RISK. EVALUATION OF OAT A COLLECTED DURING U.S. EPA's
1984 FIELD STUDY OF THE MIDLAND, MICHIGAN AREA
Prepared by:
J. Milton Clark, Ph.D
U.S. EPA Region V Pesticides and Toxic Substances Branch
October 11, 1985
-------
Acknowledgements
The author would like to thank the following Individuals who provided valuable
comments during the peer review process:
Donald Barnes, U.S. EPA
James Falco, U.S. EPA
Gregory Kew, U.S. EPA
Georgi Jones, Centers for Disease Control
-------
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
REGION V
DATE: 1 5 OCT 1985
subject Finalization of the Midland Risk Evaluation / , /
from. j. Milton Clark, Ph.D. THRU: Karl E. Bremer, Chief
Health Effects Specialist Toxic Substances Section
70 Willi am H. Sanders, III
Director, Environmental Services Division
Attached please find a copy of the report, "Risk Evaluation of Data
Collected During U.S. EPA's 1984 Field Study of the Midland, Michigan
Area". The document provides a multi-media exposure and risk assess-
ment on the chlorinated dioxin and dlbenzofuran contamination found
in Midland.
This risk assessment was produced in order to appraise the lifetime
cancer risks from individual media as well as all media combined.
The assessment has been developed in a rigorous fashion in accordance
with the proposed exposure - risk assessment guidelines of the Agency,
inducing the incorporation of uncertainties in the potential risks.
All peer review comments have been addressed in the final document.
The document, having provided a detailed discussion of risks, should
assist Region V risk managers in their decision-making regarding
possible remedial actions in Midland. It should also serve as a model
for other rigorous, multi-media risk assessments, which will become
more connicnplace in future years.
In May 1985, EPA and Centers for Disease Control signed a Memorandum
of Understanding in order to define shared responsibilities in their
hazardous waste programs. Under this agreement, EPA will have respon-
sibility for risk assessments. This presumably will include assess-
ments involving dioxins and dibenzofurans. Region V may therefore
dish to develop more in-house capacity to perform risk assessments,
perhaps even forming a special group within one of the Divisions.
Attachment
t = i '32W 3-76'
-------
Risk Evaluation of Data Collected during U.S. EPA's
1984 Field Study of the Midland, Michigan Area
Pesticides and Toxic Substances Branch •
Region V, U.S. EPA
I. Soil Data
A. Dow Chemical In-Plant Samples
1. 2,3,7,8-TCOO
Analytical results, found In Table 1 (attached), of analyses of 15
surface soil samples (0-1") taken Inside the Dow plant grounds revealed
2,3,7,8-TCDD concentrations ranging from 10 parts per trlllon (ppt)
to 30 parts per billion (ppb). The high value of 30 parts per billion,
station/sample #15, found south and east of building 874 and Immediately
north of the railroad tracks, represents an average of duplicate
analysis on the sample. The average soil concentration of the 15
samples was found to be 2.4 ppb, with three of the 15 samples having
more than 1 ppb 2,3,7,8-TCOO. The median value was 220 ppt. Sample
#5, collected southwest of building 955 and immediately southeast
of the Dow incinerator, was found to contain 3.50 ppb of 2,3,7,8-TCDD.
Sample #7, collected about 1/2 mile northeast of the Dow incinerator
at the northwest corner of 11th and J streets contained 4.60 ppb of
2,3,7,8-TCDD.
Samples #11, 12, 13, and 15 were all collected about 1/2 mile northeast
of the incinerator. Concentration of 2,3,7,8-TCDD in samples #11, 12,
and 13 were reasonably consistent ranging from 220 to 440 ppt; however,
sample 15, at 30 ppb 2,3,7,8-TCDD is disproportionately high. Samples
#4,7,8, and 9, collected in a line moving northwest from the Incinerator
to an approximate distance of one mile reveals an Inconsistent pattern.
Sample #4 was 20 ppt; sample #7, (1/3 mile) 4,600 ppt; sample #8, (3/4
mile) 150 ppt; and sample #9, (1 mile) 10 ppt.
2. Total PCDDs and PCDFs
PCDD and PCDF concentrations were determined in one soil sample, collect-
ed west of building 934 (station/sample #4). The concentration of
2,3,7,8-TCDD (0.27 ppb) was found to be 84% of the total TCDDs detected
(0.32 ppb). Concentrations of penta CDDs and hexa CDDs were 0.24 ppb
and 4.0 ppb, respectively. The ratio of hexa CDDs to 2,3,7,8-TCDD was
14.8. Hepta CDDs were found at a concentration of 75 ppb. Several PCDFs
were also detected 1n this sample. 2,3,7,8-TCDF was present at a con-
centration of 27 ppt, while total TCDFs amounted to 900 ppt. Hexa CDFs
and hepta CDFs were 3.1 and 15.4 ppb respectively.
B. Residential Soil Samples
1. 2,3,7,8-TCDD
Superficial soil sanples (0-1") were collected and analyzed from Midland,
Michigan; an industrial area in Middleton, Ohio; and four natural areas
in Minnesota. This sampling design was to determine if 2,3,7,8-TCDD
-------
-2-
contamlration was more or less unique to a location, such as Midland,
where 2,4,5-Trichlorophenol (2,4,5-TCP) and a variety of 2,4,5-T deriva-
tives had been manufactured. For more information on design and sam-
pling see the September 22, 1983 EPA document, "Soil Screening Survey at
Four Midwestern Sites" {!).
Rata from this sampling effort 1s attached and found in Tables 2-5A.
Yard composite soil samples from three locations, upwind (west/southwest)
of Dow Chemical, see Table 2, were found to contain non-detectable levels
of 2,3,7,8-TCDD. Detection limits ranged from 2 to 4 ppt. At two of
these locations soils collected near downspouts or under roof drip
lines, were found to have low levels of 2,3,7,8-TCDD; fi and 9 ppt.
Of four locations sampled from the perimeter of the Armco plant in
Middleton, Ohio only one soil sample of six collected and analyzed was
found to be positive for 2,3,7,8-TCDD and at a level of 4 ppt, (See
Table 4). In contrast, all nine samples collected along the perimeter
of the Dow-M1dland plant were found to be positive for 2,3,7,8-TCDD.
Concentrations of 2,3,7,8-TCDD ranged from a low of 10 ppt to a high
of 2.03 ppb, the average concentration being 327 ppt and the median
value, 69 ppt. (See Table 2).
Of seven public use locations sampled 1n Middleton, Ohio only one was
found to contain 2,3,7,8-TCDD at a concentration of 4 ppt. All nine
public use areas sampled in Midland were found to have 2,3,7,8-TCDD in
soils with values ranging from 3 to 170 ppt. The arithmetic mean was
56 ppt; the median value, 19 ppt. The highest concentrations of
2,3,7,8-TCDD observed were those collected from locations most near the
Dow plant or located from east northeast to north northeast of the
facility. As prevailing wind direction In Midland, see Figure 1, in "Soil
Screening Survey at Four Midwestern Sites," is from the southwest or
westerly direction, deposition of 2,3,7,8-TCDD by air is suggested.
Of the 19 residential superficial soil samples (0-1") analyzed in
Midland, 18 were positive for 2,3,7,8-TCDD. Concentrations ranged from
non-detectable to 270 parts per trillion. The arithmetic mean of this
data was 54 ppt: the median, 22 ppt. Two of seven soil samples from
residential areas in Middletown, Ohio were found to be positive for
2,3,7,8-TCDD. These samples, both found at one location, contained 4
and 5 ppt of 2,3,7,8-TCDD. The median value of all the Ohio samples
was "non-detectable" (< 3 ppt). An average value of 2.2 ppt was
reported by Dow Chemical In soils from 20 industrialized areas in the
U.S. (November 5, 1984 report) (2). None of the three natural areas
sampled in Minnesota contained detectable levels of 2,3,7,8-TCOO.
(See Table 5). The data from our studies of the four natural areas,
and Middleton, Ohio, when combined with the results on 20 industrial-
ized areas reported by Oow, reveals higher levels of 2,3,7,8-TCDD in
Midland, than in other areas studied. If fish data is included for
comparison, discussed later, this finding becomes even more apparent.
-------
-3-
Concentrations of 2,3,7,8-TCDD found In soils from yards tended to de- -
crease with distance from Dow chemical. (See Table 2). For instance,
all yard samples, B-3, D-3, ami E-3, colWcted about 3700 yards from the
How incinerator, were found to have lower concentrations of 2,3,7,8-TC00
than samples, B-l, C-l, and E-l, which were all taken about 210(1 yards
from the incinerator. This pattern suggests 2,3,7,8-TCDD deposition by
air. An air deposition hypothesis is further supported by comparing the
2,3,7,8-TCDD levels observed in soils collected near downspouts or under
dripTines with those found in yards. In most cases downspout and dripline
soil concentrations of 2,3,7,8-TCDD were approximately two to ten times
higher than those found in yards. At sample location F, the downspout
sample was less than the yard sample and could be caused by the disturbing
or addition of soils near downspouts, for Instance by planting.
2. Total PCPPs and PCDFs
Concentrations of PCDDs and PDCFs in surface soil samples are presented
in Table 6 (attached). Samples collected upwind of Dow Chemical (#13401
and #13395) contained only low levels of hepta and octa CDDs (<340 ppt)
and no detectable levels of PCDFs, All six public access areas sampled
1n Midland had measurable concentrations of hexa CDDs, ranging from 63
to 410 ppt and all samples contained PCD^s, three with 2,3,7,8-TCDF and
hexa CDFs (sample is 13375, 13393, and 13394). Of five samples analyzed
from Middletown, Ohio only one contained detectable levels of hexa CDDs
and only one with detectable levels of PCDFs. No hexa CDFs were found
in the soil samples. Surface soil samples from three wilderness areas
in Minnesota contained low levels of hepta and octa CDDs (<200 ppt) and
no detectable levels of PCOFs.
3. Computation of 2,3,7,8-TCPD Equivalent Bisks
Methods are currently being developed to assess the risk from exposure
to chlorinated dioxins and dibenzofurans other than for 2,3,7,8-TCDD.
These are discussed in the September 26, 1984 draft EPA document en-
titled "Health Hazard. Assessment for Chlorinated Dioxins and
-Dibenzoforans other than 2,3,7,8-TCDD" (3). Of the eight methods cited
in the document two, the Swiss enzyme and the EPA methods, appear to
have the best scientific foundations as they are based upon enzyme
inductive properties and carcinogenic assessments. These methods are
shown below. Where isomer specific analyses were unavailable, the
tbxicity of the 2,3,7,R-substituted homologues is utilized.
