450585002
vvEPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/5-85-002
April 1985
Air
Inorganic Arsenic
Risk Assessment
For Primary and
Secondary Lead
Smelters, Primary
Zinc Smelters,
Oxide Plants,
Cotton Gins, and
Arsenic Chemical
Plants
-------
9
-------
NOTICE
Thisdocument has not been formally released by EPA and should not now be construed to represent Agency
policy. It is being circulated for comment on its technical accuracy and policy implications.
EPA-450/5-85-002
Inorganic Arsenic Risk Assessment for
Primary and Secondary Lead Smelters,
Primary Zinc Smelters, Zinc Oxide Plants,
Cotton Gins, and Arsenic Chemical Plants
Strategies and Air Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
April 1985
-------
This report has been reviewed by the Strategies and Air Standards Division of the Office of Air Quality
Planning and Standards, EPA, and approved for publication. Mention of trade names or commercial products
is not intended to constitute endorsement or recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental Protection Agency, Research Triangle
Park, N.C. 27711, or from National Technical Information Services, 5285 Port Royal Road Sprinqfield
Virginia 22161. ' a
-------
TABLE OF CONTENTS
Title Page
1 INTRODUCTION 1
1.1 Overview 1
1.2 The Relationship of Exposure to Cancer Risk 1
1.3 Public Exposure 4
1.4 Public Cancer Risks 5
2 THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC 6
2.1 The Linear No-Threshold Model for Estimation of
Unit Risk Based on Human Data (General) 6
2.2 Unit Risk Estimates Derived from Epidemiologic Studies . 9
3 QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC ARSENIC
EMISSIONS 13
3.1 EPA's Human Exposure Model (HEM) (General) 13
3.1.1 Pollutant Concentrations Near A Source 13
3.1.2 Expansion of Analysis Area 14
3.2 Methodology for Reviewing Pollutant Concentrations ... 15
3.2.1 Use of Ambient Data 19
3.2.2 The People Living Near A Source 19
3.2.3 Exposure 20
3.3 ASARCO-East Helena 22
3.3.1 Public Exposure to Inorganic Arsenic Emissions from
Primary Lead Smelters 24
3.3.1.1 Source Data 24
3.3.1.2 Exposure Data 24
3.4 Murph Metals-Dallas and Quemetco-Seattle 29
3.4.1 Public Exposure to Inorganic Arsenic Emissions from
Secondary Lead Smelters 32
3.4.1.1 Source Data 32
3.4.1.2 Exposure Data 32
3.5 Public Exposure to Inorganic Arsenic Emissions from
Primary Zinc Smelters 39
3.5.1 Source Data 39
3.5.2 Exposure Data 39
3.6 Public Exposure to Inorganic Arsenic Emissions from
Zinc Oxide Plants 45
3.6.1 Source Data 45
3.6.2 Exposure Data 45
3.7 Methodology for Reviewing Pollutant Concentrations -
Cotton Gins 51
3.7.1 Public Exposure to Inorganic Arsenic Emissions from
Cotton Gins 53
3.7.1.1 Source Data 53
3.7.1.2 Exposure Data 53
iii
-------
Title
3.8 Public Exposure to Inorganic Arsenic Emissions from
Arsenic Plants ..... 68
3.8.1 Source Data 68
3.8.2 Exposure Data 68
4 QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC
ARSENIC EMISSIONS 74
4.1 Methodology (General) 74
4.1.1 The Two Basic Types of Risk 74
4.1.2 The Calculation of Aggregate Risk 74
4.1.3 The Calculation of Individual Risk 76
4.2 Risks Calculated for Emissions of Inorganic Arsenic ... 76
5 ANALYTICAL UNCERTAINTIES APPLICABLE TO THE CALCULATION OF PUBLIC
HEALTH RISKS CONTAINED IN THIS DOCUMENT 85
5.2 Public Exposure 86
5.2.1 General 86
5.2.2 The Public 87
5.2.3 The Ambient Air Concentrations 88
6 REFERENCES 90
-------
LIST OF TABLES
Table
1 Summary of Quantitative Risk Analyses 10
2 Combined Unit Risk Estimates for Absolute Risk Linear Models. . 12
3 Arsenic Concentrations Near ASARCO-East Helena Primary
Lead Smelter 16
4 Identification of Primary Lead Smelters 25
5 Input Data to Exposure Model Primary Lead Smelting Industry
(Assuming Baseline Controls) 26
6 Total Exposure and Number of People Exposed Primary Lead
Smelting Industry 27
7 Public Exposure for Primary Lead Smelting Industry as
Produced by the Human Exposure Model (Assuming Baseline
Controls) 28
8 Arsenic Concentrations Near Select Secondary Lead Smelters . . 35
9 Identification of Secondary Lead Smelters 34
10 Secondary Lead Industry Inputs to HEM Model
(Assuming Baseline Controls) 35
11 Total Exposure and Number of People Exposed Secondary Lead
Smelting Industry 37
12 Public Exposure for Secondary Lead Smelters as Produced
by the Human Exposure Model (Assuming Baseline Controls) ... 38
13 Arsenic Concentrations Near Select Primary Zinc Smelters ... 40
14 Identification of Primary Zinc Smelters 41
15 Input Data to Exposure Model Primary Zinc Smelting Industry
(Assuming Baseline Controls) 42
16 Total Exposure and Number of People Exposed Primary
Zinc Smelter 43
17 Public Exposure for Primary Zinc Smelters as Produced by the
Human Exposure Model (Assuming Baseline Controls) 44
18 Arsenic Concentrations Near Select Zinc Oxide Plants 46
19 Identification of Zinc Oxide Plants 47
v
-------
Table Page
20 Input Data to Exposure Model Zinc Oxide Plants
(Assuming Baseline Controls) 48
21 Total Exposure and Number of People Exposed (Zinc Oxide
Plants) 49
22 Public Exposure for Zinc Oxide Plants as Produced by the
Human Exposure Model (Assuming Baseline Controls) 50
23 Arsenic Concentrations Near Two Texas Cotton Gins ...... 52
24 Identification of Model Cotton Gins ..... 54
25 Input Data to Exposure Model Cotton Gins (Assuming Baseline
Controls) . 55
26 Public Exposure for 4 Bales/Hour Model Cotton Gin (Hutto.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 56
27 Public Exposure for 7 Bales/Hour Model Cotton Gin (Hutto.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 57
28 Public Exposure for 12 Bales/Hour Model Cotton Gin (Hutto.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 58
29 Public Exposure for 20 Bales/Hour Model Cotton Gin (Hutto.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) 59
30 Public Exposure for 4 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) .... 60
31 Public Exposure for 7 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 61
32 Public Exposure for 12 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 62
33 Public Exposure for 20 Bales/Hour Model Cotton Gin (Buckholts,
TX) as Produced by the Human Exposure Model (Assuming
Baseline Controls) 63
vi
-------
34 Public Exposure for 4 Bales/Hour Model Cotton Gin (Itasca.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) .......................... 64
35 Public Exposure for 7 Bales/Hour Model Cotton Gin (Itasca.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) .......................... 65
36 Public Exposure for 12 Bales/Hour Model Cotton Gin (Itasca.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) .......................... 66
37 Public Exposure for 20 Bales/Hour Model Cotton Gin (Itasca.TX)
as Produced by the Human Exposure Model (Assuming Baseline
Controls) .......................... 67
38 Arsenic Concentrations Near Select Arsenic Chemical Plants . . 69
39 Identification of Arsenic Chemical Plants .......... 70
40 Input Data to Exposure Model Arsenic Chemical Plants
(Assuming Baseline Controls) ................ 71
41 Total Exposure and Number of People Exposed (Arsenic
Chemical Plants) ...................... 72
42 Public Exposure for Arsenic Chemical Plants as Produced by
the Human Exposure Model (Assuming Baseline Controls) ... 73
43 Maximum Lifetime Risk and Cancer Incidence for Primary Lead
Smelters (Assuming Baseline Controls) ............ 78
44 Maximum Lifetime Risk and Cancer Incidence for Secondary
Lead Smelters (Assuming Baseline Controls) ......... 79
45 Maximum Lifetime Risk and Cancer Incidence for Primary
Zinc Smelters (Assuming Baseline Controls) ......... 80
46 Maximum Lifetime Risk and Cancer Incidence for Zinc Oxide
Plants (Assuming Baseline Controls) ............ 81
47 Maximum Lifetime Risk and Cancer Incidence for Model Cotton
Gins (Assuming Baseline Controls) ............. 82
48 Lifetime Risk for Two Texas Cotton Gins (Assuming Baseline
Controls) .......................... 83
49 Maximum Lifetime Risk and Cancer Incidence for Arsenic
Chemical Plants (Assuming Baseline Controls) ........ 84
vii
-------
LIST OF FIGURES
Figure Page
1 Group 2 BG/ED Interpolation 17
2 Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (ASARCO-East Helena, MT) 23
3 Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (Murph Metals-Dallas, TX) 30
4 Predicted Versus Measured Inorganic Arsenic Ambient
Concentrations (Quemetco, Seattle, WA) 31
vm
-------
INORGANIC ARSENIC RISK ASSESSMENT FOR PRIMARY AND SECONDARY LEAD BELTERS,
PRIMARY ZINC SMELTERS AND ZINC OXIDE PLANTS, COTTON GINS AND ARSENIC CHEMICAL
PLANTS
1 INTRODUCTION
1.1 Overview
The quantitative expressions of public cancer risks presented in this
document are based on (1) a dose-response model that numerically relates
the degree of exposure to airborne inorganic arsenic to the risk of getting
lung cancer, and (2) numerical expressions of public exposure to ambient
air concentrations of inorganic arsenic estimated to be caused by emissions
from stationary sources. Each of these factors is discussed briefly below
and details are provided in the following sections of this document.
1.2 The Relationship of Exposure to Cancer Risk
The relationship of exposure to the risk of contracting lung cancer is
derived from epidemiological studies in occupational settings rather than
from studies of excess cancer incidence among the public. The epidemiological
methods that have successfully revealed associations between occupational
exposure and cancer for substances such as asbestos, benzene, vinyl chloride,
and ionizing radiation, as well as for inorganic arsenic, are not readily
applied to the public sector, with its increased number of confounding
variables, much more diverse and mobile exposed population, lack of consoli-
dated medical records, and almost total absence of historical exposure
data. Given such uncertainties, EPA considers it improbable that any
association, short of very large increases in cancer, can be verified in
the general population with any reasonable certainty by an epidemiological
study. Furthermore, as noted by the National Academy of Sciences (NAS)1,
"...when there is exposure to a material, we are not starting at an origin
-------
of zero cancers. Nor are we starting at an origin of zero carcinogenic
agents in our environment. Thus, it is likely that any carcinogenic agent
added to the environment will act by a particular mechanism on a particular
cell population that is already being acted on by the same mechanism to
induce cancers." In discussing experimental dose-response curves, the NAS
observed that most information on carcinogenesis is derived from studies of
ionizing radiation with experimental animals and with humans which indicate
a linear no-threshold dose-response relationship at low doses. They added
that although some evidence exists for thresholds "in some animal tissues,
by and large, thresholds have not been established for most tissues. NAS
concluded that establishing such low-dose thresholds "...would require
massive, expensive, and impractical experiments ..." and recognized that
the U.S. population "...is a large, diverse, and genetically heterogeneous
group exposed to a large variety of toxic agents." This fact, coupled with
the known genetic variability to carcinogenesis and the predisposition of
some individuals to some form of cancer, makes it extremely difficult, if
not impossible, to identify a threshold.
For these reasons, EPA has taken the position, shared by other Federal
regulatory agencies, that in the absence of sound scientific evidence to
the contrary, carcinogens should be considered to pose some cancer risk
at any exposure level. This no-threshold presumption is based on the view
that as little as one molecule of a carcinogenic substance may be sufficient
to transform a normal cell into a cancer cell. Evidence is available from
both the human and animal health literature that cancers may arise from a
single transformed cell. Mutation research with ionizing radiation in cell
cultures indicates that such a transformation can occur as the result of
interaction with as little as a single cluster of ion pairs. In reviewing
the available data regarding carcinogenicity, EPA found no compelling
scientific reason to abandon the no-threshold presumption for inorganic
arsenic.
