EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
EPA-452/R-95-001
March 1995
Air
SUMMARY OF URBAN AIR TOXICS
RISK ASSESSMENT SCREENING STUDIES
TO SUPPORT THE
URBAN AREA SOURCE PROGRAM { i
I L' > . ' .4 s\ '•
-------
EPA-452/R-95-001
MARCH, 1995
SUMMARY OF URBAN AIR TOXICS RISK ASSESSMENT
SCREENING STUDIES TO SUPPORT THE URBAN AREA
SOURCE PROGRAM
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
This report has been reviewed by the Office of Air Quality Planning
and Standards, U. S. Environmental Protection Agency, and has been
approved for publication. Mention of trade names or commercial
products does not constitute endorsement or recommendation for use.
Any discussion of methods and results from studies not performed by
the U.S. EPA does not necessarily constitute EPA acceptance of
these studies. All studies and results are excerpted as faithfully
as possible, in this report, for comparison purposes only.
-------
Table of Contents
Chapter 1: Summary 1
Chapter 2: Methodology 9
2.1 Background 9
2.2 Compilation of Available Exposure and Risk Screening Studies 9
2.3 Methodology 10
2.4 Additional Study Results 10
2.5 Screening Study Assumptions and Limitations 10
Chapter 3: Summaries of New Studies 21
3.1 Southwest Chicago Study 21
3.2 lEMP-Baltimore II Study 29
3.3 The Detroit Study 41
3.4 The Houston Study 49
3.5 The Twin-City Study 52
3.6 The Texas Study 59
3.7 The EPA Noncancer Screening Study 63
References 69
List of Figures and Tables
Table
1-1 Normalized Cancer Incidence by Pollutant for Each Study 3
1-2 Normalized Cancer Incidence by Source Category for Each Study 4
1-3 Average Cancer Incidence by Pollutant Across Studies 5
1-4 Average Cancer Incidence by Source Category Across Studies 6
1-5 Pollutants Identified for Noncancer Health Effects 7
2-1 Cancer Incidence by Pollutant and Source Category for Each Study 14
3-1 Aggregate Hazard Indices by Source Category 23
3-2 Aggregate Hazard Indices by Pollutants 24
3-3 Top Five Source Contributors to Cancer Cases 25
3-4 Top Four Pollutant Contributors to Cancer Cases 26
3-5 Relative Ranking of Target Compounds by Cancer Potency-Weighted
Ambient Concentrations 31
3-6 Relative Ranking of Sources by Contribution to Cancer Potency-Weighted
Ambient Concentrations 32
3-7 Average Increased Lifetime Individual Cancer Risk Using Available
Monitoring Data 33
3-8 Estimated Annual Excess Cancer Incidence for Selected Pollutants Modelled
in the Baltimore IEMP Air Toxics Study 35
3-9 Area-Wide Annual Excess Cancer Incidence Using Available Monitoring
Data 36
3-10 Receptor Locations Warranting Further Investigation for Noncancer Effects:
Pollutant-Specific 37
-------
Table of Contents (continued..)
Table
3-11 Receptor Locations Warranting Further Investigation for Noncancer Effects:
Complex Pollutant Mixtures 38
3-12 A Comparison of Predicted Average Lifetime Individual Cancer Risk in
Baltimore with Other Studies 39
3-13 Summary of Source Categories 42
3-14 Summary of Estimated Excess Cancer Cases by Pollutant Across the Study Area 44
3-15 Pollutant Contribution to Risk at Grid Cell with Highest Individual Risk 47
3-16 Source Grouping Contribution to Grid Cell with Highest Individual Risk 48
3-17 Upper Bound Annual Predicted Cancer Cases in Harris County for a
Population of 2,055,066 51
3-18 Pollutants Included in Inventory and Their Unit Risk Estimates 53
3-19 Results of the Texas Urban Air Toxics Assessment 61
3-20 Predicted Versus Measured Concentrations of Compounds Evaluated in
Harris County . ; '. 62
3-21 Urban County Study Results of Short Term (Screen) Analysis 66
Figure
3-1 Estimated Excess 70-Year Incidence, Area Wide by Pollutant Contribution 45
3-2 Estimated Excess 70-Year Incidence, Area Wide by Source Grouping Contribution .... 46
3-3 Sources of Carcinogenic Pollutant Emissions in the 7-County Twin Cities
Metropolitan Area 55
3-4 Emissions of Carcinogenic Pollutants in the 7-County Twin Cities
Metropolitan Area 56
3-5 Estimated Excess Cancer Incidence by Source Category 57
3-6 Estimated Excess Cancer Incidence by Pollutant 58
3-7 Results of Long-Term (HEM) Modelling Analysis (Exposure from Individual
Facility Emissions) 65
-------
Chapter 1
SUMMARY
Under Section 112(k) and 112(c)(3) of the Clean Air Act Amendments of 1990, the U. S.
Environmental Protection Agency (EPA) must develop and implement a national strategy to control
hazardous air pollutants (HAPs) from area sources. The national strategy must identify 30 or more
HAPs presenting the greatest threat to public health in large urban areas and identify regulatory
actions to achieve significant reductions in area source emissions of these HAPs. Subsequent to
publication of this strategy in 1995, EPA must establish regulations to implement this strategy by the
year 2000.
This report summarizes the air toxics exposure and risk assessment screening studies
conducted in the last decade. It includes studies summarized in the 1990 EPA report entitled Cancer
Risk From Outdoor Exposure to Air Toxics' and includes more current studies performed since 1990.
Generally, the studies summarized in this report represent exposure and cancer risk assessments based
on dispersionmodeling, of emissions inventory data..J5orneof the studies represent assessments based
on ambient monitoring data. Some also address noncancer effects, although generally to a limited
extent. This report is one of several independent approaches to identify HAPs and area sources in the
national strategy. As such, this report does not select specific HAPs for inclusion in the national
strategy.
Tables 1-1 through 1-5 summarize the results of this analysis. Table 1-1 presents the
areawide cancer incidence associated with each study HAP and also rank orders the study HAPs by
areawide cancer incidence. Table 1-2 presents the areawide cancer incidence associated with each
source category and also rank orders the categories by areawide cancer incidence. For inter-city
comparison purposes, cancer incidence is normalized per million population residing in each study
area. Areawide, annual excess cancer incidence reported in the eleven studies ranges from about 0.2
to 17 cases per year^>er 'million population, with values between 1 and 5 most common. The HAPs
consistently ranking highest are polycyclic organic matter (POM), benzene, hexavalent chromium,
formaldehyde, 1,3-butadiene, ethylene oxide, arsenic, and cadmium. Gasoline vapor also ranked
relatively high in some studies, but is not a listed HAP under Section 112.
None of the cancer screening studies distinguishes between major and area source categories
per the Section 112(a) definition, which is based on an emission threshold of 10 tons/year (TPY) of
any single HAP or 25 TPY of any combination of HAPs. This is because most of the studies were
conducted before enactment of the Clean Air Act Amendments of 1990, wherein this major/area
source distinction became relevant within the context of Section 112. Hence, area sources of HAPs
could not clearly be identified or ranked by cancer incidence based on these study results, pursuant to
this 10/25 TPY threshold.
Table 1-3 presents cancer incidence averaged across all eleven screening studies, as well as
the maximum and minimum reported incidence and the number of studies reporting an incidence for a
particular HAP. Table 1-4 presents this same information, but by source category.
Table 1-5 summarizes the HAPs indicated as potential concerns based on noncancer health
endpoints, from three screening studies. Because of different endpoints identified in each study for
each pollutant, no quantitative measure of noncancer serves as a convenient common denominator for
comparison purposes. Because of the problem of comparing differing noncancer endpoints, and
-------
because of the considerable uncertainty in these noncancer screening studies, no ranking of HAPs is
made in this report. Rather, the HAPs are simply identified in Table 1-5 for which each respective
study felt there may be some cause for concern. Many of these HAPs are also carcinogens.
Inclusion of any study results herein does not indicate acceptance or validation of the study
methodology or data by EPA. The limitations associated with any of the studies are presented in this
report, and will be considered in the development of the national area source strategy. Clearly, some
of the methods employed in these studies vary widely because of age and the data and resources
available when a study was performed. No attempt was possible in this report to identify and correct
for any methods or data that may have changed since the original study was carried out.