Table A: Estimates of Relative Toxicities For PCDDs and PCDFs
Expressed as Equivalents of 2,3,7,8-TCDD
Compound
EPA Method (1984)
Swiss Method
2,3,7,8-TCDD
1
1
other TCDDs
O.Ol
0.01
2,3,7,8-PeCDDs
0.2
0.1
other PeCDDs
0.002
0.1
2 ,3,7,8-HxCDOs
0.04
0.1
other HxCDDs
0.004
o.l
2,3,7,8-HpCDDs
n
0.01
other HpCDHs
0
0.01
OCDDs
0
-------
-4-
Table A: Continued
Compound EPA Method (1984) Swiss Method
2,3,7,8-TCDFs 0.1 0.1
other TCDFs 0.001 0.1
2,3,7,8-PeCDFs 0.1 0.1
other PeCDFs 0.001 0.1
2,3,7,8-HxCDFs 0.1 0.1
other HxCDFs 0.001 0.1
HpCDFs, OCDF 0
PCDD and PDCF data from Midland, converted to 2,3,7,R-TCDD equivalents,
is presented in the following section.
Table B: 2,3,7,8-TC0D Equivalents of Other
PCDDs and PCOFs Found in Midland Soils
Location: Bullock School Virginia Park
Method Method
PCDDs/PCDFs
Level
EPA
Swiss
Level
EPA
Swiss
(ppO
(ppt)
2,3,7,R-TCDD
110
110
110
76
76
76
TCDDs
220
2
2
290
3
2
PeCDDs
120
24
12
110
20
10
KxCDDs
410
16
41
240
10
24
HpCODs
2,400
0
24
410
0
4
2,3,7,8-TCOF
15
2
2
13
1
1
TCDFs
-
»
-
-
-
-
PeCDFs
110
11
11
40
4
4
HxCDFs
170
17
17
64
17
17
Eauivalents
(ppt):
182
- 219
120
- 127
Longview School Mapelton School
2,3,7,8-TCDD
78
78
78
15
15
15
TCDDs
170
1
1
40
0
0
PeCDDs
100*
20*
10*
ND
0
0
HxCDDs
340
14
34
63
2
6
HpCDDs
2,300
0
23
380
0
4
2,3,7,8-TCOF
13
1
1
ND
0
0
TCDFs
-
-
-
-
PeCDFs
ND
0
0
ND
0
0
HxCDFs
260
26
26
ND
0
0
Equivalents
(ppt):
-
H7T
IT -
T5
* value estimated
-------
-5-
Table B: Continued
Location: Central School County Line Road
(Ball diamond)
PPt
EPA
Swiss
PPt
EPA
Swiss
2,3,7,8-TCDD
12
12
12
3
3
3
TCDDs
40
0
0
ND
0
0
PeCDDs
ND
0
0
ND
0
0
HxCDOs
8fi
3
9
67
3
7
HpCDDs
350
0
4
350
0
4
2,3,7,8-TCDF
ND
0
n
NO
0
0
PeCDFs
NO
0
0
ND
0
0
HxCDFs
ND
0
0
ND
0
0
Equivalents (ppt):
15 -
25
6 -
14
The PCDD and PCDF concentrations expressed as 2,3,7,8-TCDO equivalents
range from a low of 6-13 ppt to a high of 162-219 ppt. Relatively small
differences exist between the Swiss and EPA methods to derive 2,3,7,8-
TCOn equivalents for other PCDOs and PCDFs. Three of the locations
sampled; Bullock School, Virginia Park, and Longvfew School have
2,3,7,8-TCDD equivalents of 7 to 20 times the equivalents found in soils
at County Line Road, Central School, and Mapelton School. Similar
ratios of 2,3,7,8-TCDD equivalents to 2,3,7,8-TCDO measured soil levels
can be observed. At Bullock School this ratio is 1.82; 1.62 for Virginia
Park and 1.99 for Longview School. The average of these ratios is 1.82.
The general consistency in the ratio suggests a single primary source
for the PCDDs and PCDFs observed in these residential, public use soils.
Assuming that a ratio of 1.B 2,3,7,8-TCDD equivalents to 2,3,7,8-TCDD
exists for all other locations sampled in Midland, it is possible to
derive estimated 2,3,7,8-TCDD equivalents for residential soil samples
from measured 2,3,7,8-TCDD soil data. An average yard soil concentration
of 2,3,7,8-TCDD can be estimated assuming that concentrations of 2,3,7,8-
TCDD in soils near downspouts represent 10% of the area of the yard
contaminated at this level and 2,3,7,8-TCDD concentrations in yard com-
posite soil samples away from downspouts, represent 90% of the contami-
nated soils. These calculations are shown below in Table C.
Table C: Average 2,3,7,8-TCDD Equivalent
Levels in Residential Surface
Soil Samples in Midland, Michigan1
Estimated
Measured Average 2,3,7,8-TCDD
2,3,7,8-TCDD 2,3,7,8-TCDD Equivalents
Soil Concentration in Soils
(ppt) (PPt) Tppt)
B-l-L1 7*> 84 152
B-l-12 160
R-3-L 20 45 81
8-3-1 270
Field
Number
-------
-6-
%
Estimated
Field Measured Average 2,3,7,8-TCDD
Number 2.3,7,6-rCDD 2J.7,fi-tCDD Equivalents
Soil Concentration in Soils 2
(ppt)
(ppt)
(ppt)
C-l-L
26
28
50
C-l-1
54
C-3-L
12
35
64
C-3-1
240
D-3-L
18
19
34
D-3-1
31
E-l-L
26
28
50
£-1-1
49
E-3-L
9
10
18
E-3-1
20
F-l-L
31
29
52
F-l-1
13
UPW-l-L
ND (2)
1
2
UPW-1-1
6
UPVI-4-L
ND (4)
1
2
UPW-4-D
9
^Equivalents based upon average of EPA and Swiss Methods
L » Yard composite soil sample, 0-1"; 1 » downspout composite sample, 0-1";
UPW = Upwind samples; NO = not detected (detection limit)
As shown in Table C, superficial soil concentrations of PCDD and PCOFs,
expressed' as 2,3,7,8-TCDD equivalents, would range from 18 to 152 ppt.
Seven of the eight downwind locations have 2,3,7,8-TCDD equivalents
of 50 ppt or greater. The average is 63 ppt; the median, 51 ppt. In
contrast, downwind samples from Middletown, Ohio would have 2,3,7,8-TCDD
equivalents ranging from not detected to 5 ppt, with an average of 3 ppt
and a median of 1 ppt. For two of the sites, R-l and C-3, concentrations
of 2,3,7,8-TCDD in downspout soils make a large contribution to the
overall yard average of 2,3,7,8-TCDD.
C. Risk Assessment from Exposure to 2,3,7,8-TCDD and other PCDDs and PCDFs
in Midland Residential Soils
i
1. Background
Both EPA and Centers for Disease Control (CDC) have prepared risk assess-
ments for exposure to 2,3,7,8-TCOD 1n soils. EPA's first exposure and
risk assessments were performed by the Chlorinated Oioxin Work Group in
response to the finding of 2,3,7,8-TCDD contamination 1n Hyde Park, Love
Canal , and at the Syntex site. Based upon these assessments, EPA entered
into consent decrees and administrative orders to require cleanup and
capping of contaminated soils to the detection limit of 2,3,7,8-TCDD
(from 1-50 ppt} (4). In September 1982, EPA prepared an exposure and risk
assessment for contaminated sites found in Missouri ("Region 7
Exposure/Risk Assessment", Draft, September 28, 1982) (5). The cancer
risks from exposure to 2,3,7,8-TCDD in residential soils were determined
using the Carcinogen Assessment firoup Model. The highest cancer risk
-------
-7-
was found to be for young children who could Ingest soil by the behavior
called pica.
In November, 1984 EPA published a detailed, peer-reviewed exposure-risk
assessment approach for 2,3,7,8-TCDD contaminated soils (21). A number of
similarities exist with the EPA r 1 sit assessment and that developed by
CDC, including the utilization of a 10-12 year half-life for 2,3,7,8-
TCOD in soils. The most significant difference Is the usage of EPA's
cancer slope extrapolation, degree of dermal exposure to soils, and a
quantitative appraisal 1n the uncertanties of exposure. These differences
result 1n calculated cancer risks about two orders of magnitude greater
for a given soil concentration of 2,3,7,8-TCDD using the EPA exposure-risk
model. A two order of magnitude difference 1s significant enough
to have an Impact on the risk management decisions of the Agency. For
Instance, CDC has estimated that the excess upper 95% lifetime cancer
risk from exposure to soils containing lppb 2,3,7,8-TCDD to be approximately
10-5, while with the more conservative EPA model cancer risks could be
as great as 1G"3. The two agencies also differ In what constitutes an
unacceptable risfc, with CDC having stated that a 10-5 excess cancer
risk is unacceptable while 1n EPA's Superfund program a 10-6 risk has been
considered unacceptable (23). This effectively means that EPA could have
concern in situations where soil concentrations of 2,3,7,8-TCDD exceed lppt.
In early 1983 EPA requested CDC to determine If a 1 ppb level of 2,3,7,8-
TCDD in soils represented an appropriate level of concern. This action,
In part, was pronpted 1n part by the widespread utilization of 1 ppb detection
limits by Region 7 for their analyses of 2,3,7,8-TCDD in soils. Based
upon CDC's recommendations that a lppb soil level of 2,3,7,8-TCDD was
unacceptable, EPA evacuated Times Beech residents In February, 1983 (23}.
In January 1984, CDC published an abbreviated analysis of their risk assess-
ment for exposure to 2,3,7,8-TCDD which concluded that a level 1 ppb of
2,3,7,8-TCDD in residential soils was an appropriate level of concern
for human health (7). A more complete 2,3,7,8-TCDD exposure and risk
assessment document was released by CDC in July 1984 ("Health Implications
of 2,3,7,8-TCDD contamination of Residential Soil," Kimbrough, et
al.) (8). The authors concluded that soil levels greater than 1 ppb,
1s a level for concern, however, 1t was recommended that management
decisions should also consider an evaluation of specific circumstances
at each contaminated site. COC's risk assessment considered 2,3,7,8-TCDD
via exposure to soils Including that from dermal contact and by inhalation.
Air and water contamination of 2,3,7,8-TCDD were believed not to make
a significant contribution to TCDD Intake 1n most situations, and
therefore was not considered in the model. However, CDC did state
that TCDD could eventually end up in the food chain, particularily 1n
fish. Under these conditions CDC stated, u1f TCDD enters a food chain,
there Is an unknown additional source of exposure which rust be added
to the risk of those individuals exposed to contaminated soil and of a
larger, undefined population". CDC also commented that 1f contaminated
soils were near waterways and could contaminate these waterways, acceptable
levels of 2,3,7,8-TCDD nay need to be lower (than 1 ppb).
The CDC risk assessment 1s based upon a liner-multistage extrapolation
of the Kociba animal cancer data for 2,3,7,8-TCDD as interpreted by
Squire (8). CDC's extrapolation differs from EPA in three ways (6).
First, the curve fit was performed not on administered dose but on liver
concentration at sacrifice. Second, CDC's extrapolation does not adjust
-------
-8-
for high early mortality of test animals. Finally, CDC uses a rat-to-
man extrapolation on the basis of body weight rather than surface area.
These modifications result in a 95% upper-limit potency value estimate,
when corrected back to administered dose, yields a slope factor of 3,6 x
10^ (mg/kg-bw/day)-^ which is less potent by a factor of 4.3 than the
1.56 x 10^ factor derived by EPA. Roth agencies have used a 70 kg
Individual was used as a model. A potentially higher risk Individual,
one of lower body weight, such as a 50 kg person (female) can be used when
performing risk assessments. Such an approach protects those persons
in higher risk groups.