-------
In developing the exposure-risk relationship for inorganic arsenic, EPA
has assumed that a linear no-threshold relationship exists at and below the
levels of exposure reported in the epidemiological studies of occupational
exposure. This means that any exposure to inorganic arsenic is assumed
to pose some risk of lung cancer and that the linear relationship between
cancer risks and levels of public exposure is the same as that between cancer
risks and levels of occupational exposure. EPA believes that this assumption
is reasonable for public health protection in light of presently available
information. However, it should be recognized that the case for the linear
no-threshold dose-response relationship model for inorganic arsenic is not
quite as strong as that for carcinogens which interact directly or in
metabolic form with DMA. Nevertheless, there is no adequate basis for
dismissing the linear no-threshold model for inorganic arsenic. Assuming
that exposure has been accurately quantified, it is the Agency's belief
that the exposure-risk relationship used by EPA at low concentrations
represents only a plausible upper-limit risk estimate in the sense that the
risk is probably not higher than the calculated level and could be much
lower.
The numerical constant that defines the exposure-risk relationship
used by EPA in its analysis of carcinogens is called the unit risk estimate.
The unit risk estimate for an air pollutant is defined as the lifetime cancer
risk occurring in a hypothetical population in which all individuals are
exposed throughout their lifetimes (about 70 years) to an average concentration
of 1 ug/m3 of the agent in the air which they breathe. Unit risk estimates
are used for two purposes: 11) to compare the carcinogenic potency of several
agents with each other, and (2) to give a crude indication of the public
health risk which might be associated with estimated air exposure to these
agents.
-------
The unit risk estimate for inorganic arsenic that is used in this
appendix was prepared by combining the five different exposure-risk numerical
constants developed from four occupational studies.2 The methodology used
to develop the unit risk estimate from the four studies is described in
Section 2 below.
1. 3 Public Exposure
The unit risk estimate is only one of the factors needed to produce
quantitative expressions of public health risks. Another factor needed
is a numerical expression of public exposure, i.e., the numbers of
people exposed to the various concentrations of inorganic arsenic. The
difficulty of defining public exposure was noted by the National Task
Force on Environmental Cancer and Health and Lung Disease in their 5th
Annual Report to Congress, in 1982. 3 They reported that "...a large
proportion of the American population works some distance away from their
homes and experience different types of pollution in their homes, on the
way to and from work, and in the workplace. Also, the American population
is quite mobile, and many people move every few years." They also noted the
necessity and difficulty of dealing with long-term exposures because of
"...the long latent period required for the development and expression
of neoplasia [cancer]..." The reader should note that the unit risk estimate
has been changed from that value used in the inorganic NESHAP proposal as a
result of EPA's analysis of several occupational epidemiological studies that
have recently been completed.
EPA's numerical expression of public exposure is based on two estimates.
The first is an estimate of the magnitude and location of long-term average
ambient air concentrations of inorganic arsenic in the vicinity of emitting
sources based on dispersion modeling using long-term estimates of source
emissions and meteorological conditions. The second is an estimate of the
number and distribution of people living in the vicinity of emitting sources
based on 1980 Bureau of Census data which "locates" people by population
-------
centrolds in census tract areas. The people and concentrations are combined
to produce numerical expressions of public exposure by an approximating
technique contained in a computerized model. The methodology is described
in Section 3 below.
1.4 Public Cancer Risks
By combining numerical expressions of public exposure with the unit
risk estimate, two types of numerical expressions of public cancer risks are
produced. The first, called individual risk, relates to the person or
persons estimated to live in the area of highest concentration as estimated
by the computer model. Individual risk is expressed as "maximum lifetime
risk." As used here, the work "maximum" does not mean the greatest possible
risk of cancer to the public. It is based only on the maximum annual average
exposure estimated by the procedure used. The second, called aggregate risk,
is a summation of all the risks to people estimated to be living within the
vicinity (usually within 50 kilometers) of a source and is customarily summed
for all the sources in a particular category. The aggregate risk is expressed
as incidences of cancer among all of the exposed population after 70 years of
exposure; for convenience, it is often divided by 70 and expressed as cancer
incidences per year. These calculations are described in more detail in
Section 4 below.
There are also risks of nonfatal cancer and other potential health effects,
depending on which organs receive the exposure. No numerical expressions
of such risks have been developed.
-------
2. THE UNIT RISK ESTIMATE FOR INORGANIC ARSENIC?
The following discussion is summarized from a more detailed description
of the Agency's derivation of the inorganic arsenic unit risk estimate as
found in EPA's "Health Assessment Document for Inorganic Arsenic" (EPA-600/
8-83-021F).
2<1 The Linear No-Threshold Model for Estimation of Unit Risk Based on
Human Data (General)
The methodologies used to arrive at quantitative estimates of risk
must be capable of being implemented using the data available in existing
epidemic!ogle studies of exposure to airborne arsenic. This requires
extrapolation from the exposure levels and temporal exposure patterns in
these studies to those for which risk estimates are required. It is assumed
that the age-specific mortality rate of respiratory cancer per year per
100,000 persons for a particular 5-year age interval, i, can be
represented using the following linear absolute or additive risk model:
aj(D) = ai + lOO.OOOa'D ( 1)
With this model, a^ is the age-specific mortality rate per year of
respiratory cancer in a control population not exposed to arsenic, a1 is
a parameter representing the potential of airborne arsenic to cause
respiratory cancer, and D is some measure of the exposure to arsenic up
to the ith age interval. For example, D might be the cumulative dose
in years-pg/m3, the cumulative dose neglecting exposure during the last
10 years prior to the ith age interval, or the average dose in ug/rn3
over some time period prior to the ith age interval. The forms to be used
for D are constrained by the manner in which dose was treated in each
individual epidemiologic study. At low exposures the extra lifetime
probability of respiratory cancer mortality will very correspondingly
( e.g. , linearly).
-------
The dose-response data available in the epidemiologic studies for esti-
mating the parameters in these models consists primarily of a dose measure
Di for the jth exposure group, the person-years of observation Yj, the observed
J
number of respiratory cancer deaths Oj, and the number Ej of these deaths
expected in a control population with the same sex and age distribution as
the exposure group. The expected number Ej is calculated as
j = £ Yjjai/100,000 (2)
here YJ-J is the number of person-years of observation in the ith age cate-
gory and the jth exposure group (Yj = 2 Y ji). This is actually a simplified
representation, because the calculation also takes account of the change in
the age-specific incidence rates with absolute time. The expected number
of respiratory cancer deaths for the ith exposure group is
E(0j) = Z Yji (aj + 100,OOOa'Dj)/100,000
=Ej+a'YjDj (3)
under the linear absolute risk model. Consequently, E(0j) can be expressed
in terms of quantities typically available from the published epidemiologic
s tu d i es.
Making the reasonable assumption that Oj has a Poisson distribution,
the parameter a' can be estimated from the above equation using the method
of maximum likelihood. Once this parameter is estimated, the age-specific
mortality rates for respiratory cancer can be estimated for any desired ex-
posure pattern.
-------
To estimate the corresponding additional lifetime probability of res-
piratory cancer mortality, let bi,...,bi8 be the mortality rates, in the
absence of exposure, for all cases per year per 100,000 persons for the age
intervals 0-4, 5-9,..., 80-84, and 85+, respectively; let ai aig represent
the corresponding rates for malignant neoplasms of the respiratory system.
The probability of survival to the beginning of the ith 5-year age interval
is estimated as
i-1
n [1 - 5bj/100,000] (4)
Given survival to the beginning of age interval i, the probability of dying
of respiratory cancer during this 5-year interval is estimated as
Sai/lOO.OOO (5)
The probability of dying of respiratory cancer given survival to age
85 is estimated as ai8/&i8« Therefore, the probability of dying of respir-
atory cancer in the absence of exposure to arsenic can be estimated as:
17 i-1
PQ = Z [5a-j 7100,000) n (l-5bj/100,000)]
i=l j=l
17
+(ai8/bis) n (1 - 5bj/100,000)
j=l
Here the mortality rates a-j apply to the target population for which risk
estimates are desired, and consequently will be different from those in
-------
(l)-(5), which applied to the epidemiologic study cohort. If the 1976 U.S.
mortality rates (male, female, white, and non-white combined) are used in
this expression, then Pg = 0.0451.
To estimate the probability PEP of respiratory cancer mortality when
exposed to a particular exposure pattern EP, the formula 16) is again used,
but a-,- and b-j are replaced by a-j(D-j) and b-j(Df), where D-j is the exposure
measure calculated for the ith age interval from the exposure pattern EP.
For example, if the dose measure used in (1) is cumulative dose to the be-
ginning of the ith age interval in ug/m3-years, and the exposure pattern
EP is a lifetime exposure to a constant level of 10 ug/m , then D^ =
( i-l)(5)( 10), where the 5 accounts for the fact that each age interval has
a width of 5 years. The additional risk of respiratory cancer mortality is
estimated as
PEP - PO (7)
If the exposure pattern EP is constant exposure to 1 Mg/m , then PEP - PQ is
called the "unit risk."
This approach can easily be modified to estimate the extra probability of
respiratory cancer mortality by a particular age due to any specified
exposure pattern.
2.2 Unit Risk Estimates Derived from Epidemiologic Studies
Prospective studies of the relationship between mortality and exposure
to airborne arsenic have been conducted for the Anaconda, Montana smelter
and the Tacoma, Washington smelter. Table 1 summarizes the fit of the
absolute linear model to dose-response data from 4 different studies at the
two smelters. (See the "Health Assessment Document for Inorganic Arsenic",
-------
10
Table 1
Summary of Quantitative Risk Analyses
Exposed
Population
Anaconda
smel ter
workers
ASARCO
smel ter
workers
Study and
Data Source Model
Lee-Feldstein absolute risk
( heavy exposure
omitted)
Higgins et al. absolute risk
Brown & Chu absolute risk
Enterline 4 Marsh absolute risk
( zero lag)
Results of Goodness-of-Fit Test
x2(d. f. ) p-value "unit" risk3
12.7(5) 0.025 2.80(-3)&
1.2(3) 0.75 4.901-3)
7.01(7) 0.41 1.25(-3)
5.5(4) 0.24 6.8K-3)
Enterline 4 Marsh absolute risk
(10-year lag)
7.0(4) 0.14 7.60(-3)
Additional lifetime risk of respiratory cancer mortality from Hfetfnie environmental exposure"
to 1 ugm-5 arsenic.
b2.80 (-3) means 2.80 x 10'3
-------
11
Chapter 7, EPA-600/8-83-021F for detailed description of occupational studies.)
Table 1 also displays the carcinogenic potencies a*. It should be noted
that the potencies estimated from different models are in different units,
and are therefore not comparable.
The estimated unit risk is presented for each fit for which the chi-
square goodness-of-fit p-value is greater than 0.01. The unit risks derived
from linear models8 in allrange from 0.0013 to 0.0136. The largest of
these is from the Ott et al. study, which probably is the least reliable
for developing quantitative estimates, and which also involved exposures to
pentavalent arsenic, whereas the other studies involved trivalent arsenic.
The unit risks derived from the linear absolute-risk models are considered
to be the most reliable; although derived from 5 sets of data involving 4
sets of investigators and 2 distinct exposed populations, these estimates
are quite consistent, ranging from 0.0013 to 0.0076.
To establish a single point estimate, the geometric mean for data sets
is obtained within distinct exposed populations, and the final estimate is
taken to be the geometric mean of those values. This process is illustrated
in Table 2.