-------
Table 1-1. Normalized Cancer Incidence by Pollutant for Each Study
(excess annual cancer cases per million population within study area)
/EMP- Mod*/*4
-------
Table 1-2. Normalized Cancer Incidence by Source Category for Each Study
(excess annual cancer cases per million population within study area)
IfMP- Modeled data:
Kanawha IfMP- IfMP- Soultiocnt Southwell ICMP-
Source Category HveCtty-3' Valley* Ptiltadotphla-S Santo Ctara-6 South Coasl-t Chlcago-9 TwlnCllr-17 Chlcogo-IO Baltimore 11-7 Houtton-ll Detroll-12
Aircraft Engines
Asbestos/Demolition
Background Carbon Tettachtorlde Concenlrotfons
3afge loading
Chemical Manufacturing
Chemical Users and Producers
Chrome Platers
Cool and OH Combustion/Heating
Commercial Incinerators
Commercial Sterilization/Hospitals
Cooling Towers
Delaware River
Dry Cleaning
Gasoline Marketing
Heating/Combustion
Heatlng/Woodsloves
Industrial Solvent Coating
Iron and Steel
Manufacturing Operations
Motor Vehicles
Motor Vehicles/Diesel
Motor Vehicles/Gasoline
Municipal Waste Combuslors
Municipal Waste Landfill
Nonferrous Smelters
Nonrcod Engines
Other
Other Miscellaneous Area Sources
Other Organic Evaporation
Paint and 'Other Stripping
Pesticide Usage
Petroleum Refining
Point Sources
POTWs
Refractories
Secondary Formaldehyde Formation
Sewer Volatilization
Solvent Use/Degreasing
Specialty Steel
Surface Coating
ISDF's
Unspecified Stationary Sources
Waste Oil Combustion/Burning
Study Total
(JO) 0.04
(20) 0.002
(3) 0.5
(15) 0.02
(6) O.I
(3) 0.5
(18) 0.01
(?) 0.07
(13) 0.03
(5) 0.4
(10) 0.04
(1) 2.7
(18) 0.0)
(6) O.I
(15) 0.02
(15) 0.02
(13) 0.03
(2) 0,6
(10) 0.04
(21) 0.00006
(6) O.I
5.3
(1) 11
(3) 1.9
(4) 0.2
(6) 0.02
(6) 0.02
(6) 0.02
(2) 3.7
(5) 0.05
10) 0.0002
(9) 0.005
16.9
(9) 0.00002
(4) 0.01
(4) 0.01
(3) 0.03
(1) 0.05
(1) 0.05
(4) 0.01
(4) 0.01
(8) 0.0002
0.2
(8) 0.002
(5) 0.007
£3) 0.02
(2) 0.04
(1) O.I
(9) 0.0005
(7) 0.005
(4) 0.01
(5) 0.007
0.2
(2) 0.6
(1) 1.0
1.6
(6) 0.2
;i6) 0.001
(3) 0.5
;iO) o.oi
(7) 0.06
(8) 0.03
;ii) 0.008
(5) 0.3
(1) 1.0
(2) 0.6
;i3) 0.004
(13) 0.004
(4) 0.4
:i_2) 0.007
;I5) 0.002
117) 0.0007
(9) 0.02
3.1
(3) 0.3
(5) 0.08
(4) 0.2
(2) 0.4
(1) 1.6
(6) 0.05
(7) 0.03
2.7
(4) 0.3
(20) 0.001
(6) 0.2
(24J 0.00003
(2) 0.5
(14) 0.01
(18) 0.007
(8) 0.07
(17) 0.008
(11) 0.04
(12) 0.02
(22) 0.0007
(6) 0.2
(1) 0.8
(22) 0.0007
(4) 0.3
(10) . 0.06
(12) 0.02
(19) 0.002
(8) 0.07
(14) 0.01
(3) 0.4
(14) 0.01
(20)_ 0.001
3.0
(3) 0.2
(8) 0.02
;10) 0.006
(5) 0.09
(8) 0.02
(6) 0.08
(2) 0.7
J71 0.05
0) i.o
(4) 0.1
2.3
(3) 0.003
(1) 11
(4) 0.0002
(2) 0.2
1.3
(2) 0.2
(6) 0.08
(8) 0.001
(2) 0.2
(7) 0.02
(4) O.I
(4) 0.1
(0 0.5
1.2
Caveat: Estimates of cancer Incidence In these screening studies are not absolute predictions of cancer occurrence and are Intended to be used In o relative sense only. See Section 2.5 for a discussion
of the many assumptions associated with these estimates.
NOTES:
(t) Numbers In parentheses represent the ordinal ranking of source categories by cancer Incidence for each study.
(2) Blank cells Indicate o source category was not analyzed for that particular study.
(3) If multiple source categories have the same normalized cancer Incidence they are assigned the same ordinal ranking and subsequent numbers are assigned their true ordinal rank. For example,
V two source categories have on ordinal rank of 4. the next highest source category would be assigned on ordinal rank of 6.
'Source of Information by reference number
-------
Table 1 -3. Average Cancer Incidence by Pollutant Across Studies
(excess annual cancer cases per million population within study area)
Average Cancer
HAP Incidence
1 ,2-Dibromoethane'
1 ,2-Dichloroethane
1 ,3-Butadiene
Acetaldehyde
Acrylamide
Acrylonitrile
Arsenic, inorganic
Asbestos (friable)
Benzene
Benzo (a) pyrene
Beryllium
Cadmium
Carbon tetrachloride
Chloroform
Chromium (VI)
Coke oven emissions
Dichloroethene
Dioxins
Epichlorohydrin
Ethylene oxide
Formaldehyde
Gasoline vapor
Hexachlorobenzene
Methyl chloride
Methylene chloride
Perchloroethylene
Polychlorinated biphenyls
Polycyclic organic matter
Propylene dichloride
Propylene oxide
Styrene
Trichloroethylene
Vinyl chloride
0.004
0.04
1
0.000001
O.OOOOJ
(0.5)
0.08
0.01
0.3
0.04
0.007
0.02
0.1
0.5
0.6
0.4
0.005
0.01
0.000007
1
0.4
0.05
0.02
0.005
0.04
0.4
0.0002
0.8
0.1
0.000007
0.0001
0.01
0.001
Maximum Cancer
Incidence
0.01
0.2
4.4
0.000001
0.00001
1.9
0.2
0.02
1.1
0.04
0.02
0.04
0.2
1.9
1.7
0.9
0.01
0.04
0.00001
8.4
1
0:1
0.02
0.01
0.1
3.4
0.0003
1.8
0.1
0.000007
0.0003
0.02
0.002
Minimum Cancer
Incidence
0.001
0.0008
0.2
0.000001
0.00001
0.000003
0.02
0.005
0.008
0.04
0.0000009
0.005
0.03
0.00001
0.05
0.2
0.0006
0.0004
0.000003
0.000003
0.06
0.01
0.02
0.0002
0.0003
0.0007
0.000004
0.03
0.1
0.000007
0.000005
0.001
0.0008
Number of Studies
Reported
7
8
7
1
1
4
8
2
12
1
3
9
5
5
8
3
2
3
2
6
7
5
1
2
7
9
2
7
1
1
3
9
3
Caveat: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are
intended to be used in a relative sense only. See Section 2.5 for a discussion of the many assumptions and uncertainties
associated with these estimates.
-------
Table 1 -4. Average Cancer Incidence by Source Category Across Studies
(excess annual cancer cases per million population within study area)
Average Cancer
Source Category Incidence
Aircraft Engines
Asbestos/Demolition
Background Carbon Tetrachloride Cone.
Barge Loading
Chemical Manufacturing
Chemical Users and Producers
Chrome Platers .
Coal and Oil Combustion/Heating
Commercial Incinerators
Commercial Sterilization/Hospitals
Cooling Towers
Delaware River
Dry Cleaning
Gasoline Marketing
Heating/Combustion
Heating /Woodstoves
ndustrial Solvent Coating
Iron and Steel
Manufacturing Operations
Motor Vehicles
Motor Vehicles/Diesel
Motor Vehicles/Gasoline
Municipal Waste Combustors
Municipal Waste Landfill
Nonferrous Smelters
Nonrood Engines
Other
Other Miscellaneous Area Sources
Other Organic Evaporation
Paint and Other Stripping
Pesticide Usage
Petroleum Refining
Point Sources
POfWs
Refractories
Secondary Formaldehyde Formation
Sewer Volatilization
Solvent Use/Degreasing
Specialty Steel
Surface Coating
TSDF's
Unspecified Stationary Sources
Waste Oil Combustion/Burning
0.3
0.001
0.2
0.00003
2.2
1.0
0.5
0.1
0.08
0.04
0.2
0.01
0.01
0.03
0.06
0.2
0.04
0.4
0.02
1.1
0.08
0.7
0.0006
0.0007
0.01
0.3
0.2
0.02
0.01
0.002
0.01
0.04
0.3
0.03
0.03
0.5
0.01
0.04
0.001
0.001
0.0007
0.2
0.005
Maximum Cancer
Incidence
0.3
0.001
0.2
0.00003
11
1.9
0.5
0.2
0.08
0.1
0.5
0.01
0.02
0.07
0.2
0.4
0.04
1.0
0.02
3.7
0.08
0.7
0.0005
0.0007
0.01
0.3
1.1
0.05
0.02
0.002
0.01
0.05
1.0
0.05
0.03
0.6
0.01
0.1
0.002
0.001
0.0007
1.0
0.005
Minimum Cancer
Incidence
0.3
0.001
0.2
0.00003
0.00002
0.002
0.3
0.01
0.08
0.007
0.06
0.01
0.008
0.006
0.008
0.0007
0.04
0.04
0.02
0.003
0.08
0.7
0.0005
0.0007
0.01
0.3
0.004
0.0002
0.005
0.002
0.01
0.02
0.03
0.01
0.03
0.4
0.01
0.007
0.00006
0.001
0.0007
0.0002
0.005
Number of Studies
Reported
1
1
3
1
5
2
4
5
•1
3
3
1
4
7
7
7
1
4
1
9
1
1
1
1
1
1
6
4
2
1
1
2
4
2
1
4
1
6
2
1
]
6
1
Caveat: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are
intended to be used in a relative sense only. See Section 2.5 for a discussion of the many assumptions and
uncertainties associated with these estimates.