Using COC's exposure and risk assessment assumptions, exposure to 1 ppb
of 2,3,7,8-TCHD in soil over lifetime has an upper 95% probability of
causing cancer of 2.3 x 10-5. Expressed siightiy differently, at a con-
centration of 1 ppb, exposure to 2,3,7,8-TCDD in soils could accumulate
to a sufficient dose in just three years to increase an individual's risk
of cancer by one chance in a million (8). If no more than 10"® upper proba-
bility of causing cancer is desired, using upper 95% confidence limits,
then a soil concentration of 44 ppt would be an acceptable level of
2,3,7,8-TCDO using the CDC exposure-risk model. If EPA's extrapolation
of animal cancer is employed in the COC model these values decrease to
10 ppt for a 10"® risk. If EPA's exposure-risk assessment model is
employed, a soil level of about lppt 2,3,7,8-TCDD could give an upper
limit 95% lifetime cancer risk of cancer of 10"5, using the most conservative
exposure estimates. Using the least conservative exposure assumptions,
a soil level of lOOppt could also result in a cancer risk of 10"" with the
EPA model.
2. Risk From Exposure to Residential Soils
Table D, below, presents the upper excess 95% lifetime cancer risks from
exposure to contaminated soils observed in Midland, Michigan using (1)
EPA's 1984 exposure and risk model (2) CDC's exposure and risk model and
(3) CDC's exposure and risk model using EPA's 1984 animal extrapolation
and a 50 kg Individual as a model.
Table D: Estimation of Cancer Risks from Exposures to
Soils in Mid land, Michigan using Risk
and Exposure Models of EPA and CDC*"
Estimated
Location Average 2,3,7,8-TCDD
(Field #) Equivalents in Soil
(ppt)
Cancer Risk Models
EPA 1984
CDC CDC Modified
B-l-L
ft-3-L
C-l-L
C-3-L
D-3-L
E-l-L
E-3-L
F-l-L
152
81
50
64
34
50
18
52
2x10-4 - 2x10*6 4x10-6 2xlO"5
8xl0-5 - 8x10-7 2xl0"6 1x10-5
5x10-5 - 5x10-7 1x10-5 7x10"5
6x10-5 - 6x10-7 2xl0-6 9x10"6
3x10-5 - 3x10-7 8x10-7 5x10"5
5x10-5 - 5x10-7 IxlO-5 7xl0"5
2x10-5 - 2x10-7 4x10-7 3x10-5
5x10-5 - 5x10-7 1x10-5 7xl0"5
Upwind (2 samples) 2
2x10-5 - 2xl0-8 4x10"s 3xlO"7
-------
-9-
Table D: Continued
Location
<1
EPA 1984
CDC CDC Modified 1
3x10"® - 3x10-8 8xl0-8 5xl0"7
<10-5 - 10-8
<10-8
<10"7
* EPA slope factor of 1.56xlQ5 (mg-kg-bw/day)~1; 50 kg-bw individual.
(Modified model 1s 6.Ox CDC's original model)
All three cancer risk models yield roughly equivalent risks. The EPA
model yields the highest risks, with all downwind sample locations
in Midland having potential cancer risks exceeding 10-5t even if just
2,3,7,8-TCDD Is considered. For 2,3,7,8-TCDD equivalents greater
than 50 ppt all models produce cancer risks exceeding 10-°. Above 75
ppt the CDC modified model indicates cancer risks of 10-5 or greater,
at the 95% upper confidence limit.
It should be noted that the exposure and risk assessments for 2,3,7,8-
TCDD may not adequately consider Intake of soil from vegetables or
gardens grown in contaminated soils. While 2,3,7,8-TCDD bioaccumula-
tlon does not occur, coating of plant material by 2,3,7,8-TCDD contami-
nated soils has been documented (24,25). CDC assumes 1n their exposure
and risk assessment a soil ingestion of 100 mg/day, although it is
not clear if soil ingestion from garden work or from consumption of
foods grown 1n gardens has been Included in this figure. Some authors
have estimated soil Intake by geophagia to be as great as 3g/day (9).
If this value were substituted into CDC's assessment model the risks
from 2,3,7,8-TCDD would be dramatically increased. A range of assump-
tions can also be made to estimate dermal exposure to soils. CDC
estimated dally dermal exposure after age 15 as 100 mg/day. According
to CDC, 10 grams of soil less than 1 mm thick can be spread over an
area of about 15cm2 or about 6 inches2 (8). This is less of a
surface area which exists for one side of a hand. For a individual
working in a garden, an exposed skin surface area of 30 inches2 and
contact with 50g of soil may be more realistic. While such direct
exposure would only exist for a portion of the year, for instance
5-6 months, one or more days per week, 1t could be a very significant
route of exposure If human dermal absorption of 2,3,7,8-TCDD exists.
In animal studies, Poiger and Schlatter determined that 0.1-61 of
dermally applied 2,3,7,8-TCDD in applied soils was absorbed (10). If
these soil exposure assumptions, are placed into the CDC model, risks
from exposure to 2,3,7,8-TCDD in soils would be increased. The EPA
exposure-risk model, by using a range of potential dermal exposures
to soils as well as by ingestion, better addresses the uncertainty
of soil exposure and associated risks and would therefore be the
preferred dioxin risk model.
-------
-in-
II. Ambient Air Data
Data provided by Dow Chemical on 2,3,7,8-TCDD concentrations in ambient air
are presented below in Table E (2). A total of 10 measurements were taken:
two within the Dow Plant, six at the Dow Plant fence line and two in the
city of Midland. All samples were found to be positive for 2,3,7,8-TCDD.
The data generated indicate rather large fluctuations 1n air concentrations
Of 2,3,7,8-TCDD.
Table E: Ambient Air Concentrations (pg/m3) of 2,3,7,8-TCDD
in Midland, Michigan (March 1983-February 1984) 1
Number of Arithmetic
Location Measurements Range Mean
Dow Plant 2
Dow Plant Fence Line 6
City of Midland 2
0.019-0.022
0.010-0.21
0.019-0.16
0.021 o.oii
0.064 •***/
0.090
io L *Q a fo &
iData supplied by Dow Chemical (Reference #2)
2Data of EPA unavailable at this time
Risk assessments from inhalation of 2,3,7,8-TCDD have been made by EPA in
regards to emissions from municipal waste incinerators. Exposure, bio-
availability and other assumptions are discussed in more detail in the
document, "Assessment of Emissions from a Recent Municipal Haste Combustor,"
December 16, 1983 (17). Given these assumptions and incorporating the most re
cent change in the cancer slope determination for 2,3,7,8-TCDD, the lifetime
cancer risks from exposure to various air levels of 2,3,7,8-TCDD, measured
in Midland are shown below in Table F. Using a ratio of 1.8 to estimate
total risks from PCDDs and PCDFs as done for soils; 2,3,7,8-TCDD equivalent
risks for estimated air levels of PCDDs and PCDFs have been derived.
Table F: Lifetime Cancer Risks from Exposure
to Ambient Air Levels of 2,3,7,8-TCDD
and Estimated PCOOs and PCDFs in Mid land
Concentration Cancer Risk Cancer Risk
(pq/m3) 2,3.7.B-TCPfl PCDDs/PCDFs
0.21 (high value 6 x 10"® 1 x 10-5
at Dow Plant
Fence line)
0.16 (high value 5 x 10"* 1 x 10-5
in Midland)
Fence 1ine)
'average vali __
in Midland) ^
n.09 (average value £& 3 x 10-6 5 x i0-6 J.3
1 As upper 95* confidence limit
2 Estimated from the ratio of PCDD and PCDF to 2,3,7,8-TCDD in soil
-------
-11-
As shown from data presented in the Table F, cancer risks from
lifetime exposure to 2,3,7,8-TCDD and from the estimated concentrations
of total PCDDs and PCDFs are all in the 10-5 to 10"® range. No
information 1s available on historic air levels of PCDDs and PCDFs In
Midland, although particulate matter from Dow's rotary kiln
incinerator in 1978 operated without supplementary fuel had
reticulate concentrations of ?,3,7,ftTr.nn as frigh as Y
13-200 pph h»Y*_fnnt fi^.uuu ppbJi/i. With supplementary
fuel, the concentration of PCODs in particulate matter was
reduced to low ppb levels, or by a factor of 1000 or more.
Historic operations of these and other incinerators, as well as
fugitive PCDD emissions from the facility when 2,4,5-T manufacture
was occurring, could have given rise to far higher air levels
of PCDDs and PCDFs 1n previous years than those observed currently.
If so, exposure to PCODs and accumulated cancer risks may have
reached unacceptable levels for long term residents of Midland.
Using the 1978 emmisslon data reported by Dow Chemical, the
Michigan Department of Natural Resources, calculated
a lifetime excess cancer risk of 9x10-5 (22).
In Dow Chemical's November 5, 1984 report of point sources of
2,3,7,8-TCDD in Midland, historic waste Incinerator practices were
identified as the principal sources of 2,3,7,8-TCDD soil contamination,
as current emission rates from the rotary kiln incinerator and
three other smaller Incinerator sources could not account for
levels of 2,3,7,8-TCDD which have accumulated fn soils In
Midland or at the plant site (2). Dow reported that incinerator
combustion gases were directly vented to the atmosphere until
1968 and high temperature incineration, needed to destroy
dlwas nn+ fully "implemented until 1978. A1 thOugfThiStorlc
air concentrations of PCDDs and PCDFs are unknown at this time,
1t should be possible to produce models using records of past
disposal practices and Incinerator operation in order to produce
estimates of previous exposure and associated risks.
Adding to the risks calculated for PCDDs and PCDFs are air
concentrations of other carcinogens or suspect carcinogens
found presently or historically in Midland. The Michigan
Department of Natural Resources has accumulated air data in
Midland which should be evaluated in order to assess additional
risks. In one national survey of air emissions from industrial
facilities, using estimated emission data, the Dow facility
1n Midland was ranked as the number one air source of toxic
siijstances 1n the State of Michigan (19). As lifetime
cancer risks from exposure to only PCDDs and PCDFs in
ambient air are in the 10"® to 10*5 range, monitoring for
additional hazardous chemicals 1s strongly recommended.
-------
-12-
III. Cancer Risks from Exposure to Soils and Antlent Air Levels of PCDDs
In CDC's exposure and risk assessment It was assumed that concentrations of
2,3,7,8-TCDD 1n airborne dusts would equal those found 1n soils. Further-
more, 1t was assumed that Indoor air levels of 2,3,7,8-TCDD on particulate
matter would equal those found 1n ambient air. This assumption was viewed
by some commenters as overly protective. Assuming that Midland residential
soils average 50 ppt of 2,3,7,8-TCDD and total particulate matter averages
70 ug/m3, the expected maximum concentration of 2,3,7,8-TCDD 1n air would
be 3.5 fg/m3. Air levels In Midland exceed this value by as great as 50
fold, suggesting that residential soil alone could not be responsible for
the observed concentrations of 2,3,7,8-TCDD In ambient air in Midland. A
value of 70 ug/m3 1s the primary air standard for total suspended particulates.