-------
12
Table 2
Combined Unit Risk Estimates for Absolute-Risk Linear Models
Geometric Final
Mean Unit Estimated
Exposure Source Study Unit Risk Risk Unit Risk
Anaconda smelter Brown & Chu 1.25 x 10~3
Lee-Feldstein 2.80 x 10~3 2.56 x 10~3
Higgins et al. 4.90 x 10~3 4.29 x 10~3
ASARCO smelter Enterline &
Marsh 6.81 x 10'3
7.60 x lO'3 7.19 x 10-3
-------
13
3. QUANTITATIVE EXPRESSIONS OF PUBLIC EXPOSURE TO INORGANIC ARSENIC
EMISSIONS
3.1 EPA's Human Exposure Model (HEM) (General)
EPA's Human Exposure Model is a general model capable of producing
quantitative expressions of public exposure to ambient air concentrations
of pollutants emitted from stationary sources. HEM contains ( 1) an atmospheric
dispersion model, with included meteorological data, and (2) a population
distribution estimate based on Bureau of Census data. The input data needed
to operate this model are source data, e.g., plant location, height of the
emission release point, and volumetric rate of release temperature of the
off-gases. Based on the source data, the model estimates the magnitude and
distribution of ambient air concentrations of the pollutant in the vicinity
of the source. The model is programmed to estimate these concentrations
for a specific set of points within a radial distance of 50 kilometers from
the source. If the user wishes to use a dispersion model other than the
one contained in HEM to estimate ambient air concentrations in the vicinity
of a source, HEM can accept the concentrations if they are put into an
appropriate format.
Based on the radial distance specified, HEM numerically combines the
distributions of pollutant concentrations and people to produce quantitative
expressions of public exposure to the pollutant.
3.1.1 Pollutant Concentrations Near a Source
The HEM dispersion model is a climatological model which is a sector-
averaged gaussian dispersion algorithm that has been simplified to improve
computational efficiency.^
-------
Stability array (STAR) summaries are the principal meteorological input to
the HEM dispersion model. STAR data are standard climatological frequency-
of-occurence summaries formulated for use in EPA models and available for
major U.S. meteorological monitoring sites from the National Climatic Center,
Asheville, N.C. A STAR summary is a joint frequency-of-occurence of wind
speed, atmospheric stability, and wind direction, classified according to
Pasquill's categories. The STAR summaries in HEM usually reflect five years
of meteorological data for each of 314 sites nationwide. The model produces
polar coordinate receptor grid points consisting of 10 downwind distances
located along each of 16 radials which represent wind directions. Concen-
trations are estimated by the dispersion model for each of the 160 receptors
located on this grid. The radials are separated by 22.5-degree intervals
beginning with 0.0 degrees and proceeding clockwise to 337.5 degrees. The
10 downwind distances for each radial are 0.2, 0.5, 1.0, 2.0, 5.0, 10.0,
20.0, 30.0, 40.0, and 50.0 kilometers. The center of the receptor grid for
each plant is assumed to be the plant center. Concentrations at other
points were calculated by using a log-linear scheme as illustrated in
Figure 1.
3.1.2 Expansion of Analysis Area
At proposal, exposure and risk were estimated for people residing
within 20 kilometers of the smelter. Some comrnenters pointed out that
since people beyond 20 kilometers are exposed to some level of arsenic due
to a source's emissions, EPA's proposal analysis underestimates the total
exposure and risk. EPA agreed with the commenters and expanded its analysis
out to 50 kilometers. When applying air dispersion models, the EPA's
modeling guidelines recommend that, because of the increasing uncertainty
of estimates with distance from the modeled source and because of the
paucity of validation studies at larger distances, the impact may extend
-------
15
out to 50 kilometers but the analysis should generally be limited to this
distance from the source.^ Such site-specific factors as terrain features
(complex or flat), the objectives of the modeling exercise, and distance to
which the model has been validated will determine the appropriate distance
(whether greater than or less than the guideline distance) for which the
Agency should apply the model.
3.2 Methodology for Reviewing Pollutant Concentrations
Before making HEM computer runs, EPA reviewed small-scale U.S. Geological
Survey topographical maps (scale 1:24000) to verify locational data for each
arsenic source. Plants were given accurate latitude and longitude values which
were then incorporated into the HEM program.
After completing the HEM runs, nearby monitoring sites with ambient
air quality data were identified by a computer search of EPA's National
Aerometric Data Bank (NADB) (Table 3). At some sites, data collected over
several years along with annual averages (based on different numbers of
sample sizes for the years monitored) for each year were available. In
these instances, weighted multi-year averages were calculated to provide an
overall mean for each monitoring site. For purposes of annual mean calculations,
values measured below mimimum detection limits were considered by EPA to be
equal to one-half the detection limit. These ambient arsenic data were
»
then compared to HEM predicted values in order to gauge the accuracy of the
air dispersion model's estimates. As noted above, HEM predicted values
were based on concentrations at 160 polar coordinate receptor grid points
consisting of 10 downwind distances located along each of 16 radials which
represented wind directions. Because the actual monitoring site locations
identified in the NADB retrieval usually did not correspond to exact grid
point locations, a log-linear interpolation scheme (Figure 1) was used to
calculate an estimated concentration at the site.
-------
16
Table 3
Arsenic Concentrations Near ASARCO-East Helena
Primary Lead Smelter
Plant
ASARCO-East
Helena
Company Data
f Obs.
27
41
137
25
31
36
81
23
20
1460
1460
1460
638
1460
274
Distance1
(km)
.5
.7
.8
.9
1.4
1.5
3.9
4.7
7.2
1.1
1.3
1.3
2.1
6.1
7.2
Bearing
119.6
11.5
20.4
343.9'
45.3
156.9
176.5
270.4
273.4
275
5
145
92
275
162
Predicted2
lug/m-5)
0.230
0.078
0.056
0.050
0.076
0.047
0.0159
0.005
0.003
0.024
0.050
0.077
0.071
0.0037
0.0084
Measured3
(Ug/m^)
0.108
0.151
0.242
0.161
0.078
0.109
0.031
0.025
0.030
0.059
0.24
0.078
0.074
0.024
0.028
MDL4
(ygTm3")
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.0055
~
~
Percent! le5
30< * <50
<10
<10
-------
17
Figure 1 Group 2 BG/ED Interpolation
A
R R2
Given:
A - The angle in radians subtended clockwise about the source from due
south to the BG/ED centroid;
Al - The angle from due south to the radial line immediately counter-
clockwise of A, or passing through A if there is an exact match;
A2 - The angle from due south to the radial line immediately clockwise of
Al (A2 is 0 if 1t is due south); ; -
R - The distance in km from the source to the BG/ED centroid;
Rl - The distance from the source to the largest circular arc of radius
less than R;
R2 - The distance from the source to the smallest circular arc of
radius greater than or equal to R;
Cl - The natural logarithm of the concentration value at (Al, Rl);
C2 - The natural logarithm of the concentration value at (Al, R2);
-------
18
C3 - The natural logarithm of the concentration value at (A2, Rl);
C4 - The natural logarithm of the concentration value at (A2, R2);
then:
RTEMP - ln(R/Rl)/ln(R2/Rl);
ATEMP - (A-A1)/(A2-A1);
CA1 - exp(Cl + (C2-Cl)xRTEMP);
CA2 - exp(C3 + (C4-C3)xRTEMP); and
CX - CA1 + (CA2-CAl)xATEMP,
where CX is the interpolated concentration at the BG/ED centroid.
-------
19
3.2.1 Use of Ambient Data
Certain criteria were considered in review of ambient levels. Mean
concentration values derived from sample sizes of less than 25 data points
were disregarded. When reviewing the available monitoring data, it appeared
that monitors situated at distances greater than 15 km from the arsenic
source were considered too far from the source to gauge air dispersion
results without interference from other arsenic sources. Furthermore, at
distances greater than 15 km from the source, plant impacts were often
predicted to be significantly lower than minimum detection limits. These
data were not incorporated in the analyses. A third consideration in
reviewing ambient data concerned the percentage of monitored data which
fell below minimum detection limits. Although some monitoring sites
registered data with over 90 percent of the values above minimum detection
levels, many had about half the data points or more below such levels.
Instances where more than 50 percent of the data were below MDL were dis-
regarded. It should be noted that the various tables in subsequent sections
display, in addition to company-collected data, all ambient monitoring data
that were collected at sites within 15 kilometers of the source as identified
by EPA's computer search although not all the data were used in the final
analysis.
3.2.2 The People Living Near A Source
To estimate the number and distribution of people residing within 50
kilometers of the source, the HEM model uses the 1980 Master Area Reference
File (MARF) from the U.S. Bureau of Census. This data base consists of
enumeration district/block group (ED/BG) values. MARF contains the population
centroid coordinates (latitude and longitude) and the 1980 population of each
ED/BG (approximately 300,000) in the United States (50 states plus the District
of Columbia). HEM identifies the population around each plant, by using the
-------
20
geographical coordinates of the plant, and identifies, selects, and stores
for later use those ED/BGs with coordinates falling within 50 kilometers of
plant center.
3.2.3 Exposure^
The Human Exposure Model (HEM) uses the estimated ground level
concentrations of a pollutant together with population data to calculate
public exposure. For each of 160 receptors located around a plant, the
concentration of the pollutant and the number of people estimated by the
HEM to be exposed to that particular concentration are identified. The HEM
multiplies these two numbers to produce exposure estimates and sums these
products for each plant.
A two-level scheme has been adopted in order to pair concentrations
and populations prior to the computation of exposure. The two level approach
is used because the concentrations are defined on a radius-azimuth (polar)
grid pattern with non-uniform spacing. At small radii, the grid cells are
usually smaller than ED/BG's; at large radii, the grid cells are usually
larger than ED/BG's. The area surrounding the source is divided into two
regions, and each ED/BG is classified by the region in which its centroid
lies. Population exposure is calculated differently for the ED/BG's located
witliin each region. For ED/BG centroids located between 0.2 and 3.5 km
from the emission source, populations are divided between neighboring
concentration grid points. There are 64 (4 x 16) polar grid points within
this range. Each ED/BG can be paired with one or many concentration points.
The population associated with the ED/BG centroid is then divided among all
concentration grid points assigned to it. The land area within each polar
sector is considered in the apportionment.
-------
21
For population centroids between 3.5 and 50 km from the source, a
concentration grid cell, the area approximating a rectangular shape bounded
by four receptors, is much larger than the area of a typical EO/BG. Since
there is an approximate linear relationship between the logarithm of
concentration and the logarithm of distance for receptors more than 2 km
from the source, the entire population of the EO/BG is assumed to be exposed
to the concentration that is logarithmically interpolated radially and
arithmetically interpolated azimuthally from the four receptors bounding
the grid cell. Concentration estimates for 96 (6 x 16) grid cell receptors
at 5.0, 10.0, 20.0, 30.0, 40.0, and 50.0 km from the source along each of
16 wind directions are used as reference points for this interpolation.
In summary, two approaches are used to arrive at coincident concentration/
population data points. For the 64 concentration points within 3.5 km of the
source, the pairing occurs at the polar grid points using an apportionment
of EO/BG population by land area. For the remaining portions of the grid,
pairing occurs at the ED/BG centroids themselves through the use of log-log
and linear interpolation. (For a more detailed discussion of the model used
to estimate exposure, see Reference 5.)
-------
22
3.3 ASARCO-East Helena
Predicted (HEM) versus measured data were plotted (Figure 2) and a
least squares weighted linear regression analysis was run based on thirteen
data points (see Table 3). The least squares regression line (solid line)
was determined on the basis of a conparison of National Aerometric Data Bank
monitoring data (circumscribed dots) and ASARCO monitoring data (circumscribed
Xs) with ambient concentrations predicted by the Human Exposure Model.
The reader should note that a perfect fit for the least squares regression
analysis results in a line running through the origin at a 45° angle (dotted
line on Figure 2). This means that if the HEM model predicts the measured
data perfectly, then the data points would fall on the dotted line. In cases
where the HEM model underpredicts concentrations, data points will be located
above the 45° perfect fit line. Likewise, when the HEM model overpredicts
concentrations, data points will be located below the perfect fit line.
