TJ
in
O
a.
x
4)
C
V
m
O
ID
41
L.
a
O
E m
a u
in g
in TJ
O 4-
E 3
3 O
m
in m
< a
CM
E
U
8
CM
to
CM
E
U
8
CM
TJ
4)
in
O
a.
x
4)
c 4)
TJ ~*
* CM
in o E
u
\^ * *
O i-
ซ 8 ^
4) TJ ~
i- C .C
< O
in
00
Q\
v~
^
>ซ
a>
x
ID
X
in
L.
8
TJ
C
TJ
4)
in
O
a.
x
4>
C
V
m
0
ID
4)
l_
a
ID U
u, g
in TJ
i ^
3 0
in
in n
< a
TJ
CM C
E C ID
O 4- O)
O o c a>
O .0 U
ซr 4- > ID
Oป e 4- a.
CM o m
CM C
E ฃ ID
O 4- O)
O U C 4)
O .0 (J
ซป 4- > ID
01 c 4- a
CM ID in
TJ
4)
in CM
0 E
a u
x **
4)
4-
C
^ 3
Jฃ TJ
in a
O L.
ID 8
4) TJ
I. C
4) "O 4) TJ
.c aj x: 4>
4- ID 4- E>O 4-ID4- ElO
E to O E ID O
l-l-C !_>. L.I.C .->-.
O4) -- O4) <-
-4-Ql.TJ 3 ^Qt-TJ 3
OcO"' O c O '
me IDE me IDE
CID.C 4)^in C(Dฃ 4)ซ
OEumEUoo OEumEO
3>.C CrtOl 3 >ป C (/)
4- x = a. 4- x r a.
a. O x - o a. O x ป cj
E^Q-Om - E >- a. O m
3O C ซ >- 3O C -
m O Q ra x: o in O O 4>I*-.Q{O 4>4>I**-X}
CEI.OOIOX CEUOOID
Qin3MNffl Oin3NNCD
Einmcc - Emincc
E4>O4)4)4}(7l E4)O4)4)4)O*
Oina.x)x3jฃco o^ao
o