It 1s unclear what particulate matter concentrations were used in CDC's
inhalation exposure assessments. Therefore, with the CDC risk models 1t
may not be valid to add cancer risks from measured air concentrations of
2,3,7,8-TCDD and other PCDDs to those risks calculated from soils. If CDC
assumed dust levels of 140 ug/m3, as mentioned in their report, then a
soil concentration of 1 ppb, would result 1n air levels of 2,3,7,8-TCDD
approximately equal to those measured 1n Midland. Therefore, If the
CDC soil exposure risk model was derived with these assumptions 1t Is
inappropriate to add risks from ambient air exposure to 2,3,7,8-TCDD to
those previously found for soils. It 1s interesting to note that the
average soil concentration observed on the Dow plant site was 2.38 ppb
and 0.74 ppb along its perimeter. Given the very large area of the Dow
plant site, 1500 acres, airborne dust and soils from the site may be a
primary contrfbuter to current air levels of 2,3,7,8-TCDD. Modeling
should be conducted to determine contributions from existing air sources
in Midland to those from plant and residential, soils.
Using the EPA model, cancer risks from Inhalation could be added. For
soils containing 50 ppt 2,3,7,8-TCDD equivalents, lifetime cancer risks
could increase from an upper range of 5 x 10-5 to 6 x 1Q"5 If the cancer
risk from lifetime exposure to the peak value of 0.16 pg/m3 2,3,7,8-TCDD
converted to 2,3,7,8-TCDD equivalents for estimated air levels of other
PCDDs and PCDFs 1s included. If the Intermediate risk level of the
EPA model is used, or the CDC model, lifetime cancer risks from exposure
to soils 1s dominated by cancer risks calculated for air Inhalation.
IV. Ftfsh Analysis
1. Overview of 2,3,7,8-TCDD Fish Data
Fish from 13 Michigan rivers were analyzed In 1983 to determine flesh
concentrations of 2,3,7,8-TCDD. A total of 74 fish were analyzed. Whole
fish carp samples ranged from non-detectable to 190 ppt of 2,3,7,8-TCDD,
with the high value found 1n carp collected from the Tlttabawassee River
in Midland. This data 1s found In Table 11 (attached). The average of
2,3,7,8-TCDD found 1n whole carp 1n rivers other than the Tlttabawassee
averaged 3 ppt, or 1.6% of the concentration observed in whole fish from
the Tlttabawassee River. No fish analyzed exceeded 10 ppt. A statisti-
cal t-test of this data reveals that carp from the Tlttabawassee river
are significantly different from fish from other Michigan rivers at a p
value greatly exceeding 0.001. It therefore can be stated with virtual
certainty that Tlttabawssee fish are uniquely contaminated.
-------
-13-
Ind1vidual carp fillet samples from the Tittabawassee River ranged from 12
to 530 ppt 2,3,7,8-TCDD, with an arithmetic mean of 50 ppt and a median
of 25 ppt. A skinless fillet composite of five catfish averaged 75 ppt of
2,3,7,8-TCDD while a composite of skin-on fillets of five smallmouth bass
contained 5.1 ppt of 2,3,7,8-TCDD. Five Individual skin-on fillets of wall-
eye averaged 3.9 ppt of 2,3,7,8-TCDD.
2. Overview of PCDD and PCDF F1sh Data
A composite of five whole carp samples analyzed in 1981, downstream of
Midland were found to contain 81 ppt TCDD, 31 ppt penta CDD, 44 ppt hexa
CDD, 53 ppt hepta CDD and 14 ppt octa CDD. PCDFs were found to be as
follows: 37 ppt TCDF, 73 ppt penta CPF, 145 ppt hexa CDF, 31 ppt hepta
CDF, and 4 ppt octa CDF. About 90% of the total TCDD observed was 2,3,7,8-
TCDD (18).
3. Exposure and Risk Analysis From Consumption of Tittabawassee River Fish
Although a fish advisory against consumption of fish from the Tittabawassee
River has been in place since 1978, there have been reports of individuals
consuming fish from the river (20). This 1s not to be unexpected as many people
tend to ignore fish advisories, particularly those in poorer economic groups.
For people consuming even small amounts of bottom fish from the Tittabawassee
River over their lifetimes, the cancer risks are fairly dramatic. To provide
a perspective of such risks, Table G has been developed which gives frequency
of fish consumption and associated upper 95% excess lifetime cancer risks
for different species, and for both 2,3,7,8-TCDD and estimated total PCDDs and
PCDFs as 2,3,7,8-TCDD equivalents. PCDDs and PCDFs other than 2,3,7,8-TCDD
have been estimated from 1981 composite carp analysis by the ratios of congeners
to total TCDD, assuming that 90% of the measured TCDD is 2,3,7,8-TCDD.
Table G: Lifetime Cancer Risks From Consumption of Tittabawassee
River Fish in Midland
Frequency of Carp Catfish Sportfish
Consumption1
2,3,7 8-
TCDD*
Total
PCDDs/PCDFs3
2,3,7.8
TCDD4 '
Total
PCDDs/PCDFs5
2,3,7.8
TCDD6
Total
PCDDs/PCDFs7
Once
per week
5xl0-3
-J
X
o
1
CO
8xl0-3
lxlO-2
4x10"4
6x10-4
Once per
month
lxlO"3
2x10-3
2x10-3
3xlO-3
lxlO"4
1x10-4
Once per yr.
lxlO"4
1x10-4
2xl0"4
3xl0"4
8x10"6
1x10-5
Ten times
over
Li fetime
lxlO-5
2xl0-5
2x10-5
3x10-5
lxlO"®
2x10-6
1 Assumes 1/2 pound fish per meal; 2 Average value of^Oppt^)
3 Calculated from ratios of PCDDs/PCDFs observed in composite sample of carp
analyzed from the Tittabawassee River in 1981. Both Swiss and EPA models yield
2,3,7,8-TCDD equivalents of 22 ppt for PCDDs and PCDFs other than for 2,3,7,3-TCDD
-------
-14-
* 75 ppt
5 See footnote #3 above. Both Swiss and EPA models yield 2,3,7,8-TCDD equivalents
of approximately 35 ppt for PCDDs and PCDFs other than 2,3,7,8-TCDD.
6 4 ppt
7 See footnote #3 above. PCDDs and PCDFs other than 2,3,7,8-TCDD have equivalents
of 2 ppt.
Cancer risks for the Infrequent consumer of Tlttabawassee River fish, for
Instance ten times over lifetime, equal or exceed a 10*5 risk factor and
10-6 for sportflsh. For more frequent consumers of carp and catfish, even a 1/2
pound meal once per year, cancer risks exceed 10"*. If consumption 1s more
frequent, once a month, risks exceed 10"3. Consumers of only sportflsh would
have excessive risks, greater than 10-4, for consumption equaling or exceeding
1/2 pound meal monthly. A once a month consumption of 1/2 pound of fish 1s
equal to 7.3 grams of fish daily or slightly higher than EPA's average fish
conscription value of 6.5 g/day used 1n ambient water quality criteria docu-
ments. It 1s likely that a small fraction, of Midland's population would
consume bottom fish or sportflsh as frequently as once per year or once per
month, adding significantly to cancer risks received from exposure to PCDDs
and PCDFs in soils and ambient air.
Probably many fishermen of the Tlttabawassee River would tend to discard
carp being less desirable for eating than catfish or sportflsh; however,
this would not necessarily be true for subsistence fishermen, as Identified
by a survey conducted by Smith and Thomson (21). Perhaps the most realistic
way to generate an average risk 1s to assune that a consigner would eat equal
amounts of carp, catfish, and sportflsh. For an Individual consuming 1/2
pound of fish per year the cancer risk would be 2x10"^; on a monthly basis
the risk would be 2x10-3 for total PCDDs and PCDFs expressed as 2,3,7,8-TCDD
equivalents. Other chenricals, such as PBBs and pentachlorophenol, have also
been detected 1n Tlttabawassee fish, so the risk estimates calculated here,
may underestimate the cancer risk. For women consuming a 1/2 pound meal of
fish weekly with an average of 63 ppt 2,3,7,8-TCDD equivalents, Intake would
be over 2,000 pg/day or about 40 pg/kg-bw/day for a 50 kg woman. Short term
sub-hunan primate studies have demonstrated reproductive effects, Including
conception failures and miscarriages at a dosages as low as 1.8 ng/kg-bw/day
(6,8). Clearly woman consuming meals of f 1 sh from the Tlttabawassee in
Midland would have a high risk of adverse reproductive outcomes. If a safety
factor of 1000 is utilized for humans, as by CX, then no consumption by
wome^ of any quantity of Tlttabawassee River fish would be recommended.
Y. Cumulative Risks from Exposure to Soil, Ambient Air, and F1sh 1n Midland
and Recommendations for Remedial Actions
Total risks from exposure to PCDDs and PCDFs in all media are presented 1n
Table H. The assessment does not include exposure to other chemicals which
may be in ambient air and In fish. Average and realistic worst case risks
have been calculated. All risks represent lifetime exposure and the upper
95% probability of contracting cancers. For typical (average) concentrations
of 2,3,7,8-TCDD equivalents found in residential soil (50 ppt) and in air
(0.09 pg/m3) combined cancer risks would range from 6 x 10"° to 6 x 10-5,
using the EPA models. With the CDC modified model for soils and the EPA risk
model for air, cancer risks would be 1 x 10~5. For the worst observed soil
and air levels measured 1n Midland, cancer risks would range from 1 x 10"5
to 2 x 10"4, with the EPA models. With the CDC modified soil model and
the EPA air model, the lifetime cancer risks would be 3 x 10-5. if fish
-------
-15-
consumption of ten, 1/2 pound meals 1n a lifetime are added to those risks
found from exposure to average levels of PCDDs and PCDFs, then risks Increase.
As indicated by a review of the lifetime cancer risks shown 1n Table H, the
risks from exposure to PCDDs and PCDFs, as well as for 2,3,7,8-TCDD alone,
in either soil or air, exceed a 10-6 risk considered by EPA under Superfund
to be unacceptable or a 10-5 r-fsic considered by CDC as unacceptable (23).
Although the EPA soil exposure-risk model indicates potentially lower life-
time cancer risks, the more conservative upper range risk should be given
preference 1n risk management decision making.