The regression line resulting from our conparison of predicted and monitored
data runs nearly parallel to the perfect fit line but intersects the ordinate
axis at a value of approximately 0.05 ug/m3. This result is consistent with
the expectation that air dispersion modeling would underpredict ambient con-
centrations. The air dispersion modeling did not consider other local
sources of arsenic such as naturally-occurring arsenic in the windblown
dust and reentrained arsenic particulate matter that had settled to the
earth from past smelter emissions.
A study to determine source apportionment for particulate lead and total
suspended particulates (TSP) in East Helena was completed in 1982. High
volume TSP, low volume TSP, and dichotomous samplers were co-collected
(same time period and same site) to permit differences in sample collection
mass and chemistry to be understood. Analysis of hi-vol samples was carried
out by the State of Montana and lo-vol and dichotomous samples were analyzed
-------
23
FIGURE 2 Predicted Versus Measured
Inorganic Arsenic Ambient Concentrations
(ASARCO - East Helena, MT)
0,
J,
c
o
fC
OJ
-------
24
by NEA, Inc. In addition to participate lead and TSP, sanples were also
measured in some cases for arsenic.6
At six locations where arsenic concentrations were measured using both
lo-vol and hi-vol samplers, the ratio of lo-vol to hi-vol in percent ranged
from 104 to 133 with a mean of 118%. This loss of arsenic compounds could
have occurred in two areas: (1) the volatilization of the arsenic compounds
from the hi-vol filter itself during sampling, and (2) the loss of volatile
arsenic compounds during digestion and storage of sanples prior to analysis.
However, based on the data from the study, EPA concluded that the loss of
arsenic on hi-vol filters was relatively minor in nature and within the over-
all accuracy goal of + 15-20% considered adequate for most ambient air quality
measurements.
3*3a Public Exposure to Inorganic Arsenic Emissions from Primary Lead
Smelters
3.3.1.1 Source Data
Five primary lead smelters are included in the analysis. Table 4
lists the names and addresses of the plants considered, and Table 5 lists
the plant data used as input to the Human Exposure Model (HEM).
3.3.1.2 Exposure Data
Table 6 lists, on a plant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products of number of people times the ambient air concentration
to which they are exposed, as calculated by HEM. Table 7 sums, for the
entire source category (5 plants), the numbers of people exposed to various
ambient concentrations, as calculated by HEM. (Source-by-source exposure
results are provided in the EPA docket numbered A-83-23.)
-------
25
TABLE 4
IDENTIFICATION OF PRIMARY LEAD SHELTERS
Plant Number Code Plant Name and Address
ASARCO East Helena. MT
ASARCO El Paso, TX
St. Joe Herculaneum, MD
ASARCO Glover. MO
Amax Boss, MO
-------
26
Table 5 Input Data to Exposure Model Primary Lead Smelting Industry
(Assuming Baseline Controls)
Plant
(Furnace)
ASARCO-East
Helena, MT
ASARCO-
El Paso, TX
Latitude
(Degrees
Minutes
Seconds )
46-34-52
31-47-06
St. Joe- 38-15-47
Herculaneum, MO
ASARCO-
Glover, MO
Amax-
Boss, MO
36-29-46
37-38-31
Longitude
(Degrees
Minutes
Seconds )
111-55-12
106-37-23
90-22-59
90-41-28
91-11-35
Emission
Rate
(Kg/yr)
14700
1680
1680
5040
1680
1680
5040
2772
84
1680
42
1680
50.4
Emission
Point
Elevation
(Meters)
128
130
122
0
186
91
0
107
0
24
0
61
' o
Emission
Point
Diameter
(Meters)
2.7
3.0
3.4
~ ~
4.3
4.9
~ ~
6.1
1.5
4.6
Emission*
Point
Cross
Sectional
Area (m^)
100
100
100
10000
100
100
10000
100
10000
100
10000
100
10000
Emission
Point Gas
Exit
Velocity
m/sec
16.5
19.4
11.7
--
17.0
3.0
--
13.5
18.3
10.0
Emission
Point Gas
Temp.
(°K)
352
330
375
293
345
330
293
353
293
294
293
355
293
Emission
Point
Type
Stack
Stack
Stack
Fugitive
Stack
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
-------
27
TABLE 6 TOTAL EXPOSURE AND NUMBER OF PEOPLE EXPOSED
PRIMARY LEAD SMELTING INDUSTRY*
Total Total
Number of Exposure
Plant People Exposed (People - ug/m3)
1
2
3
4
5
48,600
497,000
1,510,000
97,300
42,700
215
715
186
27
7
* A 50-kilometer radius was used for the analysis of primary lead
smelting industry.
-------
28
TABLE 7
PUBLIC EXPOSURE FOR PRIMARY LEAD SMELTING INDUSTRY
AS PRODUCED BY THE HUMAN EXPOSURE MODEL
(ASSUMING BASELINE CONTROLS)
Concentration
Level (ug/m^)
0.437
U.25
:).!
U.05
0.025
0.01
0.005
0.0025
0.001
0.0005
U. 00025
0.0001
0 .00005
0.0000269
Population
Exposed
(Persons)*
<1
1
40
441
1240
7700
15900
71700
340000
545000
657000
1470000
2080000
2190000
Exposure
(Persons - ug/m^)**
0
0
6
33
62
144
199
398
801
945
983
1100
1150
1150
*Column 2 displays the computed value, rounded to the nearest whole number, of the
cumulative number of people exposed to the matching and higher concentration levels
found in column 1. For example, 0.5 people would be rounded to 0 and 0.51 people
would be rounded to 1.
**Column 3 displays the computed value of the cumulative exposure to the matching
and higher concentation levels found in column 1.
-------
29
3.4 Murph Metals-Dallas and Quemetco-Seattle
Predicted (HEM) versus measured data for Murph Metals-Dallas and
Quemetco-Seattle were plotted (Figures 3 and 4) and a least squares weighted
linear regression analysis was run based on a number of data points. The
least squares regression line (solid line) was determined on the basis of a
comparison of National Aerometric Data Bank monitoring data (circumscribed
dots) and State agency monitoring data (circumscribed Xs) with ambient con-
centrations predicted by the Human Exposure Model.
The reader should note that a perfect fit for the least squares
regression analysis results in a line running through the origin at a 45°
angle (dotted lines in Figures 3 and 4). This means that if the HEM model
predicts perfectly, then the data points would fall on the 45° line. In
cases where the HEM model underpredicts concentrations, data points will be
located above the 45° perfect fit line. Likewise, when the HEM model
overpredicts concentrations, data points will be located below the perfect
fit line. The regression line resulting from our comparison of predicted
and monitored data lies above the perfect fit line, intersecting the
ordinate axis at values of approximately 0.011 ug/m3 and 0.026 ug/m3 for Murph
Metals and Quemetco respectively. This result is consistent with the
expectation that air dispersion modeling would underpredict ambient con-
centrations. The air dispersion modeling did not consider other local sources
of arsenic such as naturally-occurring arsenic in the windblown dust and re-
entrained arsenic particulate matter that had settled to the earth from past
smelter emissions.
-------
30
FIGURE 3 Predicted Versus Measured
Inorganic Arsenic Ambient Concentrations
(Murph Metals - Dallas, TX)
! MODEL
I UNDERPREDICTION
I OVERPREDICTION |j|g
r rS tfft
Perfect Fit
Linear Regression
Municipal Data
0.02 0.04
Predicted Concentration
0.06
-------
Measured Concentration
-5
n>
Q.
_j.
O
O
O
O
fD
-5
CD
O
3
03
OO
o
o
o
oo
fD
fD
3 ii
O O
-s c:
U3 73
fa m
O
-s -s
fD Q.
3 ->
-" O
n <-+
fD
CT <
- fD
fD -S
3 t/1
l/J
O
O 3
3 fD
O O<
fD Wl
3 C
<-» -s
-S fD
Q) f^>
rt-
O
3
cn
-------
32
3-4.1 Public Exposure to Inorganic Arsenic Emissions from Secondary Lead
Smelters
3.4.1.1 Source Data
Thirty-five secondary lead smelters are included in the analysis.
Table 8 lists arsenic concentrations near select secondary lead smelters.
Table 9 lists the names and addresses of the plants considered, and Table 10
lists the plant data used as input to the Human Exposure Model (HEM).
3.4.1.2 Exposure Data
Table 11 lists, on a plant-by-plant basis, the total number of
people encompassed by the exposure analysis and the total exposure. Total
exposure is the sum of the products of number of people times the ambient
air concentration to which they are exposed, as calculated by HEM. Table
12 sums, for the entire source category (35 plants), the number of people
exposed to various ambient concentrations, as calculated by HEM. (Source-
by-source exposure results are provided in the EPA docket numbered A-83-9.)
-------
33
Table 8
Arsenic Concentrations Near Select
Secondary Lead Smelters
Plant
General Battery,
Reading, PA
Murph Metals-
Dallas, TX
Murph Metals-Dallas
Texas Air Control
Board Data
Quemetco-City of
Industry, CA
I Obs.
29
86
21
93
57
28
31
31
29
25
81
85
27
47
30
121
29
Distance^ Bearing
(km)
5.1
3.6
3.7
7.6
9.0
0.2
0.2
0.5
17.8
22.8
23.6
24.0
31.2
32.2
35.1
36.1
38.1
189.5
181.6
181.5
311.6
256.2
0
337.5
157.5
314.0
281.9
164.2
275.2
218.2
161.0
300.0
84.8
235.8
Predicted2
(ng/m3)
0.00104
0.0024
0.0024
0.00095
0.0004
0.062
0.042
0.014
0.00037
0.000133
0.00018
0.00023
0.000083
0.000112
0.000106
0.000113
0.000066
Measured3 MDL4
(wg/m3)
0.009
0.028
0.010
0.029
0.025
0.085
0.077
0.025
0.005
0.003
0.003
0.006
0.003
0.003
0.003
0.005
0.004
(ug/m3)
0.0055
0.05
0.0055
0.05
0.05
..
--
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
0.0055
Percentile5
30< 5
90< !
30< !
90, 5
__
--
70< 5
90< 5
70< 5
70< 5
70< 5
70< 5
I <50
C <95*
I <50*
6 <95*
>99*
6 <90*
>99*
I <95*
I <90*
>99*
6 <90*
>99*
6 <90*
i <90*
Quemetco-
Indianapol 1s, IN
64
12.5
78.2
0.00040
0.005 0.0055 50< % <70*
Quemetco-Seattle,
WA
Quemetco- Seattle
Washington State Dept.
of Ecology Data
80
60
72
60
60
1.9
3.2
13.5
0.2
1.4
150.8
30.3
2.6
157.5
180
0.0036
0.00183
0.00047
0.031
0.0075
0.041
0.038
0.020
0.09
0.03
0.0055
0.0055 30< !
0.0055 30< !
..
<10
6 <50
I <50
*Indicates data point was disregardfed - see Section 3.3.1.1.
^Distance from source to monitor (km).
Concentration predicted by Human Exposure Model (HEM). See Section 3.1.
3The measured values are weighted averages. When the sampled arsenic concentrations were
below the MDL, a value of 1/2 MDL was assumed for purposes of calculating the annual averages.
^Minimum Detection Limit.
5Percentile indicates percentage of data falling below minimum detectable levels.