-------
EPA Non-concer Screening Study (13)'
" I
Urban County Urban County
-• ' et* — *_
-------
Table 1-5. Pollutants Identified (or Potential Noncancer Health Effects (cont' d)
Pollutant
o-Xylene
Perchloroethylene
Phenol
Phthalic anhydride
Styrene
Sulfur "
Tetramethyl lead
Toluene
Toluene diisocyanote
Vinyl chloride
Xylene
EPA Non-cancer Screening Study (13)*
Broad Screening
Study
X
X
X
X
X
X
X
Urban County
Study - Long-
term
Urban County
Study - Short-
term
X
X
X
X
X
X
X
X
Texas Study (16)
Dallof-Ft. Worth
Harris County
IEMP Baltimore
II Study (7)
= pollutant identified for noncancer health risks
=source of information by reference number
= not listed as a HAP under Section 112
-------
Chapter 2
METHODOLOGY
2.1 Background
The purpose of this report is to compile and report the results of available exposure and risk
screening studies dealing with air toxic emissions. Numerous such studies have been carried out over
the past decade, and formed the basis for concern that led to an urban area source regulatory program
in Section 112 of the Clean Air Act Amendments of 1990.2 Such studies are generally considered to
be broad screening studies, with many aggressive, often very conservative, assumptions necessarily
having to be made to develop a comprehensive overview of the nature and magnitude of the air toxics
problem in the United States. Assumptions and limitations of these studies are presented in Section
2.5. Generally, these screening studies represent exposure and cancer risk assessments based on
dispersion modeling of emissions inventory data. Some of the studies summarized also factor in
assessments based on ambient monitoring data. Only a few screening studies consider noncancer
endpoints.
2.2 Compilation of Available Exposure and Risk Screening Studies
The following screening studies were compiled for this analysis:
• EPA five city controllability study"
• EPA Integrated Environmental Management Project (IEMP) studies in
KanawhaValley, WV4', Philadelphia.^5", Santa Clara, CA6', and
Baltimore, MD7
• South Coast, CA air toxics study8"
• Southeast Chicago, IL9" and Southwest Chicago, IL10 air toxics studies
Houston, TX air toxics risk assessment"
• Detroit, MI Transboundary air toxics study12
• EPA noncancer air toxics screening study13"15
• Texas air toxics assessment16
• Twin City cancer risk study17
All of these studies are cancer screening studies with the exception of the EPA noncancer study, the
Texas air toxics assessment and the Baltimore IEMP, the latter including both a noncancer and cancer
assessment component. The studies identified above with asterisks (*) were conducted before 1990
and summarized in EPA's 1990 EPA report entitled Cancer Risk From Outdoor Exposure to Air
Toxics'. This analysis utilized the results of these studies as summarized in the Cancer Risk... report,
and did not revisit the original studies. This was done because the Cancer Risk... authors revisited the
original studies and in some cases updated the original results to reflect newer health benchmarks.
-------
The studies identified above without asterisks were not included in the Cancer Risk... report, mainly
because they were completed subsequent to 1990 or were unavailable for inclusion in the Cancer
Risk... report. These additional screening study results were reported here without any attempt to
update any of the methods or data utilized in them. This is very difficult to do and available
resources precluded any such update.
2.3 Methodology
Each of the studies was reviewed and the cancer and noncancer incidence and health
benchmark exceedances were summarized. In most of the studies, the summary data needed were
directly excerpted for relative ranking purposes in this report. Where population-normalized cancer
incidence data were not presented, the aggregate, areawide cancer incidence values were divided by
the study area population to express cancer incidence per million population, for inter-city comparison
purposes. Where cancer incidence data were not presented on an annual basis, the lifetime incidence
figures were divided by 70 to adjust accordingly.
Most of the noncancer studies presented a listing of the extent or occurrence of monitored
and/or modeled ambient concentrations of HAPs exceeding some specific health benchmarks. These
specific health benchmarks varied by study, and included lowest-observed-adverse-effect-levels
(LOAEL), health-effect-levels (HEL) determined by dividing the LOAEL by an uncertainty factor,
reference concentrations (RfC), or threshold levels set by the investigators. The exposures could be
acute or chronic. Given the complex and nonuniform presentations in the noncancer risk studies, it is
inappropriate to compare the pollutants and their noncancer effects on a common basis. Therefore,
this report only lists those pollutants in each of the studies that are identified as potentially causing a
noncancer concern, but does not rank those pollutants in any prioritized order.
Salient excerpts of the results from each of these screening studies are provided in the
following chapter, along with a brief summary of the purpose, methodology and sponsoring agency.
Also presented along with these excerpts are the specific assumptions and limitations known to be
associated with each screening study.
2.4 Additional study results
The results of this analysis are summarized in Tables 1-1 through 1-5 in Chapter 1.
Additional, study-specific results are presented in Table 2-1. This table summarizes the specific
cancer incidence associated with each source category-HAP combination, in descending order of
importance, by study.
2.5 Screening study assumptions and limitations
All screening studies to date have made many assumptions regarding
emissions/exposures/risks in order to be broadly comprehensive in the face of uncertain, changing
assessment methodologies and weak, gap-laden supporting data. In fact, most of the studies have
made similar assumptions, used similar assessment methodologies, and accounted for data gaps in
similar ways. Sometimes these assumptions have involved default options in instances where specific
data were altogether missing.
10
-------
Typical assumptions and limitations commonly associated with these screening studies are
recited below. These assumptions and limitations are generally excerpted verbatim directly from
References 1 and 3, and are considered fairly representative of all studies summarized in this report,
and of such broad screening studies, in general.
• "Personal exposure to air toxics was estimated using annual-average concentration estimates
and it was assumed that exposures occur where people reside. In addition, only outdoor
exposures were modeled. Thus, this methodology ignores people's movements throughout the
urban area, travel outside the urban area, indoor exposures and seasonal or diurnal variations
in emissions."
• "The study relied solely on quantitative estimates of cancer risk associated with inhalation of
ambient air over long periods. Acute and subchronic effects were not included, and cancer
cases associated with exposure routes other than inhalation of ambient air were not
quantified."
• "Only 25 compounds were explicitly included in this study, although monitoring studies have
shown that urban atmospheres typically obtain additional carcinogenic pollutants. The
compounds selected for study were chosen because they were estimated to be the most
important contributors to excess cancer incidence."
• "This study focused on routine, continuous emissions. Accidental releases were not
modeled."
• "Unit risk factors employed in this study represent the chance of contracting cancer from a
lifetime (70 years) exposure to a given concentration of that pollutant. It was assumed that
the resulting lifetime incidence levels could be divided by 70 to represent annual incidence
levels. The carcinogenic potency estimates used in this study were developed by EPA's
Carcinogen Assessment Group, and generally represent conservative (upper bound)
dose/response relationships."
• "Unit risk factors used in this study have been generated, in most instances, using EPA
approaches or models. Most of the resulting unit risk factors are generally regarded either as
plausible, upper-bound estimates or as maximum likelihood estimates. The linearized
multistage procedure used to derive these factors leads to a plausible upper limit to the risk
that is consistent with some proposed mechanisms of carcinogenesis. Such estimates,
however, do not necessarily give a realistic prediction of the risk. The true value of the risk
is unknown, and may be as low as zero."
• "Cancer incidence estimates are presented for "existing" conditions (1980). These incidence
estimates are based on the assumption that emission levels for each scenario remain constant
for a 70-year period. In reality, emissions will vary from year to year."
• "Sources included in the exposure modeling data set for each study area were limited to those
in the counties under study. Therefore, while contributions from these sources to areas
outside the county boundaries were considered, contributions of sources outside the county
boundaries to air toxic concentrations within the study areas were not."
11
-------
"Except for secondary formaldehyde exposure formation, atmospheric transformation of toxic
compounds and precursors had been ignored. Both secondary formation and scavenging may
occur for the compounds included in this study. Thus, it is difficult to quantify how
neglecting transformation might affect the final results."