Tab!e H: Combined Lifetime Cancer Risks From Exposure
to PCDDs and PCDFs In Residential Soils, Ambient
Air and Fish in Midland, Michigan
Exposure
Scenario Soil Air Fish
1. Typical 50 ppt: lxlO"7 to 0.16pg/m3: 6x10-6 lOx: 3x10-5
(Average) 5x10-5 (EPA) life
7x10-6 (CDC)*
2. Worst 160 ppt: 2xl0-6to 0.29pg/m3: 1x10*5 R>: 2x10-3
Observed 2x10-* (EPA) Wk: 1x10-2
2x10-5 {CDC)2
Soil + Air
Typical: Soil + A1r » 6x10-6 ^o 6x10-5 (EPA)
(Average) 1x10-5 (cdc)
Worst Observed: Soil + Air = 1x10-5 to 2x10"^ (EPA)
3x10-5 (CDC)2
Soil + Air + F1sh
Typical: Soil (EPA) + Air + F1sh (lOx 1n lifetime)" 4x10-5 to 9xl0"5
(Average)
Soil (CDC) + A1r + F1sh (lOx in 11fet1me)= 4xl0'5
Worst Observed:
Soil (EPA) + Air + Fish (Once per month)= 2x10-3
Soil (EPA) + Air + Fish (Once per week)* 1x10-2
Soil (CX)2+ Air + F1sh (Once per month)* 2x10*3
Soil (CDC)2+ Air + Fish (Once per week)= 1x10-2
1 Using estimated values of some PCDDs and PCDFs expressed as 2,3,7,8-TCDD
equivalents; 2,3,7,8-TCDO cancer risks are approximately 45% less
2 CDC model as modified
-------
-16-
VI. Suggestive Adverse Health Data from Midland County Michigan
Rates of connective and soft tissue cancer among Midland women, have
been statistically significant for over two decades (11). Ho other
Michigan county and only 13 counties In the nation have had statistically
significant rates for two successive decades, 1950-1979, where there
have been two or more cases recorded per decade (12). Connective and
soft tissue cancers have been associated with exposure to 2,3,7,8-
TCDD (6). While exposure to 2,3,7,8-TCDD or other PCDDs and PCDFs
is yet to be determined in these cases, animal studies show that
females exposed to 2,3,7,8-TCDD are more sensitive than males (6).
Birth defects in Midland county have also been elevated compared to
other Michigan counties (13). In the early 1970s the rates of oral
facial and urogenital defects rose precipitously, then declined
(14,15). Such defects are consistent with those observed in animals
exposed to 2,3,7,8-TCDD (6,14). EPA's Health Review Division in 1979
conducted preliminary analyses of Midland birth defect data and
believed sufficient evidence existed to warrant a case control stuty
of these birth defects in cooperation with the Michigan Department
of Health and CDC, but the stuty was not Initiated (14). CDC's
analysis of Dow Chemical's reproductive study of wives of Midland
workers exposed to PCDD's revealed a rate of oral facial clefts
twice that expected (15,16).
This suggestive adverse health information from Midland County, while
not as yet demonstrating an association with exposure to dloxins and
dibenzofurans, deserves consideration in the risk mangement decisions
made concerning the city of Midland. Consideration should also be
given to the long duration of chemical manufacture 1n Midland and
exposure to potentially higher concentrations of PCDDs and PCDFs
historically, as well as to other carcinogens, especially for workers
and fish consumers.
VII. Recommendations
1. Consideration should be given to covering or removing soils on the Dow
Plant site and around its perimeter having 1 ppb or greater 2,3,7,8-TCDD
equivalents. This action would reduce the potential for off-site
dispersal, help decrease existing air levels 1n Midland, and reduce
worker exposure to contaminated soils.
2. As lifetime cancer risks exceed 10~5 or 10"®, using several models,
strong consideration should be given to reducing risks from exposure
to PCDDs and PCDFs found in residential soils as well as in ambient air.
Implementing additional air monitoring for PCDDs and PCDFs and other
toxic substances and assessing the need for additional controls is
also recommended.
3. Estimates of risks from historic air levels of PCDDs and PCDFs, as
well as other toxic substances, should be made using appropriate
modeling and air data. Information on past releases of PCDDs and
PCDFs and other toxic substances from the Dow facility should also
be gathered.
-------
-17-
4. A formal, written risk management decision document should be developed
using data gathered from the Midland field study which includes a com-
parison with previous actions of the Agency concerning 2,3,7,8-TCDD
contamination and exposure. A comparison should also be made with non-
dloxin contamination situations addressed under Superfund due to apparent
inconsistences with the treatment of dloxln contamination and associated
risks in relationship to cases of contamination caused by other chemicals.
5. Preliminary health studies should be considered for Midland residents at
high potential health risk. Ideally, these would be Individuals living
(1) near and immediately downwind of the Dow facility for over 20 years
and with average or higher than average soil concentrations of PCDDs and
PCDFs (2) employees and their families of Dow Chemical exposed to PCDDs
as identified in the Townsend(Dow) reproductive study and (31 frequent
consuners of Tittabawassee River fish. The incidence rates of disease,
reproductive outcomes, and blood chemistry in these groups should be
conpared to matched groups outside of Midland. Smith and Thompson have
recently identified and interviewed 128 fishermen, including subsistence
fishermen, of the T1ttabawassee and have preliminary data regarding fish
consumption (20). This data base should be reviewed to assess the
sultabllty and sucess of an epidemiological study.
Standard epidemiological approaches to gather and evaluate such information
have been developed and should not be difficult to implement in this
situation. To assess reproductive outcomes of Dow workers, the Townsend
data set, should be re-evaluated and conpared with control populations
outside of Midland. The Office of Health and Environmental Assessment of
EPA possesses a computer printout of the Townsend study (15). In the high
exposure groups identified above, fat and human milk samples could also
be collected for residents of 20 years or more to estimate PCOD and PCDF
body burdens. A unique opportunity exists in Midland for assessing
hunan health impacts from long term exposure to PCDDs, particularly from
fish consunption. Performing such studies would serve not only 1n the
assessment of public health In Midland but would contribute further to
the understanding of health Impacts from PCDDs and PCDFs.
6. The use of the lppb 2,3,7,8-TCDD safe/unsafe action level should be
abandonded and substituted with risk levels associated with a given degree
of contamination and exposure as is done with other hazardous substances.
7. The EPA risk model for 2,3,7,8-TCDD, given its employment of exposure
uncertainties would be the preferred model to assess cancer risks.
Based upon the Memorandum of Understanding signed between EPA and CDC
in May, 1985, EPA now has the responslbility for risk assessements.
Therefore, 1t would be appropriate to apply those risk models developed
by EPA for the assessment of risks from exposure to PCDDs and PCDFs.
-------
-18"
References
1. U.S. EPA, Region V "Sot! Screening Survey at Four Midwestern Sites",
September 22, 1983.
2. Dow Chemical, "Point Sources and Environmental Levels of 2,3,7,8-TCDD on
Midland Plant Site of the Dow Chemical Conpany and In the City of Midland,
November 5, 1984.
3. BelUn, J. and D. Barnes, "Health Hazard Assessment for Chlorinated
Dloxins and - Dlbenzofurans other than 2,3,7,8-TCDD, Draft, U.S. EPA,
September 26, 1984.
4. Kurent, E "Enforcement Concerns In the Process and Outcomes of Clean-up
Levels for Dloxln In Soil", memorandum toC. Kleveno, U.S. EPA,
September 17, 1982.
5. U.S. EPA, "Outline For Region 7 Exposure/Risk Assessment, "Draft #1,
September 28, 1982.
6. U.S. EPA, Ambient Water Quality Criteria For 2,3,7,8-TCCD, February 1984.
7. Centers for Disease Control, "Health-Risk Estimates for 2,3,7,8-Tetra
chlorodibenzo dloxln in Soil," Morbidity and Mortality Weekly Report,
Vol. 33, No. 3, January 27, 1984.
8. Kimbrough, R.D, et al., "Health Implementations of 2,3,7,8-Tetrachloro-
dibenzodioxln (TCDD) contamination of Residential Soil", Centers For
Disease Control, July 1984.
9. Wedeen, R. P. et al., "Geophaglc Lead Nephropathy"; Case Report,
Environmental Research, Vol. 17, pg. 409, 1978.
10. Poiger, H. and C. Schlatter, "Influence of Solvents and Absorbance on
Dermal and Intestinal Absorption of TCDD'', Food Cosmet. Toxicol., Vol. 18,
pg. 477, 1981 .
11. Mlchifan Department of Public Health, "Evaluation of Soft and Connective
Tissue Cancer Mortality Rates 1n Midland and other Selected Michigan
Counties Compared Nationally and Statewide," May 4, 1983.
12. Riggan, W. et al., U.S. Cancer Mortality Rates and Trends, 1950-1979,
Volume II, EPA-600/1-83-015b, September 1983.
13. Harger, J.R.E., "First Cut Analysis of Birth Defect Data", memorandum to
D. Isleib, Toxic Substance Control Commission, State of Michigan,
October 29, 1980.
14. Poole, C., "Case-control study of Birth Defects in Midland County,
Michigan", Health Review Division, U.S. EPA, October 4, 1979.
-------
-19-
15. EPA Oversight on Dioxln Contamination, Hearing before the Subcommittee
on Natural Resources, Agricultural Research and Environment of the
Committee on Science and Technology U.S. House of Representatives, First
Session, Number 88, March 23, 1983, U.S. Government Printing Office,
29-9560, Washington, 1984.
16. Townsend, et al., "Survey of Reproductive Events of Wives of Employees
Exposed to Chorlnated Dioxins", Amer. Jour, of Epidem., Vol. 115, No. 5,
pg. 695, 1982.
17. Cook, M. "TCDD Emissions for Municipal Waste Combustors", memorandum,
U.S. EPA, December 16, 1983.
18. U.S. EPA, Region V, "A Report on Polychlorinated D1benzo-p-D1ox1ns (PCDDs)
and Polychlorinated Dlbenzofurans (PCDFs): A Summary of Studies conducted
in the Great Lakes Area, July 1981.
19. National Clean Air Coalition," Major Industrial Sources of Potential Toxic
Air Pollutants", April 16, 1982.
20. Smith. F. and W. Thompson, "Fisherman of the Tittabawassee", Environment,
Vol. 26, No. 5, 1984.
21. U.S. EPA, "Risk Analysis of TCOD Contaminated Soil", Office of Health and
Environmental Assessment, Office of Research and Development, November, 1984,
EPA-600/8-84-031.
22. Teoh, R., Communication to A. Wilson, Michigan Department of Natural Resources,
July 19, 1984.
23. Testimony by V. Houk and L. Thomas before the Public Works and Transportation
Subcommittee on Investigations and Oversight, on November 9, 1983 and November
15, respectively, Environmental Reporter, November 18, 1983.
24. Cocucci, S. et. al, "Absorption and Translocation of Tetrachlorodibenzo-p-
dioxin by Plants from Polluted Soil", Experientia, Vol. 35, No. 4, pg. 482, 1979.
25. W1ph, H., "TCDD-Levels 1n Soil and Plant Samples from the Seveso Area",
Pergamon Series on Environmental Science, Vol.5, Oxford Press, pg. 115., 1982.