-------
34
Table 9
Identification of Secondary Lead Smelters
Plant Number Code
2
4
0
6
1
10
12
14
Ib
16
I/
18
19
20
21
22
23 x
24
25
26
27
28
29
30
31
32
33
34
35
Plant Name and Address
Alco Pacific Gardena, CA
Bergsoe St Helens, OR
Chloride Metals Columbus, (iA
Chloride Metals Tampa, FL
Dixie Metals Dallas, TX
East Penn Lyons Station
Federated Metals San Fran, CA
General Battery Reading, PA
General Smelting college Grov, TN
Gopher Eugene, Minn
Gould Frisco, TX
Gould Vernon, CA
Gulf Coast Tampa, FL
Hyman Viener Richmond, VA
Interstate Lead Leeds, AL
Lancaster Lancaster, PA
Master Metals Cleveland, OH
Murph Metals Dallas, TX
National Smelting Atlanta, GA
National Smelting Pedricktown, NJ
Quemetco City of Industry, CA
Quemetco Indianapolis, IN
Quemetco Seattle, WA
Refined Metals Beach Grove, IN
Refined Metals Memphis, TN
Revere Wall , NY
Ross Metals Rossville, TN
Sanders Lead Troy, Al
Schuylkill Baton Rouge, LA
Schuylklll Forest City, MO
Standard San Antonio, TX
Taracorp Atlanta, GA
Taracorp Granite City, IL
Tonolli Nesquehoning, PA
USS Lead E. Chicago, IN
-------
35
Table 10
Secondary Lead Industry Inputs to HEM Model
(Assuming Baseline Controls)
Process
As
Emission
Latitude /Longitude
Plant (Deqrees-Minutes-Seconds)
Alco Pacific 33-50-20/118-18-07
Bergsoe 45-50-58-122-49-3
Chloride 32-26-00/84-56-00
Metal s/GA
Chloride 27-54-5/82-24-12
Metal s/FL
Dixie Metals 32-44-49/96-46-37
East Penn 40-28-19/75-58-23
Federated 37-44-/122-23
Metals/CA
General Battery 40-22-45/75-54-50
General Smelting 35-48-00/86-40-05
Gopher 40-50-/93-7-30
Gould/TX 33-08-38/96-49-44
Gould/CA 34-00-14/118-13-45
Gulf Coast 27-57-44/82-22-53
Hyman Viener 37-31-10/77-24-54
Interstate Lead 33-31-58/86-32-00
Lancaster 40-03-11/76-19-52
Master Metals 41-28-52/81-40-48
Murph Metals 32-46-40/96-52-21
National 33-47-31/84-24-18
Smelting/GA
National 39-45-30/75-25-30
Smelt ing/NJ
Rate
kg/yr
7.6
86.6
34.9
34.9
52.3
52.3
14.5
91.5
17.4
43.6
87.2
53.4
36.3
11.6
69.7
0.4
69.7
174.3
43.6
174.3
Stack
Ht.
m
31
31
31
31
31
31
31
26
31
31
31
26
31
31
31
31
31
26
31
26
Stack
Diam.
m
0.62
0.92
0.92
0.92
0.92
0.92
0.62
1.2
0.62
0.92
0.92
1.2
0.92
0.62
0.92
0.62
0.92
1.2
0.92
1.2
Stack
Vel.
m/sec
29
25.9
25.9
25.9
25.9
25.9
29
16.6
29
25.9
25.9
16.6
25.9
29
25.9
29
25.9
16.6
25.9
16.6
Exit
Tenp.
°K
331
400
400
400
400
400
400
400
400
400
400
331
400
400
400
400
400
400
400
400
Process
Emission Stack
Rt.
kg/yr
.25
2.6
0.5
0.5
0.75
0.75
0.21
2.98
0.25
0.62
1.24 -
3.83
0.52
0.17
0.99
0.01
0.99
2.48
0.62
2.48
Ht.
m
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
Fugitive
Stack
Diam.
m
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
Stack
Vel.
m/sec
10.4
13.3
13.3
13.3
13.3
13.3
10.4
10.6
10.4
13.3
13.3
10.6
13.3
10.4
13.3
10.4
13.3
10.6
13.3
10.6
Area
As
Exit Emission
Temp . Rt .
°K k/yr
311 4.0
311 0
311 17.8
311 13.8
311 168.9
311 20.0
311 4.0
311 253.8
311 7.1
311 46.2
311 13.8
311 0
311 16.9
311 7.1
311 67.6
311 7.1
311 1.3
311 692.1
311 18.7
311 115.6
Area
m
2,945
5,862
5,862
5,862
5,862
5,862
2,945
8,788
2,945
5,862
5,862
8,788
5,862
2,945
5,862
2,945
5,862
8,788
5,862
8,788
-------
36
Plant
Quemetco/CA
Quemetco/IN
Quemetco/WA
Refined
Metals/IN
Refined
Metals/TN
Revere.NY
Ross Metals
Sanders Lead
Schuylkill/LA
Schuylkill/MO
Standard
Taracorp/GA
Taracorp/IL
Tonolli
USS Lead
Lat/Lonq
34-01-30/117-58-58
39-45-14/86-17-59
47-34-44/122-21-04
39-42-36/86-03-5 4"
35-05-13/90-04-10
41-27-37/74-21-35
35-02-42/89-34-30
31-47-28/85-58-16
30-58-08/91-14-40
40-01-59/95-13-59
29-20-00/98-29-38
33-47/84-22
38-42-05/90-08-37
40-51-03/75-52-46
41-36-58/87-27-47
Table 10 (Continued)
Secondary Lead Industry Inputs to HEM Model
(Assuming Baseline Controls)
As
Emission
Rate
kg/yr
61.0
139.5
69.7
52.3
87.2
139.5
27.9
116.2
232.4
104.6
23.5
87.2
244.1
57.2
58.1
P
Stack
Ht.
m
26
26
31
31
31
26
31
31
26
31
31
31
26
26
31
roc ess
Stack
Oiam.
m
1.2
1.2
0.92
0.92
0.92
-1.2
0.62
0.92
1.2
0.92
0.62
0.92
1.2
1.2
0.92
Stack
Vel.
m/sec
16.6
16.6
25.9
25.9
25.9
16.6
29
25.9
16.6
25.9
29
25.9
16.6
16.6
25.9
Exit
Terrp.
°K
331
400
400
400
400
400
400
400
400
400
400
400
400
331
400
Emission Stack
Rt. Ht.
kg/yr m
1.99
1.99
0.99
0.75
1.24
1.99
0.39
1.66
3.31
1.49
0.34
1.24
3.48
1.86
0.83
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
15.4
fugitive
Stack Stack
Diam. Vel.
m m/sec
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
10.6
10.6
13.3
13.3
13.3
10.6
10.4
13.3
10.6
13.3
10.4
13.3
10.6
10.6
13.3
Exit
Tenp.
°K
311
311
311
311
311
311
311
311
311
311
311
311
311
311
311
Area
As
Emission
Rt. Area
692.1
692.1
461.4
76.9
76.9
692.1
0
29.3
20.9
13.8
7.6
13.8
23.1
69.3
61.3
8,788
8,788
5,862
5,862
5,862
8,788
2,945
5,862
8,788
5,862
2,945
5,862
8,788
8,788
5,862
-------
37
Plant
Table 11
Total Exposure and Number of People Exposed
Secondary Lead Smelting Industry*
Total
Number of
People Exposed
Total
Exposure
(People - ug/m3)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
8,450,000
1,120,000
314,000
1,670,000
2,350,000
1 , 310 ,000
3,370,000
1,260,000
679,000
58,200
1,800,000
8,900,000
1,690,000
766,000
844,000
1,160,000
2,530,000
2,560,000
1,920,000
4,210,000
8,860,000
1,150,000
2,060,000
1,180,000
927 ,000
948,000
902,000
92,000
143,000
149,000
1,050,000
1,920,000
2,190,000
934,000
5,280,000
32
15
17
19
228
16
22
250
3
1
17
189
32
15
41
6
59
668
51
197
2300
460
576
67
115
249
4
11
4
2
24
83
159
21
103
* A 50-kilometer radius was used for the analysis of secondary lead
smelters.
-------
38
Table 12
Public Exposure for Secondary Lead Smelters
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
0.101
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
Population
Exposed
(Persons)*
<1
<1
256
2880
16000
53100
152000
7 43000
1940000
4510000
12800000
21700000
31100000
44000000
55700000
64100000
70800000
73800000
7 4700000
Exposure
(Persons - ug/rn^)**
0
0
16
104
300
543
878
1770
2590
3480
4750
5390
5720
5930
6010
6050
6060
6060
6060
Column 2 displays the computed value, rounded to the nearest whole number
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For exarple, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the conputed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
39
3.5 Public Exposure to Inorganic Arsenic Emissions from Primary Zinc
Smelters
3. 5.1 Source Data
Five primary zinc smelters are included in the analysis. Table 13
lists ambient arsenic concentrations near select primary zinc smelters.
Table 14 lists the names and addresses of the plants considered, and Table
15 lists the plant data used as input to the Human Exposure Model (HEM).
3.5.2 Exposure Data
Table 16 lists, on a pi ant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products of number of people times the ambient air
concentration to which they are exposed, as calculated by HEM. Table 17
sums, for the entire source category (5 plants), the numbers of people
exposed to various ambient concentrations, as calculated by HEM. I Source-
by-source exposure results are provided in the EPA docket numbered A-83-23. )
-------
Plant
St . Joe-
Monaca , PA
ASARCO-
Corpus Christi ,
TX
Amax-
Sauget, IL
Jersey Miniere
Zinc Co-
Clarksville, TN
National Zinc-
Bartlesville, OK
Distance-^
# Obs (km)
40
Table 13
Arsenic Concentrations Near
Select Primary Zinc Smelters
Predicted^
Bearing (ug/m3)
No data within Ib km
36 3.0
299 3.8
33 5.5
83 6.2
319 6.9
26 14.9
190 14.9
27 4.1
No data within
No data within
262.8 0.000024
252.1 0.0000149
299.4 0.0000156
289.7 0.0000125
173.0 0.0000067
187.7 0.0000029
187.7 0.0000029
314.4 0.000042
15 km
15 km
Measured3 MDL4
(ug/m3) (|jg/m3) Percentile 5
0.025 0.05 100*
0.026 0.05 95< % <99*
0.025 0.05 100*
0.029 0.05 95< % <99*
0.026 0.05 95< % <99*
0.008 0.0055 30< % <50*
0.028 0.05 90< % <95*
0.007 0.0055 50< % <70*
* Indicates data point was disregarded; see Section 3.1.
* Distance from source to monitor (km).
2 Concentration predicted by Human Exposure Model (HEM).
The measured values are weighted averages. When the sampled arsenic concentrations were below the MDL, a
value of 1/2 MDL was assumed for purposes of calculating the annual averages.
^Minimum detention limit.
5 Percentile indicates percentage of data falling below minimum detectable levels.
-------
41
Table 14
Identification of Primary Zinc Smelters
Plant Number Code Plant Name and Address
1 St. Joe - Monaca, PA
2 ASARCO - Corpus Christi, TX
3 Amax - Sauget, IL
4 Jersey Miniere Zinc Co - Clarksville, TN
5 National Zinc - Bartlesville, OK
-------
42
Table 15
Input Data to Exposure Model
Primary Zinc Smelting Industry
(Assuming Baseline Controls)
Plant Lati
(Deg
Min
Sec
tude Longitude
rees (Degrees
utes Minutes
onds) Seconds)
St. Joe - 40-40-12 80-20-10
Monaca, PA
ASARCO - 27-48-00 97-23-46
Corpus Christi,
TX
Amax - 38-36-07 90-10-16
Sauget, IL
Jersey 36-30-54 87-24-14
Miniere
Zinc Coup .
Clarksville, TN
Emission
Rate
(Kg/yr)
109
8.4
67.7
8.4
8.4
. 23.5
25.2
Emission
Point
Elevation
(Meters)
61
31
37
32
92
46
61
Emission
Point
Diameter
(Meters)
3.4
1.8
2.1
2.0
2.0
1.5
1.8
Emission
Point
Cross
Sectional
Area (m^)
100
100
100
100
100
100
100
Emission
Point Gas
Exit
Velocity
m/sec
7.0
1.8
10.7
1.1
2.7
9.7
6.3
Emission
Point Gas
Temp .
325
325
336
325
389
373
334
Emission
Point
Type
Stack
Stack
Stack
Stack
Stack
Stack
Stack
National Zinc 36-44-24 95-58-59
Bartelsvi 1 le,
OK
11.3
31
1.3
100
5.7
325
Stack
-------
43
Table 16
Total Exposure and Number of People Exposed
Primary Zinc Smelter*
Total Total
Number of Exposure
Plant
1
2
3
4
5
People Exposed
2,000,000
336,000
2,200,000
235,000
120,000
V people - ug/m~j
47
2
16
3
2
* A 50-kilometer radius was used for the analysis of primary
zinc smelters.