"Incidence from formaldehyde exposure was estimated using ambient monitoring data and
assuming that everyone with an urban area is exposed to the same concentration. This is a
relatively crude technique, but because so much of formaldehyde is formed secondarily, this
procedure was judged to be preferable to modeling direct formaldehyde concentrations and
ignoring secondary formation."
"It has been suggested that global background concentrations of some toxic compounds,
notably carbon tetrachloride, may be contributing significantly to observed ambient readings.
No attempt has been made in the dispersion modeling performed for this study to account for
background concentrations."
"The handling of some point sources as area sources for modeling purposes may introduce
some upward bias in the resulting exposure/risk estimates since HEM (EPA's Human
Exposure Model) distributes area source emissions by populations and since area sources are
emitted closer to ground level."
"Any study such as this represents a "snapshot in time" on one's collective understanding of
the urban air toxics problem. In fact, the emission estimates and dose-response relationships
used in this study are subject to frequent revision as newer data become available. Hence,
care should be taken when interpreting any results from this study or comparing these results
to those from other studies where different data have been used."
"All risks are assumed to be additive. This assumption can lead to substantial errors in risk
estimated if synergistic or antagonistic interactions occur. Although dose additivity has been
shown to predict the acute toxicities of many mixtures of similar compounds, some marked
exceptions have been noted. Consequently, additivity assumptions may substantially
overestimate risk in some cases and underestimate it in others. The available data on
mixtures are insufficient for estimating the magnitude of these errors. Based on current
compounds that induce similar types of effects at the same sites of action."
"In the case of chromium, only the hexavalent form has been proven to be carcinogenic. The
percentage of total chromium that is hexavalent is known to vary considerably depending on
the source. For example, hexavalent chromium is less than 1 percent of total chromium
emissions from coal and oil burning combustion, while it is nearly 100 percent of total
chromium emissions from cooling towers and electroplating. Nevertheless, considerable
uncertainty remains as to the exposure to hexavalent chromium versus total chromium
emissions."
"In the case of PIC [products of incomplete combustion], there are several sources of
uncertainty. There are a number of methodologies available to estimate risk from PIC. Some
of these methodologies use BaP as a surrogate for both PIC emissions and unit risk value.
Others use PIC-specific emission factors and unit risk factors or comparative potency factors.
The estimates of cancer incidence are seen to vary by a factor of 200 depending on which
12
-------
methodology is used. While no one methodology has been shown to be a better methodology
for estimating risk from PIC, this study uses the methodology that relies on PIC-specific
emission factors and unit risk factor or comparative potency factors."
• "While the comparative potency approach used to estimate POM risks for several important
sources is judged to be an improvement, conceptually, over previous techniques which used
B(a)P as a surrogate for POM, available comparative potency factors are, in fact, based on
few measurements, especially for the important motor vehicle categories, and the uncertainty
in these values should be recognized."
• Dioxins/furans are only handled in a few studies9-10-12, and the details for
emissions/exposure/risk assessment are not described in these reports. The unit risk factors
were based on a variation of EPA's "Interim Procedures for Estimating Risk Associated with
Exposures to Mixtures of Chlorinated Dibenzo-p-dioxins and Dibenzofurans (CDDs and
CDFs).18
Another limitation not explicitly mentioned in these studies, with respect to using the study
results in the urban area source program, is that none of the cancer screening studies expressly
distinguishes between major and area source categories per the Section 112(a) definition, which
is based on an emission threshold of 10 tons/year (TPY) of any single HAP or 25 TPY of any
combination of HAPs. Hence, area sources of HAPs could not clearly be identified or ranked by
cancer incidence based on these study results, pursuant to this 10/25 TPY threshold.
All of the studies summarized in this report were completed before publication of the National
Academy of Science's (NAS), 1994 report "Science and Judgement in Risk Assessment. 19 The NAS
report was carried out pursuant to the 1990 Clean Air Act Amendments to address important evolving
issues regarding risk assessment of hazardous air pollutants. While this report encourages certain
improvements in risk assessment methods, it does not conclude that assessments cannot be useful even
if many default assumptions are made. Instead, the report recommends that "EPA should continue to
regard the use of default options as a reasonable way to deal with uncertainty about underlying
mechanisms in selecting methods and models for use in risk assessment." However, the NAS also
recommends that EPA should explicitly identify and state the scientific and policy basis for each use
of a default option in risk assessments and, of course, improve its methods where resources and data
make this possible. Hence, it is important to be aware of the many limitations and assumptions in the
screening studies reported herein before considering the results in National Strategy development.
13
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
Pollutant
Source category
Study (Reference)
Cancer Incidence
(cases/million people)
Polycyclic organic matter
1,3-Butadiene
Formaldehyde
Chromium (VI)
Chromium (VI)
Benzene
Polycyclic organic matter
Formaldehyde
Ethylene oxide
Polycyclic organic matter
Benzene
1,2-Dichloroethane
Gasoline vapor
Arsenic, inorganic
Chromium (VI)
Carbon tetrachloride
Formaldehyde
Arsenic, inorganic
1,3-Butadiene
Polycyclic organic matter
Benzene
Methylene chloride
Benzene
Benzene
Formaldehyde
Benzene
Perchloroethylene
Formaldehyde
Perchloroethylene
Chloroform
Trichloroethylene
Formaldehyde
Ethylene oxide
Cadmium
Formaldehyde
Benzene
Formaldehyde
Polycyclic organic matter
Ethylene oxide
Perchloroethylene
Chloroform
1,2-Dichloroethane
Arsenic, inorganic
Vinyl chloride
Carbon tetrachloride
Arsenic, inorganic
1,2-Dichloroethane
Benzene
Chromium (VI)
Cadmium
Trichloroethylene
Chromium (VI)
Arsenic, inorganic
Motor Vehicles
Motor Vehicles
Secondary Formaldehyde Formation
Chrome Platers
Cooling Towers
Motor Vehicles
Heating/Woodstoves
Motor Vehicles
Commercial Sterilization/Hospitals
Other
Heating/Woodstoves
Unspecified Stationary Sources
Gasoline Marketing
Other
Refractories
Unspecified Stationary Sources
Heating/Combustion
Coal and Oil Combustion/Heating
Chemical Manufacturing
Iron and Steel
Iron and Steel
Solvent Use
Gasoline Marketing
Other Organic Evaporation
Heating/Woodstoves
Chemical Manufacturing
Dry Cleaning
Petroleum Refining
Solvent Use/Degreasing
Unspecified Stationary Sources
Solvent Use/Degreasing
Nonferrous Smelters
Unspecified Stationary Sources
Heating/Combustion
Unspecified Stationary Sources
Petroleum Refining
Chemical Manufacturing
Coal and Oil Combustion/Heating
Chemical Manufacturing
Chemical Manufacturing
Chemical Users and Producers
Gasoline Marketing
Heating/Woodstoves
Unspecified Stationary Sources
Chemical Users and Producers
Nonferrous Smelters
Chemical Manufacturing
Heating/Combustion
Heating/Combustion
Unspecified Stationary Sources
Unspecified Stationary Sources
Specialty Steel
Solvent Use
5-City (Reference 1)
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
5-City
1
0.8
0.6
0.5
0.5
0.4
0.3
0.3
0.1
0.06
0.06
0.05
0.05
0.05
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.01
0.009
0.009
0.005
0.003
0.002
0.002
0.001
0.001
0.0008
0.0008
0.0008
0.0006
0.0006
0.0005
0.0004
0.0003
0.0002
0.0002
0.00006
0.00006
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5, for a discussion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA's Cancer Risk...., report (Ref 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
14
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Vinyl chloride
Formaldehyde
Trichloroethylene
1,3-Butadiene
Formaldehyde
1,3-Butadiene
Formaldehyde
Benzene
Benzene
Methyl chloride
Formaldehyde
Methyl chloride
Benzene
Benzene
Formaldehyde
Polycyclic organic matter
Polycyclic organic matter
Chromium (VI)
Polycyclic organic matter
Polycyclic organic matter
Methylene chloride
Benzene
Arsenic, inorganic
Benzene
Polycyclic organic matter
Formaldehyde
Methylene chloride
Cadmium
Arsenic, inorganic
Arsenic, inorganic
Perchloroethylene
Polycyclic organic matter
Methylene chloride
Trichloroethylene
Benzene
Formaldehyde
Cadmium
Perchloroethylene
Ethylene oxide
1,3-Butadiene
Acrylonitrile
Chloroform
1,3-Butadiene
Benzene
Arsenic, inorganic
Methylene chloride
. Perchloroethylene
Cadmium
Polycyclic organic matter
Polycyclic organic matter
Methylene chloride
1,2-Dibromoethane
Trichloroethylene
Source category
Chemical Manufacturing
Solvent Use
Chemical Manufacturing
Other
Other
Point Sources
Point Sources
Point Sources
Other
Other
Motor Vehicles
Point Sources
Motor Vehicles
Other Miscellaneous Area Sources
Other Miscellaneous Area Sources
Point Sources
Motor Vehicles/Gasoline
Point Sources
Coal and Oil Combustion/Heating
Motor Vehicles/Diesel
Solvent Use
Motor Vehicles/Gasoline
Coal and Oil Combustion/Heating
Point Sources
Heating/Combustion
Motor Vehicles/Gasoline
Other
Coal and Oil Combustion/Heating
Heating/Combustion
Point Sources
Dry Cleaning
Heating/Woodstoves
Solvent Use/Degreasing
Solvent Use/Degreasing
Gasoline Marketing
Motor Vehicles/Diesel
Point Sources
Solvent Use/Degreasing
Chemical Manufacturing
Motor Vehicles
Chemical Manufacturing
Chemical Users and Producers
Chemical Manufacturing
Motor Vehicles
Coal and Oil Combustion/Heating
Chemical Users and Producers
Solvent Use/Degreasing
Heating/Combustion
Heating/Woodstoves
Motor Vehicles
Solvent Use
Motor Vehicles
Solvent Use/Degreasing
Study (Reference)
5-City
5-City
5-City
Harris County, TX (Ref. 11)
Harris County, TX
Harris County, TX
Harris County, TX
Hams County, TX
Harris County, TX
Harris County, TX
Harris County, TX
Harris County, TX
Harris County, TX
Harris County, TX
Harris County, TX
lEMP-Baltimore II (Ref. 7)
TEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
ESMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Baltimore II
lEMP-Kanawha Valley (Ref.