-------
APPENOIX 0
Table 1
2378-TCDO
Dow Chemical - Midland Plant
In-Pi ant Surface Soil Samples
Si pie Field 2378-TCDD (DL)
er
Identification
Location
(ppb)
% Recovery
% Solids
13404
Station
1
South of 492 Building
0.018
(0.003)
71
92.9
14176
Station
2
South of 1005 Building; Southwest of 703 Building
0.074
(0.005)
52
96.2
14190
Station
3
South of 703 Building
0.42
(0.013)
90
96.1
14192
Station
4
Uest of 703 Building
0.020
(0.003)
63
95.0
13406
Station
5
Southwest of 956 Building; East of 703 Building
3.50
(0.039)
100
96.4
14180
Station
6
Northwest of 1159 Building; North of Shot Pond
0.13
(0.013)
103
98.5
14178
Station
7
11th and J Streets - Northwest Corner
4.60
(0.083)
101
96.3
14193
Station
8
8th and G Streets - Northwest Corner at Steam Pipeline
0.15
(0.010)
65
99.1
14194
Station
9
Northwest of 1050 Building at F Street
0.010
(0.003)
61
90.2
13405
Station
10
South of 543 Building; Uest of 14th Street
0.045
(0.005)
80
99.3
14187
Station
11
16th and G Streets - Southwest Corner
0.44
(0.016)
119
99.6
14182
Station
12
16th and G Streets - Northwest Corner
0.46
(0.012)
94
96.0
13407
Station
13
17th and G Streets - Northwest Corner
0.22
(0.007)
51
99.3
13412
Station
14
Vest of 934 Building
0.27
(0.007)
100
100.0
13413
Station
15
South and East of 874 Building North of RR Tracks
25.0
(0.50)
66
85.5
[36.0JR
(0.28)
Notes: (1) 2378-TCDO concentrations and detection levels (DL) reported In parts per
billion (ppb).
(2) * Recovery - Recovery of Internal standard (CI37 2378-TCDD or 13C 2378-TC0D)
expressed as percent.
(3) % Solids - Solids content of sample determined after sample homogenfzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) [ JR ¦ Repeat analysis of same sample.
-------
APPENDIX 0
Table 2
2378-TCOO
SITE #1 - Mfdland, Michigan Area
Surface Soil Samples
Sampie Field 2378-TCOD (01)
Number Identification (ppb) 1 Recovery 1 Solids
Upwind Areas
13354
UPJM-L
NO
(0.002)
821
94.91
13343
UPW-1-1
0.006
(0.002)
761
98.61
13401
UPH-2-L
NO
(0.004)
901
94.01
13395
UPW-4-L
NO
(0.004)
841
98.11
13342
UPW-4-D
0.009
(0.001)
841
84.21
Track Out and Perimeter Samples -
Dow Chemical
-Midland Plant
13353
T0-4-G
0.011
(0.003)
521
99.31
13360
T0-6-S
0.25
(0.018)*
211
99.01
13367
T0-9-G
0.014
(0.002)
681
99.01
14188
PER-2-L
0.31
(0.014)
841
B2.81
14177
PER-2-1
0.069
(0.003)
551
99.01
14191
PER-5-L
0.21
(0.008)
621
99.21
13402
PER-8-L
0.010
(0.005)
1041
99.01
13389
PER-9-G
2.03
(0.042)
721
95.61
14181
PER-10-L
0.040
(0.002)
731
97.11
Public Use Areas
13362
P-l-L
0.019
(0.002)
641
99.51
13364
P-2-L
0.028
(0.002)
641
89.71
13374
P-5-L
0.003
(0.001)
1041
99.71
13392
P-6-L
0.01S
(0.003)
861
80.21
13393
P-7-L
0.078
(0.003)
881
98.81
13340
P-8-L
0.17
(0.006)
941
96.51
13375
P-9-L
0.076
(0.003)
581
89.21
13391
P-10-L
0.012
(0.003)
921
96.61
13394
P-ll-L
0.108 (0.002)
811 100.0%
-------
APPENDIX D
Table 2 (continued)
2378-TCDD
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Sample
Field
2378-TCD0 (01)
Number
Identification
(ppb)
% Recovery
% Solids
Residential Areas
13007
A-l-L
0.075
(0.006)
1021
83.0%
13008
A-l-1
0.090
(0.008)
52%
89.7%
13101
A-3-L
0.009
(0.002)
1021
88.5%
13102
A-3-1
0.112
(0.008)
921
76.5%
13328
B-l-L
0.076
(0.007)
1001
93.2%
13305
B-l-1
0.16
(0.006)
63%
97.0%
13306
B-3-L
0.020
(0.001)
71%
97.8%
13325
B-3-1
0.27
(0.013)
74%
87.1%
13103
B-4-L
0.019
(0.002)
88%
84.7%
13104
B-4-1
0.028
(0.004)
54%
88.1%
13317
C-l-L
0.026
(0.001)
86%
78.7%
13303
C-l-1
0.054
(0.003)
64%
84.4%
13314
C-3-L
0.012
(0.001)
100%
98.6%
13307
C-3-1
0.24
(0.015)
71%
98.1%
13105
C-4-L
0.024
(0.002)
86%
91.0%
13106
C-4-1
0.032
(0.005)
88%
88.9%
13318
0-1-L
NO
(0.001)
94%
84.0%
13331
D-l-D
0.024
(0.001)
100%
99.5%
13319
D-2-6
0.028
(0.001)
92%
94.9%
13316
D-3-L
0.018
(0.001)
100%
90.9%
13329
0-3-1
0.031
(0.004)
86%
99.8%
13312
E-l-L
0.026
(0.003)
96%
97.5%
13330
E-l-1
0.049
(0.001)
80%
97.0%
13301
E-3-L
0.009
(0.001)
75%
97.6%
13304
E-3-1
0.020
(0.001)
70%
99.1%
13302
F-l-L
0.013
(0.001)
73%
99.0%
13313
F—1-1
0.013
(0.001)
92%
96.8%
M1 seel1aneous
14175 Sludge
0.021 (0.008)* 491
88.72
-------
APPENDIX D
Table 2 (continued)
2378-TCDD
SITE #1 - Midland, Michigan Area
Surface Soil Samples
Notes: (1) 2378-TCDD concentrations and detection levels (DL) reported 1n parts per
billion (ppb).
(2) * Recovery - Recovery of Internal standard (C1^7 2378-TCDD or 2378-TCDD)
expressed as percent.
(3) 5 Solids - Solids content of sample determined after sample homogenlzatlon,.
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field Identification of samples:
Location
Type
TO - Track Out
PER - Perimeter
P - Public Use
A-K - Residential
UPW • Upwind
L - Yard, lawn, or open area composite
1 or D - Downspout or drlpHne composite
G - Open area grab sample
* Data not valid. Quality assurance objective not achieved.
-------
APPENDIX 0
Table 4
2378-TCDO
SITE #3 - Middletown, Ohio Area
Surface Soil Samples
Sample
Field
2378-TCDD (OL)
Number
Identification
(ppb)
I Recovery
1 Solid
Upwind Area - Hamilton,
Ohio
13400
UPW-5-1
0.004 (0.002)
98%
79.3%
13377
UPW-5-L
0.003 (0.001)
1061
79.9%
Perimeter - ARMCO, Inc.
, Middletown Plant
13366
PER-12-L
NO
(0.002)
70%
87.4%
13337
PER-12-1
ND
(0.001)
881
88.4%
13387
PER-13-L
0.004
(0.003)
942
86.4%
13363
PER-13-D
ND
(0.002)
76%
87.4%
13372
PER-14-L
ND
(0.002)
106?
84.9%
13376
PER-15-L
ND
(0.002)
76%
88.4%
Public Use Areas
13349
6-1 -L
ND
(0.001)
94%
77.5%
13386
P-ll-L
ND
(0.002)
72%
87.5%
13327
P-12-L
0.004
(0.001)
82%
87.5%
13338
P-13-L
NO
(0.001)
96%
83.2%
13339
P-14-L
ND
(0.001)
96%
82.0%
13390
P-15-L
ND
(0.002)
106%
78.8%
13352
P-15-1
ND
(0.002)
82%
67.3%
Residential Areas
13335
G-2-L
0.004
(0.001)
92%
87.7%
13368
6-2-1
0.005
(0.002)
84%
86.7%
13353
H-l-L
ND
(0.002)
78%
95.9%
13403
H—1-1
ND
(0.002)
70%
94.7%
13261
K-l-L
ND
(0.002)
86%
81.3%
13351
K-l-1
ND
(0.003)
54%
88.3%
13350
K-2-L
ND
(0.002)
86%
86.4%
Miscellaneous
14189
SIudge
ND
(0.031)*
6%
92.3%
-------
APPENDIX 0
Table 3
2378-TCDD
SITE #2 - Henry, Illinois Area
Surface Sofl Samples
Sample Field 2378-TCDO (DL)
Number Identification (ppb) 1 Recovery 1 Solids
Upwind Area - Henry, Illinois
12901 UPW-7-L ND (0.001) 100% 89.51
12902 UPW-7-1 NO (0.003 ) 58% 88.11
Perimeter Samples
13002 PER-16 ND
13003 PER-17 ND
13004 PER-18 ND
13012 PER-18[D] ND
(0.0011 901 88.01
(0.002) 701 98.71
(0.001) 881 90.81
(0.001) 801 90.81
Residential Areas
12903 M-l-DL 0.002
12904 M-l-L ND
12905 N-l-L ND
12907 N-l-UD] ND
12906 N-l-1 ND
13001 P-l-L NO
12908 P-l-1 ND
(0.001) 1121 90.61
(0.001) 1001 92.71
(0.001) 1041 87.51
(0.001) 981 87.31
(0.002) 621 83.51
(0.001) 921 90.11
(0.001) 941 88.61
-------
APPENDIX D
Table 4 (continued)
2378-TCDD
SITE #3 - Middletown, Ohio Area
Surface Soil Samples
Notes: (1) 2375-TCDO concentrations and detection levels (OL) reported fn parts per
billion (ppb).
(2) I Recovery - Recovery of Internal standard (CI3' 2378-TCDD or *3C 2378-TCDO)
expressed as percent.
(3) * Solids - Solids content of sample determined after sample homogenlzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field identification of samples:
Location
Type
TO - Track Out
PER - Perimeter
P - Public Use
A-K - Residential
UPW - Upwind
L - Yard, lawn, or open area composite
I or D - Downspout or dripHne composite
G - Open area grab sample
* Data not valid. Quality assurance objective not achieved.
-------
APPENDIX 0
Table 5
2378-TCDO
SITE #4 • Minnesota Natural Areas
Surface Soil Samples
Sample
Number
Field
Identification
Location
2378-TCDD (DL)
(ppb)
I Recovery
1 Solids
13373
IWD-l-L
Itasca Wilderness
ND (0.003)
641
94.7%
13378
8SP-4-L
Bluesten Prairie
ND (0.001)
981
98.01
13379
KR-l-L
Kettle River
ND (0.002)
106%
79.5%
13388
PT-l-L
Pembina Trail Preserve
ND (0.002)
80S
53.5%
Notes: (1) 237B-TCDD concentrations and detection levels (DL) reported In parts per
billion (ppb).
(2) % Recovery - Recovery of Internal standard {CI37 2378-TCDD or 13c 2378-TCDO)
expressed as percent.
(3) % Solids - Sol Ids content of sample determined after sample homogenlzatlon,
expressed as percent. Analytical results not adjusted for moisture content.