-------
44
Table 17
Public Exposure for Primary Zinc Smelters
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (ug/m3)
0.00182
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.00000005
Population
Exposed
(Persons)*
7
109
2350
19400
91700
212000
450000
1870000
3170000
3960000
4650000
4800000
4850000
4870000
4890000
Exposure
(Persons - ug/m^**
0
0
2
8
18
27
35
55
65
68
69
69
69
69
69
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
45
3.6 Public Exposure to Inorganic Arsenic Emissions from Zinc Oxide
Plants
3.6.1 Source Data
Two zinc oxide plants are included In the analysis. Table 18 lists
ambient arsenic concentrations near select zinc oxide plants. Table 19
lists the names and addresses of the plants considered, and Table 20 lists
the plant data used as input to the Human Exposure Model (HEM).
3.6. 2 Exposure Data
Table 21 lists on a pi ant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total
exposure is tHe sum of tHe products of number of people times the ambient
air concentration to which they are exposed, as calculated by HEM.
Table 22 sums, for the entire source category (2 plants), the numbers of
people exposed to various ambient concentrations, as calculated by HEM.
I Source-by-source exposure results are provided in the EPA docket numbered
A-83-11.)
-------
Table 18
Arsenic Concentrations Near
Select Zinc Oxide Plants
Plant
ASARCO-
Columbus, OH
# Obs
127
Distance1
(km)
3.8
Bearing
206.0
Predicted2
(uq/m3)
0.0000124
Measured3
(ug/m3)
0.006
MDL4
(ug/m3)
0.0055
Percentile^
70< 51
', <90*
New Jersey Zinc- No data within 15 km
Palmerton, PA
* Indicates data point was disregarded; see Section 3.5.1.1.
* Distance from source to monitor (km).
2 Concentration predicted by Human Exposure Model (HEM).
J The measured values are weighted averages. When the sanpled arsenic concentrations
were below the MDL, a value of 1/2 MDL was assumed for purposes of calculating the
annual averages.
4 Minimum detection limit.
5 Percentila indicates percentage of data falling below minimum detectable levels.
-------
47
TABLE 19
Identification of Zinc Oxide Plants
Plant Number Code Plant Name and Address
1 ASARCO-Columbus, OH
New Jersey Zinc
Palmerton, PA.
-------
48
Table 20
Input Data to Exposure Model Zinc Oxide Plants
(Assuming Baseline Controls)
Plant
Latitude
(Degrees
Minutes
Seconds)
Longtitude
(Degrees
Minutes
Seconds )
Emission
Rate
(Kg/yr)
Emission
Point
Elevation
(Meters)
Emission
Point
Diameter
(Meters)
Emission*
Point
Cross
Sectional
Area (m^)
Emission
Point Gas
Exit
Velocity
m/sec
Emission
Point Gas
Temp .
ASARCO-
Columbus, OH 39-59-53 82-58-48 11.3 61 1.2 100 7.5 333
Emission
Point
Type
Stack
New Jersey 3155 24 5 100 7.3 411 Stack
Zinc- 40-49-41 75-35-22 2656 18 1.8 100 7.9 364 Stack
Palmer-ton, PA 2754 9 1.2 100 17.4 466 Stack
-------
49
Table 21
Total Exposure and Number of People Exposed
(Zinc Oxide Plants)*
Plant
1
2
Total Number of
People Exposed
1,210,000
907,000
Total Exposure
(People - ug/m^)
8
1260
* A 50 kilometer radius was used for the analysis of zinc oxide plants,
-------
50
Table 22
Public Exposure for Zinc Oxide Plants
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/mj) ; (Persons)*
**
0.269
0.25
0.1
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
2
2
138
1160
3990
11700
21300
54000
392000
732000
883000
908000
913000
976000
1160000
1360000
1610000
1840000
1960000
2110000
o
(J
17
-L /
79
180
300
366
474
921
-s I J.
1200
1250
1260
1260
1260
1260
1260
1270
1270
1270
1270
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the conputed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
51
3.7 Methodology for Reviewing Pollutant Concentrations - Cotton Gins
A total of 320 cotton gins were identified as processors of arsenic
desiccated cotton. Due to the large number of gins, EPA determined that it
was impractical to obtain the location data necessary for arsenic risk
assessment. Based on information regarding the range of processing rates
possible, four model plants operating at 4, 7, 12 and 20 bales/hour were
designed that are representative of the operations and emissions of the gin
population.7 These were located at each of three sites typical of the areas
in which the gins are located. Of the 320 gins, it was assumed that 32
processed 4 bales per hour, 96 processed 7 bales per hour, 160 processed 12
bales per hour, and 32 processed 20 bales per hour. The Human Exposure
Model was run for each scenario to establish a range of exposure and risk
estimates for individual sources. To provide data for validating the model
plant exposure estimates, two operating gins in south central Texas were
chosen for test sites over a one year period. Monitors were arranged in a
fan-like array of sites positioned at distances of 100, 200 and 400 meters
downwind of the gin. Upwind sites were placed at 400m (one gin only) and
100m. This configuration provided a total of 13 sanpling sites. The study
was conducted over a period of one year with intense sampling (4 hour
intervals) for 15 days during the short ginning season followed by 6 day
interval sampling for the remainder of the year.
Data from these two gins were compared to Human Exposure Model
calculated values (Table 23). The comparison was hampered somewhat by
the large number of monitored values which fell below minimum detection
limits -- only 298 measurements out of 708 were above the MDL of 0.05
ug/m3. To circumvent this problem, a range of mean measured values was
developed. At one end, all values below MDL were considered as zero
values, and at the other end, all such values were considered equal to the
MDL of 0.05 ug/m3.
-------
52
Table 23
Arsenic Concentrations Near Two Texas Cotton Gins
Plant
A ( =9 bales/hr)
B ( =12 bales/hr)
Distance ( km)1
0.1
0.2
0.2
0.1
0.1
0.2
0.1
Predicted (HEM)
( uq/m3)
._
0.011
0.011
_
0.011
Measured2
( ug/m3)
0.083-0.088
_
0.051-0.060
0.12 -0.12
0.015-0.024
0.013-0.022
0.013-0.022
Distance from source to monitor.
Weighted mean concentrations for one calendar year. Lesser value
represents weighted mean concentration calculated with values less
than minimum detection limit set equal to zero. Greater value
represents weighted mean concentration calculated with values less
than minimum detection limit set equal to MDL (0.0065 ug/m3).
-------
53
When conparing the measured arsenic values to the predicted con-
centrations from the appropriate model gin exposure analysis, EPA found
that the predicted values were reasonably close to concentrations measured
very near the gins. The monitoring study data also showed that the
arsenic concentrations fell off very rapidly with distance from the gins.
This result suggests that people living at some distance from the gins
are not being significantly exposed to the gins' emissions. Such a
result, coupled with the observation that many gins are in rural areas
supports the Agency's conclusion that the aggregate risks for this source
category are low.
3.7.1 Public Exposure to Inorganic Arsenic Emissions from Cotton Gins
3.7.1.1 Source Data
Four model cotton gins at each of three geographic locations are
included in the analysis. Table 24 lists the names and addresses of the
plants considered, and Table 25 lists the plant data used as input to the
Human Exposure Model (HEM).
3.7.1.2 Exposure Data
Tables 26 - 37 sum, for the entire source category (12 plants), the
numbers of people exposed to various ambient concentrations, as calculated
by HEM. (Model plant-by-model plant exposure results are provided in the
EPA docket numbered A-83-10.)
-------
54
Table 24
Identification of Model Cotton Gins
Model Plant Location
Model Plant Production
Hutto, TX
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
Buckholtz, TX
4 Rales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
Itasca, TX
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales/Hour
-------
55
Table 25
Input Data to Exposure Model Cotton Gins
Plant Latitude
(Degrees
Minutes
Seconds )
Hutto, TX 30-33-00
4 Bales/Hour
7 Bales/Hour
12 Bales/Hour
20 Bales /Hour
Buckholts, TX 30-52-00
4 Bales/Hour
7 Bales /Hour
12 Bales /Hour
20 Bales/Hour
Itasca, TX 32-10-00
4 Bales /Hour
7 Bales /Hour
12 Bales /Hour
20 Bales/Hour
Longitude Emission
Rate
.(Degrees
Minutes
Seconds) (Kg/yr)
97-33-00
1.0
1.0
2.7
2.7
4.6
4.6
10.2
10.2
97-OB-OO
1.0
1.0
2.7
2.7
4.6
4.6
10.2
10.2
97-09-00
1.0
1.0
2.7
2.7
4.6
4.6
10.2
10.2
(Assuming
Emission
Point
Elevation
(Meters)
9
5
9
5
9
5
10
5
9
5
9
5
9
5
10
5
9
5
9
5
9
5
10
5
Baseline Controls)
Emission
Point
Diameter
(Meters)
0.3
0.4
0.4
0.4
0.3
0.4
0.4
0.4
0.3
0.4
0.4
0.4
Emission
Point
Cross
Sectional
Area (itr)
25
12
25
12
25
27
25
27
25
12
25
12
25
27
25
27
25
12
25
12
25
27
25
27
Emission
Point Gas
Exit
Velocity
m/sec
20.4
20.4
20.4
20.4
20.4
20.4
20.4
20.4
20.4
20.4
20.4
20.4
Emission
Point Gas
Temp.
(°K)
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
298
Emission
Point
Type
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
Stack
Fugitive
-------
56
Table 26
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/m3) (Persons)* (Persons-ug/nr3)**
0.00263
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
1
1
1
6
23
112
177
433
1810
1810
3390
46800
285000
506000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would he
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
57
Table 27
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
(Persons)* (Persons-ug/m-*)**
O.OU682
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.000000528
1
1
1
8
28
112
282
799
1810
2420
7220
55400
448000
506000
0
0
0
0
0
0
0
0
0
0
0
0
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
58
Table 28
Public Exposure for 12 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/mj) (Persons)* (Persons-ug/m3)**
0.011
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
O.OOOU5
0.000025
0.00001
0.000005
0.0000025
0.000001
1
1
1
5
25
102
161
523
1810
1810
3820
39500
232000
506000
0
0
0
0
0
0
0
0
0
0
0
1
1
2
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
59
Table 29
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Hutto.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
(Persons)* (Persons-ug/m-3)**
0.0234
0.01
0.005
0.0025
0.001
0 .0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
1
1
5
23
102
169
433
1810
1810
3390
46800
300000
506000
0
0
0
0
0
0
0
1
1
1
1
3
4
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
60
Table 30
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Buckholts,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (u9/m3) (Persons)* (Persons-ua
**
0.00263
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.000000196
<1
<1
<1
2
10
49
77
190
1050
1050
4020
10600
86700
129000
131000
n
\J
o
o
o
o
o
o
o
o
o
o
o
n
\J
o
0
* Column 2 displays the computed value, rounded to the nearest whole number
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
61
Table 31
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Buckholts.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
(Persons)* (Persons-ug/nr3)**
0 .00682
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.000000528
<1
<1
<1
3
12
49
124
269
1050
1050
6020
15500
121000
131000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
62
Table 32
x f°r 1Z Bales/Hour Model Cotton Gin
.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration
Level (uq/m3)
0.011
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
Population Exposed
(Persons)*
2
11
45
71
230
1050
1050
6020
10100
81500
131000
Exposure
(Persons-ug/rn3)**
0
0
0
0
0
0
0
0
0
0
0
0
0
1
Co.lumn 2 delays the corrputed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For exanple, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the corrputed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
63
Table 33
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Buckholts.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/m3) (Persons)* (Persons-ug/m3)**
0.0234
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000002
<1
<1
2
10
45
74
190
1050
1050
4020
11300
89500
129000
131000
0
0
0
0
0
0
0
0
0
0
1
1
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
64
Table 34
Public Exposure for 4 Bales/Hour Model Cotton Gin
(Itasca.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/m3) (Persons )* (Persons-no
0.0011
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.0000000634
1
1
5
19
57
153
489
1280
2140
2660
6520
38900
107000
156000
162000
yrci auiia ~ \J\}/ HI )
o
n
\J
o
o
\J
0
o
\J
Q
\J
o
o
o
o
0
o
o
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
65
Table 35
Public Exposure for 7 Bales/Hour Model Cotton Gin
(Itasca,TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (ug/m^) (Persons)* (Persons-ug/nr)**
0.00285
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.00000025
0.000000171
1
1
7
23
70
167
587
1280
2140
3870
6520
65200
120000
159000
162000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column .3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
66
Table 36
Public Exposure for 12 Bales/Hour Model Cotton Gin
(Itasca.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
Level (pg/m-3) (Persons)* fPersons-uo
0.00461
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0.0000005
0.000000293
1
4
22
42
118
489
948
1980
2660
4970
20300
114000
153000
162000
yrci auiia ~n^)/ in /
n
-------
67
Table 37
Public Exposure for 20 Bales/Hour Model Cotton Gin
(Itasca.TX) as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Concentration Population Exposed Exposure
,T /.._/_3\ (Persons)* (Persons-ug/rrr)**
0.0097
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0.000005
0.0000025
0.000001
0 .0000006 49
1
4
14
46
146
489
1280
2140
2660
6520
40100
109000
156000
162000
0
0
0
0
0
0
0
0
0
1
1
1
1
1
* Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
** Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
63
3.8 Public Exposure to Inorganic Arsenic Emissions from Arsenic
Plants
3.8.1 Source Data
Eight arsenic chemical plants are included in the analysis. Table 38
lists ambient arsenic concentrations near select arsenic chemical plants.