lEMP-Kanawha Valley
lEMP-Kanawha Valley
DEMP-Kanawha Valley
lEMP-Kanawha Valley
lEMP-Kanawha Valley
EEMP-Kanawha Valley
lEMP-Kanawha Valley
lEMP-Kanawha Valley
lEMP-Kanawha Valley
lEMP-Kanawha Valley
ESMP-Kanawha Valley
lEMP-Kanawha Valley
lEMP-Kanawha Valley
' lEMP-Kanawha Valley
Cancer Incidence
(cages/million people)
0.00002
0.00002
0.000001
0.8
0.3
0.07
0.06
0.02
0.01
0.01
0.002
0.002
0.001
0.0001
0.00006
0.6
0.5
0.3
0.09
0.08
0.08
0.07
0.06
0.06
0.06
0.06
0.05
0.03
0.03
0.03
0.02
0.02
0.02
0.02
0.006
0.006
0.006
0.006
1) 8
3
2
2
1
0.4
0.2
0.03
0.03
0.02
0.02
0.01
0.01
0.01
0.01
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5, for a discussion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA'a Cancer Risk..... report (Ref 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
15
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Source category
Study [Reference)
Cancer Incidence
(cases/million people)
Dichloroethene
Benzene
Cadmium
Arsenic, inorganic
1,2-Dichloroethane
Trichloroethylene
1,2-Dibromoethane
Cadmium
1,2-Dichloroethane
1,2-Dichloroethane
Gasoline vapor
Gasoline vapor
1,2-Dichloroethane
Perchloroethylene
1,2-Dichloroethane
Trichloroethylene
1,2-Dichloroethane
Benzene
Perchloroethylene
Benzene
1,2-Dichloroethane
Perchloroethyl ene
Benzene
Trichloroethylene
1,2-Dichloroethane
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Chromium (VI)
Benzene
Benzene
Chromium (VI)
Cadmium
1,2-Dibromoethane
Coke oven emissions
Chromium (VI)
Formaldehyde
Polycyclic organic matter
Carbon tetrachloride
1,3-Butadiene
Polycyclic organic matter
Gasoline vapor
Benzene
Benzene
Chromium (VI)
Arsenic, inorganic
Formaldehyde
Chemical Manufacturing
Gasoline Marketing
Motor Vehicles
Waste Oil Combustion/Burning
Gasoline Marketing
Chemical Manufacturing
Gasoline Marketing
Waste Oil Combustion/Burning
Unspecified Stationary Sources
POTW's
Petroleum Refining
Gasoline Marketing
Delaware River
Dry Cleaning
Sewer Volatilization
Solvent Use/Degreasing
Petroleum Refining
Petroleum Refining
Solvent Use/Degreasing
Gasoline Marketing
Gasoline Marketing
POTW's
POTW's
Unspecified Stationary Sources
Chemical Manufacturing
Motor Vehicles
Industrial Solvent Coating
Heating/Combustion
Pesticide Usage
Unspecified Stationary Sources
Gasoline Marketing
Other Organic Evaporation
Chemical Manufacturing
Municipal Waste Combustors
Unspecified Stationary Sources
Motor Vehicles
Unspecified Stationary Sources
Motor Vehicles
Motor Vehicles
Motor Vehicles
Iron and Steel
Chrome Platers
Secondary Formaldehyde Formation
Heating/Woodstoves
Background Concentrations
Motor Vehicles
Motor Vehicles
Motor Vehicles
Iron and Steel
Motor Vehicles
Cooling Towers
Iron and Steel
Motor Vehicles
lEMP-Kanawha Valley 0.01
lEMP-Kanawha Valley 0.009
lEMP-Kanawha Valley 0.005
lEMP-Kanawha Valley 0.004
lEMP-Kanawha Valley 0.004
lEMP-Kanawha Valley 0.003
lEMP-Kanawha Valley 0.003
lEMP-Kanawha Valley 0.0009
lEMP-Kanawha Valley 0.0002
lEMP-Philadelphia (Ref. 1) 0.05
lEMP-Philadelphia 0.03
lEMP-Philadelphia 0.03
lEMP-Philadelphia 0.01
lEMP-Philadelphia 0.01
lEMP-Philadelphia 0.01
lEMP-Philadelphia 0.01
lEMP-Philadelphia 0.007
lEMP-Philadelphia 0.006
lEMP-Philadelphia 0.002
lEMP-Philadelphia 0.002
lEMP-Philadelphia 0.0005
lEMP-Philadelphia 0.0005
lEMP-Philadelphia 0.0002
lEMP-Philadelphia 0.0002
lEMP-Philadelphia 0.00002
lEMP-Santa Clara (Ref. 1) 0.1
lEMP-Santa Clara 0.04
lEMP-Santa Clara 0.02
lEMP-Santa Clara 0.01
lEMP-Santa Clara 0.007
lEMP-Santa Clara 0.007
lEMP-Santa Clara 0.005
lEMP-Santa Clara 0.002
lEMP-Santa Clara 0.0005
South Coast (Ref. 1) 0.6
South Coast 0.4
South Coast 0.4
South Coast 0.1
South Coast 0.04
South Coast 0.002
Southeast Chicago (Ref. 1) 0.9
Southeast Chicago 0.5
Southeast Chicago 0.4
Southeast Chicago 0.3
Southeast Chicago 0.2
Southeast Chicago 0.3
Southeast Chicago 0.1
Southeast Chicago 0.1
Southeast Chicago 0.08
Southeast Chicago 0.07
Southeast Chicago 0.06
Southeast Chicago 0.05
Southeast Chicago 0.05
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5, for a discussion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA's Cancer Risk...., report (Ref 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
16
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Gasoline vapor
Cadmium
Ethylene oxide
Formaldehyde
Formaldehyde
Trichloroethylene
Ethylene oxide
Arsenic, inorganic
Benzene
Chromium (VI)
Benzene
1,2-Dibromoe thane
1,2-Dichloroethane
Perchloroethylene
Cadmium
Trichloroethylene
2,3,7,8-TCDDioxin
Cadmium
Chromium (VI)
Formaldehyde
1,3-Butadiene
Polycyclic organic matter
1,3-Butadiene
Coke oven emissions
1,3-Butadiene
Carbon tetrachloride
Polycyclic organic matter
Formaldehyde
Chromium (VT)
Benzene
Polycyclic organic matter
Polycyclic organic matter
Benzene
Gasoline vapor
Formaldehyde
Ethylene oxide
Arsenic, inorganic
Arsenic, inorganic
Hexachlorobenzene
Benzene
Chromium (VI)
1,3-Butadiene
Trichloroethylene
Polycyclic organic matter
Chloroform
Perchloroethylene
Formaldehyde
Formaldehyde
Cadmium
Formaldehyde
Ethylene oxide
Benzene
Carbon tetrachloride
Source category
Gasoline Marketing
Iron and Steel
Commercial Sterilization/Hospitals
Heating/Combustion
Unspecified Stationary Sources
Solvent Use/Degreasing
Unspecified Stationary Sources
Other
Other Miscellaneous Area Sources
Specialty Steel
Unspecified Stationary Sources
Motor Vehicles
Unspecified Stationary Sources
Chemical Manufacturing
Motor Vehicles
TSDF's
TSDF's
Unspecified Stationary Sources
Chrome Platers
Background Concentrations
Motor Vehicles
Motor Vehicles
Aircraft Engines
Steel Mills/Iron and Steel
Nonroad Engines
Background Concentrations
Nonroad Engines
Aircraft Engines
Cooling Towers
Motor Vehicles
Aircraft Engines
Other
Steel Mills/Iron and Steel
Gasoline Marketing
Motor Vehicles
Point Sources
Steel Mills/Iron and Steel
Point Sources
POTW's
Nonroad Engines
Coal and Oil Combustion/Heating
Other
Solvent Use/Degreasing
Heating/Combustion
Other Miscellaneous Area Sources
Dry Cleaning
Other Miscellaneous Area Sources
Heating/Combustion
Steel Mills/Iron and Steel
Nonroad Engines
Commercial Sterilization/Hospitals
Aircraft Engines
Point Sources
Cancer Incidence
(cases/million people)
0.03
Study (Reference)
Southeast Chicago
Southeast Chicago 0.03
Southeast Chicago 0.01
Southeast Chicago 0.008
Southeast Chicago 0.008
Southeast Chicago 0.007
Southeast Chicago 0.007
Southeast Chicago 0.004
Southeast Chicago 0.004
Southeast Chicago 0.002
Southeast Chicago 0.002
Southeast Chicago 0.002
Southeast Chicago 0.002
Southeast Chicago 0.001
Southeast Chicago 0.0004
Southeast Chicago 0.0004
Southeast Chicago 0.0004
Southeast Chicago 0.0004
Southwest Chicago (Ref. 10) 0.5
Southwest Chicago 0.4
Southwest Chicago 0.3
Southwest Chicago 0.3
Southwest Chicago 0.2
Southwest Chicago 0.2
Southwest Chicago 0.2
Southwest Chicago 0.2
Southwest Chicago 0.1
Southwest Chicago 0.07
Southwest Chicago 0.07
Southwest Chicago 0.06
Southwest Chicago 0.06
Southwest Chicago 0.05
Southwest Chicago 0.04
Southwest Chicago 0.04
Southwest Chicago 0.03
Southwest Chicago 0.02
Southwest Chicago 0.02
Southwest Chicago 0.02
Southwest Chicago 0.01
Southwest Chicago 0.01
Southwest Chicago 0.01
Southwest Chicago 0.01
Southwest Chicago 0.01
Southwest Chicago 0.009
Southwest Chicago 0.009
Southwest Chicago 0.008
Southwest Chicago 0.008
Southwest Chicago 0.008
Southwest Chicago 0.007
Southwest Chicago • 0.007
Southwest Chicago 0.007
Southwest Chicago 0.006
Southwest Chicago 0.006