(4) Field Identification of samples:
Location Type
UPW - Upwind L - Yard, lawn, or open area composite
TO - Track Out 1 or D - Downspout or drlpllne composite
PER - Perimeter G - Open area grab sample
P - Public Use
A-K - Residential
-------
Sample
Number
14196
14184
13368
13308
13320
13333
13344
13344
133S7
13380
13396
13396
13409
13385
14183
14195
13309
13321
13332
13345
13356
13369
13381
13397
13408
13414
13415
13416
13417
13384
Notes:
APPENDIX D
Table 6
2378-TCDD
Control and Blank Samples
Field
Identification
2378-TCDD (DL)
% Recovery
% Solids
Control Soil
ND (0.006)*
461
98.4%
Control Soil
0.013 (0.004)
591
98.4%
Control Soil
0.006 (0.002)
661
98.4%
Control Soil
0.032 (0.003)
58%
98.4%
Control Soil
0.012 (0.003)
50%
98.4%
Control Soil
ND (0.010)*
40%
98.4%
Control Soil
ND (0.007)*
34%
98.4%
Control Soil
0.015 (0.005)*
22%
98.4%
Control Soil
ND (0.001)*
23%
98.4%
Control Soil
0.005 (0.003)
66%
98.4%
Control Soil
ND (0.019)*
10%
98.4%
Control Soil
ND (0.007)*
19%
98.4%
Control Soil
ND (0.054)*
22%
98.4%
Control Soil
ND (0.042)*
13%
98.4%
Blank Soil
ND (0.002)
97%
53.7%
Blank Soil
ND (0.003)
81%
53.7%
Blank Soil
ND (0.001)
83%
53.7%
Blank Soil
0.002 (0.001)
80%
53.7%
Blank Soil
ND (0.002)
73%
53.7%
Blank Soil
ND (0.002)
96%
53.7%
Blank Soil
ND (0.001)
100%
53.7%
Blank Soil
ND (0.002)
70%
53.7%
Blank Soil
ND (0.001)
100%
NR
Blank Soil
ND (0.001)
106%
53.7%
Blank Soil
NO (0.001)
84%
53.7%
Can Blank (PER-5)
ND (0.027)
102%
Can Blank (In-Piant) ND (0.024)
100%
Can Blank
ND (0.025)
96%
(Mlddletown)
Can Blank #6
ND (0.030)
80%
Blank Soil
ND (0.003)
103%
53.7%
(1) 2378-TCDO concentrations and detection levels (DL) reported 1n
parts per billion (ppb).
(2) % Recovery - Recovery of Internal standard (CI37 2378-TCDO or
13c 2378-TCDO) expressed as percent.
(3) % Solids - Solids content of sample determined after sample
homogenlzatlon, expressed as percent. Analytical results not
adjusted for moisture content.
(4) NR - Not reported.
* Data not valid. Quality assurance objective not achieved.
See Quality Assurance Sixmary, Appendix C.
-------
APPENDIX D
Table 7
PCDDs and PCDFs
SITE #1 - Midland. Michigan Area
Surface Soil Samples
Upwind
Dow Chemical
In-Piant
Sample No.:
Field ID.:
13401
UPW-2-L
13395
UPW-4-L
13406
Station 5
Location:
Pleasant
View School
4853 W. Kent
Incinerator
PCDDs (DL)
2378-TCDD
Total Iso TCDDs
Total penta CDOs
Total hexa CDOs
Total hepta CDOs
OCDD
NO (0.004)
ND (0.004)
ND (0.024)
NO (0.024)
0.15 (0.024)
0.34 (0.026)
ND (0.004)
ND (0.023)
ND (0.023)
0.17 (0.034)
0.33 (0.034)
3.5 (0.039)
PCDFs (DL)
2378-TCOF
Total TCDFs
Total penta CDFs
Total hexa CDFs
Total hepta CDFs
OCDF
ND (0.004)
ND (0.008)
ND (0.022)
ND (0.031)
ND (0.051)
ND (0.004)
ND (0.008)
NO (0.023)
ND (0.028)
ND (0.045)
0.45 (0.06)
13412
Station 14
0.27
0.32
0.24
4.0
75.0
375.0
(0.007)
(0.067)
(0.067)
(0.9)
(1.3)
0.027 (0.007)
0.90
3.1
15.4
8.6
(0.14)
(0.13)
(0.38)
(0.48)
Notes: (1) Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported 1n parts per billion (ppb).
-------
APPENDIX 0
Table 7
PCDOs and PCOFs
SITE 11 - Midland. Michigan Area
Surface Soil Samples
Public Use Areas
Sanple No.:
13374
13392
13393
13375
13391
13394
Field ID.:
P-5-L
P-
-6-L
P-7-L
P-
-9-L
P-]
10-L
P-
11-L
Central School
Location:
County Line Rd.
Mapleton School
Longview School
Virginia Park
(ball
diamond)
Bullock School
PCDDs (DL)
2378-TCD0
0.003 (0.001)
0.015
(0.003)
0.078 (0.003)
0.076
(0.003)
0.012
(0.003)
0.11
(0.002)
Total Iso TCDDs
ND (0.001)
0.040
0.17
0.29
0.040
0.22
Total penta CDDs
ND (0.014)
ND
(0.035)
Interference
0.10
(0.018)
ND
(0.034)
0.12
(0.022)
Total hexa CDOs
0.067 (0.007)
0.063
(0.035)
0.34 (0.02)
0.24
(0.018)
0.086
(0.034)
0.41
(0.022)
Total hepta CDDs
0.35 (0.013)
0.38
(0.028)
2.3 (0.055)
0.41
(0.093)
0.35
(0.031)
2.4
(0.042)
OCDD
3.1 (0.096)
0.86
(0.027)
7.0 (0.068)
12.0
(1.5)
0.68
(0.031)
7.0
(0.052)
PCDFs (DL)
2378-TCDF
ND (0.002)
ND
(0.005)
0.013 (0.007)
0.013
(0.002)
ND
(0.005)
0.015
(0.003)
Total TCDFs
ND (0.002)
Total penta CDFs
ND (0.01)
ND
(0.008)
ND (0.025)
0.040
(0.01)
NO
(0.007)
0.11
(0.017)
Total hexa CDFs
ND (0.01)
ND
(0.024)
0.26 (0.036)
0.064
(0.01)
ND
(0.029)
0.17
(0.037)
Total hepta CDFs
0.065 (0.02)
0.14
(0.043)
0.72 (0.021)
0.50
(0.034)
0.16
(0.062)
0.82
(0.045)
OCDF
0.044 (0.023)
0.10
(0.071)
0.64 (0.037)
0.37
(0.049)
0.11
(0.070)
0.66
(0.045)
Notes: (1) Concentrations of PCDDs, PCOFs, and detection
levels (01) reported In parts per billion (ppb).
-------
APPENDIX D
Table 8
PCDDs and PCDFs
SITE #3 - Middletown, Ohio Area
Surface Soil Samples
Upwind
Perimeter
Sample Ho.:
Field ID*:
Location:
PCDDs (PL)
2378-TC00
Total Iso TCDDs
Total penta CDDs
Total hexa CODs
Total hepta CDDs
OCDD
13377
UPW-5-L
13400
UPW-5-L
13387
PER-13-L
13372
PER-14-L
13376
PER-1S-L
S130 Princeton 5130 Princeton 2620 Packaging Perimeter 14 - West Oxford Road
0.003 (0.001) 0.004 f0.002) 0.004 (0.003)
ND (0.001) ND (0.002) ND (0.003)
NO (0.010) ND (0.044) NO (0.020)
NO (0.010) 0.072 (0.044) NO (0.020)
0.023 (0.007) 0.20 (0.17) 0.15 (0.014)
.<0.20 (0.009) 10.6 (0.23 ) 0.28 (0.014)
ND
ND
0.073
0.17
(0.002)
(0.002)
ND (0.014)
NO (0.009)
(0.018)
(0.026)
NO
ND
ND
ND
0.12
0.B4
(0.002)
(0.002)
(0.039)
(0.039)
(0.009)
(0.071)
PCDFs (DL)
2378-TC0F
0.002
(0.001)
ND
(0.004)
0.006
(0.004)
ND
(0.001)
ND
Total TCDFs
Total penta COFs
ND
(0.008)
ND
(0.016)
NO
(0.010)
ND
(0.012)
ND
Total hexa CDFs
ND
(0.008)
ND
(0.036)
ND
(0.028)
ND
(0.012)
ND
Total hepta CDFs
ND
(0.018)
ND
(0.023)
ND
(0.026)
ND
(0.017)
0.043
OCDF
ND
(0.026)
ND
(0.056)
ND
(0.046)
ND
(0.024)
0.050
ND (0.005)
Notes: (1) Concentrations of PCODs, PCDFs, and detection
levels (DL) reported In parts per billion (ppb).
-------
APPENDIX 0
Table 9
PCDDs and PCDFs
SITE #4 - Minnesota Natural Areas
Surface Soil Samples
Sample No.:
Field 10.:
.Location:
PCPDs (PL)
2378-TCDO
Total fso TCDDs
Total penta CDDs
Total hexa CDDs
Total hepta CDDs
OCDD
13373
IWD-l-L
Itasca Wilderness
ND (0.0031
MD 10.003)
ND (0.010)
NO (0.005)
0.047 (0.009)
0.13 (0.019)
13378
BSP-4-L
Dluestern Prairie
ND
ND
ND
ND
0.091
0.20
(0.001)
(0.001)
(0.012)
(0.012)
(0.010)
(0.012)
13379
KR-l-L
Kettle River
ND
ND
ND
ND
0.02S
0.092
(0.002)
(0.002)
(0.004)
(0.011)
(0.009)
(0.011)
PCDFs (PL)
2370-TCDF ND (0.002) NO (0.002) NO (0.001)
Total TCDFs ND (0.002)
Total penta COFs ND (0.010) ND (0.009) ND (0.013)
Total hexa COFs ND (0.010) NO (0.009) NO (0.013)
Total hepta CDFs ND (0.017) ND (0.009) ND (0.009)
OCDF NO (0.024) ND (0.010) NO (0.014)
Notes: (1) Concentrations of PCDDs, PCDFs, and detection
levels (DL) reported 1n parts per billion (ppb).