Table 39 lists the names and addresses of the plants considered, and Table
40 lists the plant data used as input to the Human Exposure Model (HEM).
3.8.2 Exposure Data
Table 41 lists, on a plant-by-plant basis, the total number of people
encompassed by the exposure analysis and the total exposure. Total exposure
is the sum of the products of numbers of people times the ambient air con-
centration to which they are exposed, as calculated by HEM. Table 42 sums,
for the entire source category (8 plants), the numbers of people exposed to
various ambient concentrations, as calculated by HEM. (Source-by-source
exposure results are provided in the EPA docket numbered A-83-23.)
-------
69
Table 38
Arsenic Concentrations Near Select
Arsenic Chemical Plants
Distance1 Predicted2 Measured3
Plant
Diamond Shamrock-
Greens Bayou, TX
Koppers Co.-
Conley, GA
Koppers Co.-
Valparaiso, IN
Mineral R&D-
Concord, NC
Osmose Wood
Preserving Co.-
Memphis, TN
Pennwalt Inc.-
Bryan, TX
Vineland Chemical-
Vineland, NJ
# Obs
302
20
37
143
261
45
26
30
76
32
72
107
(km)
1.
5.
5.
6.
8.
8.
11.
12.
13.
No
No
5.
9.
0.
No
6
5
9
2
3
7
5
0
0
data
data
9
3
1
data
Bearing
291.1
229.7
186.3
197.3
70.3
62.2
195.9
143.0
336.1
within 15
within 15
27.7
20.8
296.4
within 15
0
0
0
0
0
0
0
0
0
km
km
0
0
km
(yg/m3)
.00000042
.000000042
.000000051
.000000029
.0000000118
.0000000105
.0000000121
.0000000090
.0000000079
.000099
.000054
(ug/m3)
0
0
0
0
0
0
0
0
0
0
0
0
.027
.026
.008
.027
.025
.026
.025
.025
.008
.005
.004
.026
MOL4
(ug/m3)
0.05
0.05
0.0055
0.05
0.05
0.05
0.05
0.05
0.0055
0.0055
0.0055
0.05
Percent! le^
90< %
50< %
95< %
95< %
100
100
70< %
70< °k
50< 5
95< 51
<95*
>99*
<70*
<99*
>99*
*
*
*
<90*
; <9o*
; <7o*
i <99*
Group-Bonham.TX
* Indicates data point was disregarded; see Section 3.7.1.1. ~ ~~~
* Distance from source to monitor (km).
2 Concentration predicted by Human Exposure Model (HEM).
3 The measured values are weighted averages. When the sampled arsenic concentrations were below the MDL, a
value of 1/2 MDi was assumed for purposes of calculating the annual averages.
4 Minimum detection limit.
levels*116 1ndicates Percentage of data falling below minimum detectable
-------
70
Table 39
Identification of Arsenic Chemical Plants
Plant Number Code Plant Name and Address
1 Diamond Shamrock - Greens Bayou, TX
2 Koppers Co. - Conley, GA
3 Koppers Co., - Valparaiso, IN
4 Mineral Research & Development Co. -
Concord, NC
5 Osmose Wood Preserving Co., -Memphis, TN
6 Pennwalt Inc. - Bryan, TX
7 Vineland Chemical - Vine!and, NJ
8 Voluntary Purchasing Group - Bonham, TX
-------
71
Table 40
Input Data to Exposure Model Arsenic Chemical Plants
(Assuming Baseline Controls)
Plant
Latitude
(Degrees
Minutes
Seconds )
Longitude
(Degrees
Minutes
Seconds)
Diamond Shamrock- 29-45-58 95-12-22
Greens Bayou ,TX
Koppers Co.-
Conley, GA
Koppers Co.-
Valparaiso.IN
Mineral R&D Co
Concord, NC
Osmose Wood
Preserving Co
Memphis.TN
Pennwalt Inc.-
Bryan, TX
Vineland
Chemical-
Vineland, NJ
Voluntary
Purchasing
Group-
Bonham.TX
33-38-42 84-19-34
41-28-34 87-04-40
.- 35-24-29 80-34-44
35-05-13 90-04-19
>""
30-40-30 96-22-12
39-55-59 74-44-53
33-34-41 96-10-41
Emission
Rate
(Kg/yr)
0.030
0.027
0.054
0.022
51.3
0.019
0.001
0.019
Emission
Point
Elevation
(Meters)
13
31
9
11
5
10
13
10
Emission
Point
Diameter
(Meters)
0.38
0.61
0.76
0.50
0.36
0.53
0.38
0.53
Emission
Point
Cross
Sectional
Area (m2)
3000
3000
3000
3000,
3000
3000
3000
3000
Emission
Point Gas
Exit
Velocity
m/sec
11.9
0.1
1.6
8.8
14.3
8.9
0.1
8.9
Emission
Point Gas
Temp .
(°K)
298
298
298
298
298
298
298
298
Emission
Point
Type
Stack
Stack
Stack
Stack
Stack
Stack
Stack
Stack
-------
Plant
1
2
3
4
5
6
7
8
72
Table 41
Total Exposure and Number of People
(Arsenic Chemical Plants)*
Total Number of
People Exposed
2,680,000
1,900,000
1,190,000
813,000
927,000
138,000
4,230,000
152,000
Exposed
Total Exposure
(People - ug/m3)
0
0
0
0
68
0
0
0
* A 50-kilometer radius was used for the analysis of arsenic chemical
olants.
-------
73
Concentration
Table 42
Public Exposure for Arsenic Chemical Plants
as Produced by the Human Exposure Model
(Assuming Baseline Controls)
Population Exposed
(Persons)*
Exposure
(Persons-gg/m3)**
0.0541
0.05
0.025
0.01
0.005
0.0025
0.001
0.0005
0.00025
0.0001
0.00005
0.000025
0.00001
0 .000005
0.0000025
0.000001
0.0000005
0.00000025
0.0000001
0.00000005
0.000000025
0.00000001
0.000000005
0.0000000025
0.000000001
0.0000000005
0.00000000025
0.0000000001
0.00000000005
0.0000000000395
31
31
62
472
1440
2720
8130
16700
30900
105000
210000
374000
653000
852000
908000
935000
952000
984000
1040000
1100000
1320000
2230000
3670000
5880000
7750000
7890000
8320000
9910000
11900000
12000000
2
2
3
10
15
19
28
34
38
49
56
61
66
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
68
*Column 2 displays the computed value, rounded to the nearest whole number,
of the cumulative number of people exposed to the matching and higher
concentration levels found in column 1. For example, 0.5 people would be
rounded to 0 and 0.51 people would be rounded to 1.
**Column 3 displays the computed value of the cumulative exposure to the
matching and higher concentration levels found in column 1.
-------
74
4 QUANTITATIVE EXPRESSIONS OF PUBLIC CANCER RISKS FROM INORGANIC ARSENIC
EMISSIONS
4.1 Methodology (General)
4.1.1 The Two Basic Types of Risk
Two basic types of risk are dealt with in the analysis. "Aggregate
risk" applies to all of the people encompassed by the particular analysis.
Aggregate risk can be related to a single source, to all of the sources in
a source category, or to all of the source categories analyzed. Aggregate
risk is expressed as incidences of cancer among all of the people included
in the analysis, after 70 years of exposure. For statistical convenience,
it is often divided by 70 and expressed as cancer incidences per year.
"Individual risk" applies to the person or persons estimated to live in the
area of the highest ambient air concentrations and it applies to the single
source associated with this estimate as estimated by the dispersion model.
Individual risk is expressed as "maximum lifetime risk" and reflects the
probability of getting cancer if one were continuously exposed to the
estimated maximum ambient air concentration for 70 years.
4.1.2 The Calculation of Aggregate Risk
Aggregate risk is calculated by multiplying the total exposure produced
by HEM (for a single source, a category of sources, or all categories of
sources) by the unit risk estimate. The product is cancer incidences among
the included population after 70 years of exposure. The total exposure,
as calculated by HEM, is illustrated by the following equation:
N
Total Exposure = Z (P-jCj)
-------
75
where
£ = summation over all grid points where exposure is calculated
Pi = population associated with grid point i,
Cj = long-term average inorganic arsenic concentration at grid point i,
N = number of grid points to 2.8 kilometers and number of ED/BG
centroids between 2.8 and 50 kilometers of each source.
To more clearly represent the concept of calculating aggregate risk, a
simplified example illustrating the concept follows:
EX/MPLE
This example uses assumptions rather than actual data and uses only
three levels of exposure rather than the large number produced by HEM. The
assumed unit risk estimate is 4.29 x 10'3 at 1 gg/m3 and the assumed
exposures are:
ambient air number of people exposed
concentrations to given concentration
2 ug/m3 1,000
1 ug/m3 10,000
0.5 ug/m3 100,000
The probability of getting cancer if continuously exposed to the assumed
concentrations for 70 years is given by:
concentration unit risk probability of cancer
2 pg/m3 x 4.29 x 10-3 (ug/rn3)'1 = 9 x 10~3
1 ug/m3 x 4.29 x 10'3 " = 4 x 10'3
0.5 ug/m3 x 4.29 x 10'3 " = 2 x 10'3
-------
76
The 70 year cancer incidence among the people exposed to these concentrations
is given by:
probability of cancer number of people at after 70 years
at each exposure level each exposure level of exposure
9 x
4 x
2 x
10-3
10~3
10-3
x
x
X
1
10
100
,000
,000
,000
9
40
200
TOTAL = 249
The aggregate risk, or total cancer incidence, is 249 and, expressed
as cancer incidence per year, is 249 * 70, or 3.6 cancers per year. The
total cancer incidence and cancers per year apply to the total of 111,000
people assumed to be exposed to the given concentrations.
4.1. 3 The Calculation of Individual Risk
Individual risk, expressed as "maximum lifetime risk," is calculated
by multiplying the highest concentration to which the public is exposed, as
reported by HEM, by the unit risk estimate. The product, a probability of
getting cancer, applies to the number of people which HEM reports as being
exposed to the highest listed concentration. The concept involved is a
simple proportioning from the 1 ug/m3 on which the unit risk estimate is
based to the highest listed concentration. In other words:
maximum lifetime risk the unit risk estimate
highest concentration to 1 ug/m3
which people are exposed
4.2 Risks Calculated for Emissions of Inorganic Arsenic
The explained methodologies for calculating maximum lifetime risk and
cancer incidences were applied to each plant, assuming a baseline level of
emissions. A baseline level of emissions means the level of emissions after
-------
77
the application of controls either currently in place or required to be in
place to comply with current state or Federal regulations but before application
of controls that would be required by a NESHAP.