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5 for a discussion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA's Cancer Risk... report (Ref. 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
17
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Formaldehyde
Hexachlorobenzene
Chromium (VI)
Asbestos (friable)
Methylene chloride
2,3,7,8-TCDDioxin
Methylene chloride
Cadmium
Perchloroethylene
Methylene chloride
Trichloroethylene
Asbestos (friable)
Benzene
1,2-DichJoroethane
Formaldehyde
Perchloroethylene
1,2-Dibromoethane
Gasoline vapor
Methylene chloride
Vinyl chloride
Polycyclic organic matter
Benzene
Acrylonitrile
Vinyl chloride
Trichloroethylene
Hexachlorobenzene
Chromium (VI)
Chromium (VI)
Dichloroethene
Vinyl chloride
1,3-Butadiene
Methyl chloride
Methylene chloride
Dichloroethene
Formaldehyde
Methylene chloride
Benzene
Chloroform
Benzene
Cadmium
Styrene
Perchloroethylene
Methylene chloride
Perchloroethylene
Trichloroethylene
Trichloroethylene
Benzene
Chloroform
Methylene chloride
1,3-Butadiene
1,2-Dichloroethane
2,3,7,8-TCDDioxin
Acrylamide
Source category
Point Sources
TSDF's
Heating/Combustion
Motor Vehicles
Other Miscellaneous Area Sources
Point Sources
Paint and Other Stripping
Point Sources
Solvent Use/Degreasing
Solvent Use/Degreasing
Point Sources
Asbestos/Demolition
Point Sources
Point Sources
Coal and Oil Combustion/Heating
Point Sources
Point Sources
Point Sources
Surface Coating
TSDF's
Heating/Woodstoves
Gasoline Marketing
Point Sources
Point Sources
TSDF's
Point Sources
Other
Steel Mills/Iron and Steel
TSDF's
Municipal Waste Landfill
TSDF's
Surface Coating
TSDF's
Municipal Waste Landfill
Steel Mills/Iron and Steel
Point Sources
TSDF's
POTWs
Surface Coating
Motor Vehicles
Point Sources
Municipal Waste Landfill
Municipal Waste Landfill
POTWs
Municipal Waste Landfill
POTWs
Barge Loading
TSDF's
POTWs
Steel Mills/Iron and Steel
POTWs
TSDF's
Point Sources
Study (Reference)
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Cancer Incidence
(cases/million people)
0.006
0.005
0.004
0.004
0.003
0.003
0.002
0.002
0.002
0.002
0.001
0.001
0.001
0.001
0.001
0.001
0.001
0.0009
0.0008
0.0007
0.0007
0.0007
0.0006
0.0005
0.0005
0.0005
0.0005
0.0005
0.0004
0.0004
0.0003
0.0002
0.0002-
0.0002
0.0001
0.0001
0.0001
0.0001
0.00009
0.00006
0.00005
0.00005
0.00005
0.00003
0.00003
0.00003
0.00003
0.00003
0.00003
0.00002
0.00002
0.00001
0.00001
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5 for a diacuasion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA'a Cancer Risk... report (Ref. 1), rather than the primary study reference itself, cite
"Reference 1' in the above table. Rationale for this are explained in Section 2.2.
18
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Epichlorohydrin
Benzene
Benzene
Propylene oxide
Styrene
Asbestos (friable)
Dichloroethene
1,2-Dichloroethane
Styrene
PercbJoroethylene
Polychlorinated biphenyls
Methyl chloride
Chloroform
Arsenic, inorganic
Beryllium
Methyl chloride
Formaldehyde
Cadmium
Acrylonitrile
Polychlorinated biphenyls
Polychlorinated biphenyls
Beryllium
Methyl chloride
Chromium (VI)
Epi chlorohy drin
Diesel PM
Gasoline PM
Woodstove PM
1,3-Butadiene
Chromium (VI)
Polycyclic organic matter
Benzene
Polycyclic organic matter
Formaldehyde
Formaldehyde
Chromium (VI)
Arsenic, inorganic
Trichloroethylene
Cadmium
Arsenic, inorganic
Ethylene oxide
Perchloroethylene
Ethylene oxide
Benzene
Methylene chloride
1,2-Dibromoethane
1,3-Butadiene
Formaldehyde
Trichloroethylene
PAHCs
Formaldehyde
Cadmium
Arsenic, inorganic
Source category
Study (Reference)
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
.Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Southwest Chicago
Twin City (Ref. 17)
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Cancer Incidence
(cases/million people)
Point Sources
Municipal Waste Landfill
POTW's
Point Sources
POTW's
TSDFs
POTW's
TSDF's
TSDFs
TSDFs
TSDF's
Point Sources
Point Sources
TSDFs
Point Sources
POTW's
TSDFs
TSDF's
TSDF's
Point Sources
Municipal Waste Landfill
TSDFs
TSDF's
TSDFs
TSDF's
Motor Vehicles
Motor Vehicles
Heating/Woodstoves
Motor Vehicles
Chrome Platers
Heating/Combustion
Motor Vehicles
Commercial Incinerators
Motor Vehicles
Heating/Woodstoves
Other Miscellaneous Area Sources
Heating/Combustion
Other Miscellaneous Area Sources
Heating/Combustion
Point Sources
Other Miscellaneous Area Sources
Other Miscellaneous Area Sources
Point Sources
Point Sources
Other Miscellaneous Area Sources
Point Sources
Point Sources
Point Sources
Point Sources
Commercial Incinerators
Heating/Combustion
Point Sources
Heating/Woodstoves
0.00001
0.000009
0.000009
0.000007
0.000007
0.000005
0.000005
0.000004
0.000004
0.000004
0.000004
0.000003
0.000002
0.000002
0.0000009
0.0000008
0.0000005
0.0000004
0.0000004
0.0000004
0.00000004
9E-09
3E-09
3E-09
3E-09
0.7
0.4
0.4
0.3
0.3
0.2
0.1
0.08
0.06
0.03
0.02
0.01
0.01
0.009
0.005
0.005
0.005
0.004
0.004
0.004
0.004
0.004
0.003
0.003
0.002
0.002
0.002
0.002
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5 for a discussion of the many assumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA's Cancer Risk... report (Ref. 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
19
-------
Table 2-1. Cancer Incidence by Pollutant and Source Category for Each Study
(continued)
Pollutant
Benzene
1,2-Dichloroethane
Methylene chloride
1,2-Dibromoethane
Chromium (VI)
Perchloroethylene
Beryllium
Arsenic, inorganic
1,2-Dichloroethane
Cadmium
Chromium (VI)
Beryllium
Acrylonitrile
PAHCs
Cadmium
Formaldehyde
Styrene
PAHCs
Acetaldehyde
Formaldehyde
Coke oven emissions
1,3-Butadiene
Carbon tetrachloride
Chromium (VI)
Polycyclic organic matter
2,3,7,8-TCDDioxin
Arsenic, inorganic
Beryllium
Asbestos (friable)
Benzene
Gasoline vapor
Cadmium
1,2-Dibromoethane
Vinyl chloride
Trichloroethylene
Perchloroethylene
Polychlorinated biphenyls
Styrene
Methylene chloride
Chloroform
Acrylonitrile
Epichlorohydrin
Ethylene oxide
Source category
Other Miscellaneous Area Sources
Other Miscellaneous Area Sources
Point Sources
Other Miscellaneous Area Sources
Point Sources
Point Sources
Heating/Combustion
Other Miscellaneous Area Sources
Point Sources
Heating/Woodstoves
Heating/Combustion
Other Miscellaneous Area Sources
Point Sources
Heating/Combustion
Other Miscellaneous Area Sources
Other Miscellaneous Area Sources
Point Sources
Other Miscellaneous Area Sources
Point Sources
Secondary Formaldehyde Formation*
Iron and Steel*
Background Concentrations*
Motor Vehicles*
Other*
Coal and Oil Combustion/Heating*
Manufacturing Operations*
Heating/Woodstoves*
Study (Reference)
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Twin City
Detroit (Ref. 12)
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Detroit
Cancer Incidence
(cases/million people)
0.001
0.0007
0.0006
0.0006
0.0005
0.0003
0.0002
0.0001
0.0001
0.00008
0.00008
0.00007
0.00002
0.00001
0.000007
0.000006
0.000005
0.000002
0.000001
0.4
0.2
0.2
0.2
0.05
0.04
0.04
0.02
0.02
0.02
0.02
0.01
0.005
0.002
0.001
0.001
0.0007
0.0003
0.0003
0.0003
0.00001
0.000003
0.000003
0.000003
0.5
0.2
0.2
0.1
0.1
0.08
0.02
0.001
' The Detroit study did not joint!; analyze pollutant and source category effects. Therefore, what is presented are specific chemical effects
for all source categories combined and specific source category effects for all chemicals combined. The sum of cancer incidences for
both of these subcategories are necessarily equal and inclusion of both subcategories to obtain a total cancer incidence would lead
to double counting.