-------
APPENDIX D
Table 10
PCOOs and PCOFs
Control and Blank Samples
Blank and Control Samples
Sample No.:
13380
13381
13414
13415
13416
13417
Field 10.:
Control Soil
Blank Soil
Can
Blank
Can
Blank
Can Blank
Can
B1 ank
Location:
PER-5
In-Pi ant Blank
Blank Mlddletown
Blank #6
PCDDs (DL)
2378-TCDD
0.005
(0.003)
ND
(0.001)
ND
(0.027)
ND
(0.024)
ND
(0.025)
ND
(0.03)
Total Iso TCDDs
ND
(0.003)
ND
(0.001)
ND
(0.027)
ND
(0.024)
NO
(0.025)
NO
(0.03)
Total penta CDDs
0.081
(0.033)
ND
(0.015)
ND
(0.097)
ND
(0.10)
ND
(0.098)
ND
(0.11)
Total hexa CDDs
0.64
(0.033)
ND
(0.015)
ND
(0.097)
ND
(0.10)
ND
(0.098)
ND
(0.11)
Total hepta CDDs
NA
0.031
(0.005)
ND
(0.17)
ND
(0.15)
ND
(0.17)
ND
(0.19)
OCDD
NA
0.27
(0.007)
0.63
(0.22)
0.74
(0.19)
0.89
(0.23)
0.44
(0.24)
PCDFs (DL)
2378-TCDF
0.017
(0.005)
ND
(0.001)
NO
(0.039)
ND
(0.022)
ND
(0.023)
ND
(0.027)
Total TCOFs
ND
(0.039)
NO
(0.022)
ND
(0.023)
NO
(0.027)
Total penta CDFs
NA
ND
(0.011)
ND
(0.021)
ND
(0.032)
ND
(0.14)
NO
(0.074)
Total hexa CDFs
NA
ND
(0.011)
ND
(0.35)
ND
(0.14)
ND
(0.13)
NO
(0.12)
Total hepta CDFs
NA
ND
(0.012)
NA
ND
(0.15)
ND
(0.19)
ND
(0.16)
OCDF
NA
ND
(0.013)
NA
ND
(0.15)
ND
(0.19)
ND
(0.16)
Notes: (1) Concentrations of PCOOs, PCDFs, and detection
levels (DL) reported 1n parts per billion (ppb).
(2) NA - Not analyzed. Peak broadening In GC/HS
system prevented quantitation.
-------
Sample Identification Fish/
EPA HONR Species Sample
St. Joseph River at Berrln Springs
13216 SJR-1W Carp 5
Huron River at Flatrock
13281 HR-2W Carp 6
River Raisin at Monroe
13282 RR-4U Carp 3
Clinton River at Ht. Clemens
13283 CR-6U Carp 1
Kalamatoo River at Alleqtn
13284 KR-4W Carp 5
Pine River at Alma
13286 PR-IW Carp 5
Muskegon River at Rogers Dam
13287 UMR-1W Carp 5
St. Clair River at Algonac
13288 SCR-1W Carp 5
T/UJLE U
MICHIGAN 0I0XIN STUDY
1983 FISH COLLECTION
RESULTS FOR 2,3,7,8-TCOO
Lipid
Sample Length Weight Content 2,3,7,8-TCOO
Type (cm) (gm) .(*) , PP*
-------
1983 FISH COLLECTION
RESULTS FOR 2,3,7,8-TCDD
(continued)
Sample Identification
EPA HDNR Species
Ausable
River at M1o
13289
AR-IH
Carp
Shiawassee River at Chesanlng
13290
SR-iW
Carp
Muskegon Lake at Muskegon
13420
ML-2A
Carp
13422
ML-5W
Pike
13421
ML-5A
Pike
Grand River at Grand Ledge
13245
GR-1W
Carp
13219
6R-1
Carp
13220
GR-2
Carp
13221
GR-3
Carp
13222
Gft-4
Carp
13223
GR-5
Carp
13224
GR-6
Carp
13225
GR-7
Carp
13226
GR-8
Carp
13227
GR-9
Carp
13228
GR-10
Carp
13229
GR-11
Carp
13230
GR-12
Carp
13231
GR-13
Carp
13232
GR-14
Carp
13233
GR-15
Carp
Lipid
Ffsh/ Sample Length Weight Content 2,3,7,8-TCDO
Sample Type (cin) (gro) (*) PPt (PL)
WHL
57-72,5
2680-5370
NO (<0.8)
WHL
52-68
2360-4840
NO (1.2)
8
3
2
SF-R
WHL
SF-R
54-66
57-63
55-76
2040-4420
1140-1720
920-3080
8.1
5.3
5.2 (2.0)
3.9 (1.6)
HO (2.4)
WHL
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
SF-L
34.5-J9.5
30
31
31.5
31.5
27
32
29
29
40
40
42.5
41
38.5
36.5
37
680-1040
400
420
510
540
300
S10
380
410
1010
1020
1140
1140
880
77 Q
830
5.4
1.9
2.9
1.9
2.2
3.7
1.6
0.6
6.0 (0.6)
NO (2.5)
6.9 (2.5)
NO
NO
NO
NO
NO
(2.5)
(2.5)
(2.5)
(2.5)
(2.5)
ND (<0.4)
3.7 (0.3)
(0.2)
(0.2)
3.5 (0.2)
1.5 (0.5)
ND {l.Z)
0.6 (0.3)
0.4
7.2
-------
1983 FISH COLLECTION
RESULTS FOR 2,3,7,8-TCDD
(continued)
Sample
EPA
Identification
MDNR
Species
Fish/
Sample
Sampl e
Type
Length
(cm)
Weight
(gm)
npia
Content
w
2,3,7,8-TCl
PPt (DL)
Tlttabawassee River at
Midland (con't)
13262
TR-15"
Carp
1
SF-L
41
1000
7.9
46
(0.6
13263
TR-16
Carp
1
SF-L
39.5
820
3.4
19
(0.7
13264
TR-17
Carp
1
SF-L
63,5
3300
8.0
34
(1.2
13265
TR-18
Carp
1
SF-L
64.5
3450
2.3
12
(1.6
13266
TR-19
Carp
1
SF-L
51.5
1820
4.0
24
(1.6
13267
TR-20
Carp
1
SF-L
48
1560
3.3
25
(1.4
13268
TR-21
Carp
1
SF-L
47
1500
4.6
90
(0.3
13269
TR-22
Carp
1
SF-L
44
1320
11
530
(1.6
13270
TR-23
Carp
1
SF-L
51.5
1680
3.6
41
(1.3
13271
TR-24
Carp
1
SF-L
49
1540
3.6
70
(1.4
13272
TR-25
Carp
1
SF-L
46.5
1380
1.4
17
(1.6
13274
TR-4A
Ch. Catfish
SF-R
45-55
800-1760
7.3
75
(0.4
13275
TR-6A
Sm. Bass
SOF-R
24-38
220-060
1.0
5.1
(0.2
13276
TR-66A
Walleye
1
SOF-R
56
1580
1.4
5.1
(0.2
13277
TR-67A
Walleye
1
SOF-R
52
1320
2.0
3.6
(0.4
13278
TR-68A
Walleye
1
SOF-R
49
1070
1.3
3.9
(0.2
13279
TR-69A
Walleye
1
SOF-R
39
570
0.6
2.8
(0.3
13280
TR-70A
Walleye
1
SOF-R
38
500
0.6
4.1
(0.3
Sample Types
SF-R - Skinless Fillet - Right Side
SF-L - Skinless Fillet - Left Side
SOF-R ¦ Skin on Fillet - Right Side
SOF-L - Skin on Fillet - Left Side
WHL - Whole Fish
DL «¦ Detection Limit
-------
1983 FISH COLLECTION
RESULTS FOR 2,3,7,8-TCOD
(continued)
Sample
EPA
Identification
MDNR
Species
Fish/
Sample
Sample
Type
Length
M
Grand River at Grand Ledge
(con't)
13234
GR-16
Carp
1
SF-L
36
13236
GR-17
Carp
1
SF-L
53
13237
GR-18
Carp
1
SF-L
45
132.18
GR-19
Carp
1
SF-L
57
13239
GR-20
Carp
1
SF-L
55
13240
GR-21
Car p
1
SF-L
54
13241
GR-22
Carp
1
SF-L
40
13242
GR-23
Carp
I
SF-L
45.5
13243
GR-24
Carp
1
SF-L
46
13244
GR-25
Carp
1
SF-L
35
13246
GR-5A
Bullhead
5
SF-R
24.5-29
13247
GR-6A
Bass
5
SOF-R
28-36
Tlttabawassee River at Midland
13273
7R-1W
Carp
5
UHL
34-56
13248
TR-1
Carp
1
SF-L
37
13249
TR-2
Carp
1
SF-L
39
132SO
TR-3
Carp
I
SF-L
17
13251
TR-4
Carp
SF-L
V
13252
TR-5
Carp
1
SF-L
42
13253
TR-6
Carp
1
SF-L
37
13254
TR-7
Carp
1
SF-L
35
13255
TR-8
Carp
1
SF-L
36,5
13256
TR-9
Carp
1
SF-L
43
13257
TR-10
Carp
I
SF-L
45
13258
TR-11
Carp
1
SF-L
41
13259
TR-12
Carp
1
SF-L
43
13260
TR-13
Carp
1
SF-L
41.5
13261
TR-14
Carp
1
SF-L
43
Lipid
Weight Content 2,3,7,8-TCDD
(qm) U) PPt (PL)
700
ND
(0.7)
2190
NO
(0.5)
1600
8.1
11
(0.9)
2940
1.2
1.8
(0.3)
2900
HD
(1.8)
2940
NO
(0.4)
1100
—
NO
2.7
1400
---
NO
(1.5
1540
1.3
1.2
(0.5
660
1.7
12
(0.3
240*340
—
ND (<0.2
280-690
ND (<0.2!
>00-2860
lot
190
(0.6)
700
4.5
38
(0.4)
840
1.7
18
(0.3)
780
2.7
28
(0.2)
860
5.3
27
(1.5)
860
3.5
16
(0.3)
720
1.2
16
(1.0)
560
1.8
12
(0.6)
640
0.9
18
(0.2)
1040
6.6
45
(0.3)
1240
1.2
38
(0.6)
880
3.8
13
(0.3)
1030
4.3
14
(0.6)
1060
6.4
26
(0.4)
1070
4.9
22
(0.6)
-------
MICHIGAN OIOX1N STUOf
Sttapl« Merit i flCitlOfl
EPA TO
Saginaw 8
-------
2378-TCDD
SUMMARY OF RECENT ENVIRONMENTAL MEASUREMENTS
MIDLAND, MICHIGAN AREA
Number
of
Measurements
Range
Ambient Air* (March 1983-February 1984)
Total - all measurements
Dow Plant
Dow Plant Fence Line
City of Midland
10 0.010-0.21 pg/m3
2 0.019-0.022 pg/in3
6 0.010-0.21 pg/m3
2 0.019-0.16 pg/m3
Arithmetic
Mean
0.061
0.021
0.064
0.090
Arithmetic
Standard Geometric
Deviation Mean
0.069
0.075
0.036
0.038
* Data provided by Dow Chemical Company
Tittabawassee River Fish (August-September 1983)
Carp - whole fish composite (5 fish)
Carp - individual skinless fillets
Catfish - skinless fillet composite (5 fish)
Smallmouth Bass - skin-on fillet composite (5 fish)
Walleye - individual skin-on fillets
1
190
PPt
25
12-530
PPt
53
102
28.6
1
75
PPt
1
5.1
ppt
5
2.8-5.1
PPt
3.9
0.8
3.8
Surface Soils (October-December 1983)
Oow Plant
15
0.01-25.0
PPb
2.38
6.41
0.24
Dow Plant Perimeter
9
0.01- 2.03
PPb
0.74
1.43
0.17
Public Use and Residential Areas
17
0.003-0.17
ppb
0.037
0.042
0.026
Residential Areas*
8
0.009-0.076
ppb
0.025
0.022
0.020
Residential Downspouts*
0
0.0130.27
ppb
0.104
0.103
0.062
* Data for four additional sites available 12/15/84.
------- |