Tables 43-49 summarize the calculated risks for each source category.
To understand the relevance of these numbers, one should refer to the
analytical uncertainties discussed in section 5 below. Note that the annual
incidence is not calculated for cotton gins. As mentioned earlier in this
document, it was impractical to identify and locate all the gins handling
arsenic-acid-desiccated cotton ( = 300 gins). The Agency does not have enough
available data to provide an estimate of annual cancer incidence that would
be comparable in accuracy to the other source category estimates. As outlined
in Section 3.7, three model gins operating at each of four production rates
were used to establish a range of exposure and risk estimates for individual
sources. Likewise, two operating gins in south central Texas were chosen
for ambient air monitoring in order to validate the model plant exposure
estimates. Maximum lifetime risk estimates were calculated for each of the
three model plants (Table 47) and for the two operating gins (Table 48).
-------
78
Table 43
Maximum Lifetime Risk and Cancer Incidence for Primary Lead Smelters
(Assuming Baseline Controls)
Maximum
Lifetime Cancer Incidences
P1ant Risk Per Year
1 2 x 10-3 0>Q13
2 4 x 10-5 0>Q44
3 2 x 10-5 0
-------
79
Table 44
Maximum Lifetime Risk and Cancer Incidence for Secondary Lead Smelters
(Assuming Baseline Controls)
Plant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Maximum
Lifetime
Risk
5 x 10-6
4 x ID'6
8 x 10-6
8 x 10~6
8 x ID'5
1. 1 x 10-5
5 x ID'6
1. 1 x 10-4
2 x 10~7
1.6 x 10~5
9 x 10~6
4 x Kr6
1.1 x 10"5
4 x ID'6
6 x lO-5
9 x lO"6
4 x 10-6
3 x 10-4
8 x 10-6
4 x ID'5
4 x 10-4
2 x 10-4
4 x 10 ~4
2 x ID'5
7 x ID'5
3 x lO'4
2 x 10-6
1.2 x ID'5
1.7 x 10-6
1. 1 x 10~6
1.5 x 10-5
6 x ID"6
1.5 x lO-5
3 x lO'5
2 x lO-5
Cancer Incidences
Per Year
0.0019
0.0009
0.0010
0.0011
0.014
0.0010
0.0013
0.015
0.0002
<0.0001
0.0010
0.011
0.0019
0. 0009
0.0024
0.0004
0.0035
0.040
0.0031
0.012
0.14
0.028
0.035
0. 0040
0.0069
0.015
0.0002
0.0007
0.0002
0.0001
0.0015
0.0050
0. 0095
0.0013
0.0062
-------
80
Table 45
Maximum Lifetime Risk and Cancer Incidence for Primary Zinc Smelters
(Assuming Baseline Controls)
Plant
Maximum
Lifetime
Risk
Cancer Incidences
Per Year
1
2
3
4
5
8 x lO-6
1.9 x ID'7
1.1 x ID'6
9 x 10-7
3 x ID'6
0.0029
0.0001
0.0010
0.0002
0.0001
-------
81
Table 46
Maximum Lifetime Risk and Cancer Incidence for Zinc Oxide Plants
(Assuming Baseline Controls)
Maximum
Lifetime Cancer Incidences
Plant Risk Per Year
1 4 x 10-7 0.005
2 1.2 x 10'-3 0.077
-------
82
Table 47
Maximum Lifetime Risk and Cancer Incidence for Model Cotton Gins
(Assuming Baseline Controls)
Model
Plant
Hutto, TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Buckholts.TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Itasca,TX
4 Bales/Hr
7 Bales/Hr
12 Bales/Hr
20 Bales/Hr
Maximum
Lifetime
Risk
1.1 x lO-5
3 x ID'5
5 x ID'5
1.0 x ID'4
1.1 x ID'5
3 x 10-5
5 x ID'5
1.0 x 10-4
5 x 10-6
1.2 x lO-5
2 x ID'5
4 x 10-5
Cancer Incidences
Per Year
O.OOOl
0.0001
0.0001
0.0002
<0.0001
<0.0001
<0.0001
0.0001
<0.0001
<0.0001
<0.0001
0.0001
-------
83
Table 48
Lifetime Risk for Two Texas Cotton Gins
(Assuming Baseline Controls)
Plant Maximum Lifetime Risk
A 5 x HT4*
R 1.0 x 10-4
* Represents final risk estimate as incorporated
by EPA.
-------
84
Table 49
Maximum Lifetime Risk and Cancer Incidence for Arsenic Chemical Plants
(Assuming Baseline Controls)
Plant
1
2
3
4
5
6
7
8
Maximum
Lifetime
Risk
4 x lO'8
7 x ID'9
3 x 10-8
3 x 10-8
2 x 1CT4
3 x ID'8
9 x 1CT10
3 x lO-8
Cancer Incidences
Per Year
<0.0001
<0.0001
<0.0001
<0.0001
0.0042
<0.0001
<0.0001
<0.0001
-------
85
5 ANALYTICAL UNCERTAINTIES APPLICABLE TO THE CALCULATIONS OF PURLIC
HEALTH RISKS CONTAINED IN THIS DOCUMENT
5.1 The Unit Risk Estimate
The procedure used to develop the unit risk estimate is described in
referenced. The model used and its application to epidemiological data
have been the subjects of substantial comment by health scientists. The
uncertainties are too complex to be summarized sensibly in this appendix.
Readers who wish to go beyond the information presented in the reference
should see the following Federal Register notices: (1) OSHA's "Supplemental
Statement of Reasons for the Final Rule", 48 FR 1864 (January 14, 1983);
and (2) EPA's "Water Quality Documents Availability" 45 FR 79318 (November
28, 1980).
The unit risk estimate used in this analysis applies only to lung
cancer. Other health effects are possible; these include skin cancer,
hyperkeratosis, peripheral neuropathy, growth retardation and brain
dysfunction among children, and increase in adverse birth outcomes. No
numerical expressions of risks relevant to these health effects is included
in this analysis.
Although the estimates derived from the various studies are quite
consistent, there are a number of uncertainties associated with them. The
estimates were made from occupational studies that involved exposures only
after employment age was reached. In estimating risks from environmental
exposures throughout life, it was assumed through the absolute-risk model
that the increase in the age-specific mortality rates of lung cancer was a
function only of cumulative exposures, irrespective of how the exposure was
accumulated. Although this assunption provides an adequate description of
-------
86
all of the data, it may be in error when applied to exposures that begin
very early in life. Similarly, the linear models possibly are inaccurate
at low exposures, even though they provide reasonable descriptions of the
experimental data.
The risk assessment methods employed were severely constrained by the
fact that they were based only upon the analyses performed and reported by
the original authorsanalyses that had been performed for purposes other
than quantitative risk assessment. For example, although other measures of
exposure might be more appropriate, the analyses were necessarily based
upon cumulative dose, since that was the only usable measure reported. Given
greater access to the data from these studies, other dose measures, as well
as models other than the sinple absolute-risk model, could be studied. It
is possible that such wide analyses would indicate that other approaches
are more appropriate than the ones applied here.
5.2 Public Exposure
5.2.1 General
The basic assumptions implicit in the methodology are that all exposure
occurs at people's residences, that people stay at the same location for 70
years, that the ambient air concentrations and the emissions which cause
these concentrations persist for 70 years, and that the concentrations are
the same inside and outside the residences. From this it can be seen that
public exposure is based on a hypothetical premise. It is not known whether
this results in an over-estimation or an underestimation of public exposure.
-------
87
5.2.2 The Public
The following are relevant to the public as dealt with in this analysis:
1. Studies show that all people are not equally susceptible to cancer.
There is no numerical recognition of the "most susceptible" subset of the
population exposed.
2. Studies indicate that whether or not exposure to a particular
carcinogen results in cancer may be affected by the person's exposure to
other substances. The public's exposure to other substances is not
numerically considered.
3. Some members of the public included in this analysis are likely to
be exposed to inorganic arsenic in the air in the workplace, and workplace
air concentrations of a pollutant are customarily much higher than the
concentrations found in the ambient, or public air. Workplace exposures
are not numerically approximated.
4. Studies show that there is normally a long latent period between
exposure and the onset of lung cancer. This has not been numerically
recognized.
5. The people dealt with in the analysis are not located by actual
residences. As explained previously, people are grouped by census districts
and these groups are located at single points called the population centroids
The effect is that the actual locations of residences with respect to the
estimated ambient air concentrations are not known and that the relative
locations used in the exposure model may have changed since the 1980 census.
However, for the population sectors estimated to be at highest risk, U.S.
-------
88
Geological Survey topographical maps were checked to verify that people did
live or could live in locations near the sources as modeled predictions
estimated. Maps in certain instances were old and the possibility could
not be excluded that additional areas near sources have been developed
since publication of the maps.
6. Many people dealt with in this analysis are subject to exposure to
ambient air concentrations of inorganic arsenic where they travel and shop
(as in downtown areas and suburban shopping centers), where they congregate
(as in public parks, sports stadiums, and schoolyards), and where they work
outside (as mailmen, milkmen, and construction workers). These types of
exposures are not numerically dealt with.
5.2.3. The Ambient Air Concentrations
The following are relevant to the estimated ambient air concentrations
of inorganic arsenic used in this analysis:
1. Flat terrain was assumed in the dispersion model. Concentrations
much higher than those estimated would result if emissions impact on elevated
terrain or tall buildings near a plant.
2. The estimated concentrations do not account for the additive impact
of emissions from plants located close to one another.
3. The increase in concentrations that could result from re-entrainment
of arsenic-bearing dust from, e.g., city streets, dirt roads, and vacant
lots, is not considered.
-------
89
4. Meteorological data specific to plant sites are not used in the
dispersion model. As explained, HEM uses the meteorological data from the
STAR station nearest the plant site. Site-specific meteorological data
could result in significantly different estimates, e.g., the estimated
location of the highest concentrations.
5. tn some cases, the arsenic emission rates are estimates that are based
on assumptions rather than on measured data.
-------
90
6 REFERENCES
1. National Academy of Sciences, "Arsenic," Committee on Medical and
Biological Effects of Environmental Pollutants, Washington, D.C., 1977.
Docket Number (OAQPS 79-8) II-A-3.
2. Health Assessment Document for Inorganic Arsenic - Final Report EPA-600/
8-83-021F March 1984, OAQPS Docket Number OAQPS 79-8, II-A-13.
3. U.S. EPA, et.al., "Environmental Cancer and Heart and Lung Disease,"
Fifth Annual Report to Congress by the Task Force on Environmental Cancer
and Health and Lung Disease, August, 1982.
4. OAQPS Guideline Series, "Guidelines on Air Quality Models". Publication
Number EPA-450/2-78-027, (OAQPS Guideline No. 1.2-080).
5. Systems Application, Inc., "Human Exposure to Atmospheric Concentrations
of Selected Chemicals." (Prepared for the U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina). Volume I, Publication
Number EPA-2/250-1, and Volume II, Publication Number EPA-1/250-2.
6. NEA, Inc., "East Helena Source Apportionment Study Particulate Source
Apportionment Analysis Using the Chemical Mass Balance Receptor Model."
(Prepared for the Department of Health and Environmental Sciences, State
of Montana.) Volume I, September, 1982.
7. RADIAN Corporation, "Preliminary Study of Sources of Inorganic Arsenic."
(Prepared for the U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina.) Publication Number EPA-450/5-82-005, August 1982.
-------
-------
United States Office of Air and Radiation
Environmental Protection Office of Air Quality Planning and Standards
A9encV Research Triangle Park, NC 27711
Official Business
Penalty for Private Use
$300
If your address is incorrect, please change on the above label;
tear off; and return to the above address.
If you do not desire to continue receiving this technical report
series, CHECK HERE D; tear off label; and return it to the
above address.
------- |