CAVEAT: Estimates of cancer incidence in these screening studies are not absolute predictions of cancer occurrence and are intended to be
used in a relative sense only. See Section 2.5, for a discussion of the many asBumptions and uncertainties associated with these estimates.
NOTE: Studies whose results were excerpted from EPA's Cancer Risk...., report (Ref 1), rather than the primary study reference itself, cite
"Reference 1" in the above table. Rationale for this are explained in Section 2.2.
20
-------
Chapter 3
SUMMARIES OF NEW STUDIES
Seven studies were compiled for this report in addition to those studies that were already
summarized in EPA's 1990 EPA report entitled Cancer Risk From Outdoor Exposure to Air Toxics'.
The studies summarized in Cancer Risk... are not summarized again herein. Only the seven
additional studies not included in Cancer Risk... are summarized below. For each of these seven
studies, a brief discussion of background and methodology is presented, along with study results and
the assumptions and limitations associated with the study. Much of the material presented here,
particularly the graphics showing each study's results, is taken verbatim from these studies (and
labeled as such).
3.1 Southwest Chicago Study10
Report Title/Date: Estimation and Evaluation of Cancer Risks Attributed to Air Pollution
in Southwest Chicago, April 1993
Conducted by: U.S. EPA Region 5, Air and Radiation Division
Background and Methodology: The purpose of this study was to estimate the cancer risks
associated with 30 carcinogenic air pollutants in the Southwest Chicago area. The study was
conducted by ViGYAN Inc. in Falls Church, Virginia for U.S. EPA Region 5, Air and Radiation
Division.
The study area included Midway Airport and the neighboring suburbs that is bordered on the
north by Pershing Road, on the south by 70th Street, on the west by Harlem Avenue, and on the east
by Pulaski Avenue. The area is approximately a 4-mile square grid (16 square miles) that is divided
into 64 (8x8) rectangular grid cells. A total population of 93,845 resided in the study area.
The emission inventory included point sources, area sources, and mobile sources. The point
source types included those traditionally inventoried in air pollution studies as well as some source
types that are not traditionally inventoried such as volatilization from wastewater at sewage treatment
plants; hazardous waste treatment, storage, and disposal facilities (TSDFs); abandoned hazardous
waste sites; and landfills storing municipal waste. Using a broad emission inventory as the base
inventory, the investigators searched the Toxic Release Inventory (TRI) database for new or additional
sources to add to the inventory. Information from Resource Conservation and Recovery Act (RCRA)
permits and Illinois EPA records were reviewed and six additional point sources were added to the
inventory.
The inventory also included several area source categories such as road vehicles, gasoline
marketing, barge loading, commercial and residential heating, residential wood combustion, dry
cleaners, degreasing, surface coating, hospitals, paint strippers, demolition, chrome platers, and other
area sources inventoried on a per capita basis. The mobile sources included road vehicles, nonroad
engines, and aircraft engine emissions at Midway Airport. Aircraft engine emissions by aviation
category from all phases of the landing and takeoff (LTO) cycle among all aircraft in 1990 were
estimated. This study also considered the effects of carcinogenic air emissions from both point and
area sources as far as 16 kilometers (10 miles) to the north of the Southwest Chicago receptor area.
21
-------
This study used the Industrial Source Complex - Long Term (1SCLT) dispersion model and
the Climatological Dispersion Model (COM) for air dispersion modeling. These two dispersion
models incorporated selected meteorological and emission inventory data to predict air pollutant
concentrations in the receptor grid network. The ISCLT model was used for predicting
concentrations from point sources and the CDM model was used for nonpoint or volume sources.
The modeled ambient concentrations were based on the combination of the point source contribution,
the nonpoint or volume source contribution, and the assumed background concentrations for
formaldehyde and carbon tetrachloride. (Background formaldehyde results from photochemical
reactions and background carbon tetrachloride results largely from atmospheric accumulation of
"historical" emissions.) This study assumed background levels of 2.23 ug/m3 for formaldehyde and
0.76 ug/m3 for carbon tetrachloride, as documented in the Southeast Chicago study report. This
assumption was made because these two areas are very close to each other. Risks due to background
formaldehyde were added to these risks associated with directly emitted formaldehyde.
The study compared available monitored concentrations to modeled concentrations to assess
the quality of model predictions. Monitored ambient concentrations for arsenic and cadmium were
available for 1988 and 1991 in two locations. Monitored data for benzene, carbon tetrachloride,
formaldehyde, methyl chloride, methylene chloride, perchloroethylene, styrene, and trichloroethylene
were available for one location in the 1988/89 period. The study found that modeled concentrations
generally are close to (e.g., on the same order of magnitude) or less than monitored concentrations.
The investigators felt that the emission inventory might have underestimated emissions affecting the
study area by either underestimating emissions at sources or failing to identify some sources of
emissions.
Based on the estimated ambient concentrations, lifetime individual risks and lifetime cancer
incidence were calculated for all pollutants and their contributing source categories at the receptors
grid cells. The areawide cancer risk estimate within the study area was estimated by multiplying the
lifetime individual risks in each receptor grid cell by the population in that grid cell and then
summing the lifetime cancer cases over all 64 receptor grid cells from all pollutants and source
categories.
Results: The study showed that the total number of cancer cases attributable to air pollution
in this study area is about 20 cases over 70 years. Tables 3-1 and 3-2, adapted directly from the
study, show the aggregated lifetime individual risks, aggregated lifetime cancer cases, and percent
cancer cases by source category and by pollutant, respectively. Table 3-3 shows the pollutants
attributed to road vehicles, background formaldehyde and carbon tetrachloride, chrome platers,
nonroad mobile sources, and aircraft engines, which are the top five sources contributing to cancer
cases. Table 3-4 shows the contributing source categories for 1,3-butadiene, POM, chromium VI,
and formaldehyde, which are the top four pollutants causing cancers. The average lifetime individual
risk across the area due to air pollution is approximately 2.1 x 10"*.
Limitations: There are some limitations in this study. In terms of source category, this
study may have underpredicted the cancer incidence caused by residential woodstoves and fireplaces
because their cancer risks are two to three orders of magnitude lower than those in other urban air
toxics studies. Because the modeled ambient air concentrations are generally less than the monitored
concentrations, it is likely that some sources and pollutants were not captured in the inventory. This
study may disproportionably emphasize airport emissions and the estimated risks because of the
22
-------
Excerpted directly from the Southwest Chicago Estimation and
Evaluation of Cancer Risk Attributed to Air Pollution Study
FINAL SUMMARY REPORT
Table 3-1. Aggregate Hazard Indices by Source Category*
Source Category
Road Vehicles'1
Background Concentration
Chrome Platers
Nonroad Engines
Aircraft Engines
Steel Mills
Other Industrial Points
Cooling Towers
Residential Heating
Gasoline Marketing
Industrial Heating
Wastewater Treatment
Per Capita Area Sources
Commercial Heating
De greasing
Hospitals
Dry Cleaners
Paint Strippers
Other Hazardous Waste TSDFs
Surface Coating
Demolition
Residential Wood Combustion
Municipal Landfills
RCRA Hazardous Sites
Barge Loading
Lifetime
Individual
Risks
3.34E-03
2.58E-03
1.78E-03
1.34E-03
1.42E-03
9.50E-04
4.62E-04
3.00E-04
1.99E-04
1.66E-04
9.47E-05
1.72E-04
7.94E-05
7.94E-05
6.99E-05
3.27E-05
2.54E-05
1.05E-05
7.82E-06
6.36E-06
6.28E-06
2.96E-06
8.94E-07
2.28E-07
1.17E-07
Lifetime
Cancer
Cases
5.03
3.79
3.13
2.14
2.11
1.41
0.68
0.45
0.31
0.26
0.14
0.13
0.12
0.12
0.10
0.049
0.039
0.016
0.012
0.010
0.0098
0.0044
0.0013
0.00034
0.00017
Percent
Cancer
Cases
25.08
18.88
15.60
10.65
10.52
7.03
3.41
2.24
1.54
1.28
0.70
0.63
0.62
0.61
0.52
0.25
0.19
_e
-
-
-
-
-
-
-
Aggregated over all pollutants among all receptor grids
Including vehicular emissions from parking lots and Helen Mikols Drive at Midway
Airport
"-" indicates less than 0.1%
23
-------
Excerpted directly from the Southwest Chicago Estimation and
Evaluation of Cancer Risk Attributed to Air Pollution Study
FINAL SUMMARY REPORT
Table 3-2. Aggregate Hazard Indices by Pollutants1
Pollutant
1.3-Butadiene
POM*
Hexavalent Chromium
Formaldehyde
Coke Oven Emissions
Carbon Tecrachloride
Benzene
Arsenic
Gasoline Vapors
Ethylene Oxide
Hexachlorobenzcne
Trichloroethylene
Perchloroethylene
Cadmium
Methylene Chloride
Chloroform
Asbestos
Dioxins
Vinyl Chloride
Eihylene Dichloride
Ethylene Dibromide
Vinylidene Chloride
Acrylonitrile
Methyl Chloride
Styrene
Acrylaraide
Epichlorohydrin
Propylene Oxide
PCB$
Beryllium
Individual
Concentrations
(ug/mj)
.09E+01
.24E+02
.79E-01
.87E+O2
.30E+00
4.87E+01
6.29E401
4.37E-02
2.35E+02
1.29E+00
3.68E-01
3.07E+01
8.12E+01
2.36E-02
8.25E+01
1.68E+00
3.03E-03
4.34E-07
8.61E-02
3.21E-01
2.28E-02
4.98E-02
3.90E-02
5.31E-01
6.58E-01
4.34E-05
3.51E-02
8.46E-03
8.76E-06
1.59E-06
Lifetime
Individual
Risks
3.06E-03
2.50E-03
2.15E-03
2.44E-03
8.05E-04
7.30E-04
5.22E-04
1.88E-04
1.55E-04
1.29E-04
1.69E-04
5.22E-05
4.71E-05
4.25E-05
3.88E-OS
3.86E-OS
2.30E-05
1.43E-05
7.23E-06
8.34E-06
5.02E-06
2.49E-06
2.65E-06
9.56E-07
3.75E-07
5.64E-08
4.21E-08
3.13E-08
1.93E-08
3.82E-09
Lifetime
Cancer
Cases
4.69
3.78
3.69
3.60
1.20
1.07
0.80
0.27
0.24
0.19
0.13
0.077
0.074
0.062
0.058
0.057
0.035
0.021
0.010
0.0087
0.0067
0.0038
0.0037
0.0015
0.00041
0.000083
0.000071
0.000045
0.000027
0.0000057
Percent
Cancer
Cases
23.36
18.83
18.38
17.94
5.96
5.33
3.96
1.35
1.19
0.94
0.66
0.38
0.37
0.31
0.29
0.28
0.17
0.11
_«
•
-
-
-
-
-
-
-
-
-
-
Aggregated over all source categories among all receptor grids
Aggregated over all inventoried POM emission sources
"-" indicates less than 0.1%
24
-------
Excerpted directly from the Southwest Chicago Estimation and
Evaluation of Cancer Risk Attributed to Air Pollution Study
Table 3-3. Top Five Source Contributors to Cancer Cases
Pollutant
SOURCE CATEGORY
1 ,3-Butadiene
Diesel Paniculate
Gasoline Paniculate
Benzene
Formaldehyde
Asbestos
Cadmium
SOURCE CATEGORY
Formaldehyde
Carbon Tetrachloride
SOURCE CATEGORY
Hexavalent Chromium
SOURCE CATEGORY
1,3 -Butadiene
Diesel Paniculate
Gasoline Paniculate
Benzene
Formaldehyde
SOURCE CATEGORY
1,3-Butadiene
Formaldehyde
Turbine Paniculate
Benzene
Piston Paniculate
Concentrations
(ug/m3)
- ROAD VEHICLES
5.23E+00
6.68E+01
5.93E-^
3.29E-H01
1.11E+01
2.20E-03
9.49E-Q4
Lifetime
Individual Risks
1.46E-03
1.13E-03
3.03E-04
2.73E-04
1.45E-04
1.67E-05
1.71E-06
Lifetime
Cancer Cases
2.22
1.69
0.46
0.42
0.22
0.025
0.0026
- BACKGROUND CONCENTRATIONS
1.43E-K32
4.86E+01
- CHROME PLATERS
1.48E-01
- NONROAD MOBILE
2.68E-KX)
2.69E-KU
3.36E+00
6.39E+01
2.27E-KI1
1.86E-03
7.30E-04
1.78E-03
SOURCES
7.52E-04
4.57E-04
5.38E-05
5.31E-05
2.95E-05
2.72
1.07
3.13
1.19
0.73
0.085
0.084
0.047
. AIRCRAFT ENGINES
2.87E400
2.38E401
1.63E401
3.40E+00
4.61E-01
8.03E-04
3.09E-04
2.76E-04
2.82E-05
7.37E-06
1.21
0.47
0.39
0.041
0.0082
25
-------
Excerpted directly from the Southwest Chicago Estimation and
Evaluation of Cancer Risk Attributed to Air Pollution Study
Table 3-4. Top Four Pollutant Contributors to Cancer Cases
Source Category
POLLUTANT - U-BUTADBENE
Road Vehicles
Aircraft Engines
Nonroad Engines
Other Industrial Points
Other Hazardous Waste TSDFs
Steel Mills
POLLUTANT - POM/Particulate Matter
Road Vehicles
Diesel Paniculate
Gasoline Paniculate
Nonroad Engines
Diesel Paniculate
Gasoline Paniculate
Aircraft Engines
Piston Paniculate
Turbine Paniculate
Others POM Sources (B(a)P Surrogate)
Residential Heating (Distillate Oil Use)
Woodstoves
Concentrations
(ug/m*)
5.23E-HOO
2.87E-MX)
2.68E-MX)
1.54E-01
3.88E-03
3.29E-04
7.27E«01
6.68E+01
5.93E+00
3.03E+01
2.69E+01
3.36E+00
1.68E+01
1.63E+01
4.61E-01
1.37E-01
4.24E-MX)
1.02E-01
Lifetime
Individual Risks
1.46E-03
8.03E-04
7.52E-04
4.30E-05
1.09E-06
9.20E-08
1.43E-03
1.I3E-03
3.03E-04
5.11E-04
4.57E-04
5.38E-OS
2.83E-04
2.76E-04
7.37E-06
2.32E-04
3.82E-05
2.96E-06
Lifetime
Cancer Cases
2.22
1.21
1.19
0.071
0.0018
0.00014
2.15
0.82
0.40
0.35
0.059
0.0044
1.69
0.46
0.73
0.065
0.39
0.0082
continued
26
-------
Excerpted directly from the Southwest Chicago Estimation and
Evaluation of Cancer Risk Attributed to Air Pollution Study
Table 3-4. Top Four Pollutant Contributors to Cancer Cases (continued)
Source Category
Concentrations
(ug/m1)
Lifetime
Individual Risks
Lifetime
Cancer Case*
POLLUTANT • HEXAVALENT CHROMIUM
Chrome Platers
Cooling Towers
Commercial Heating
Residential Heating
Industrial Heating
Steel Mills
Other Industrial Points
RCRA Hazardous Sites
1.48E-01
2.50E-02
3.14E-03
1.57E-03
8.79E-04
1.69E-04
1.72E-04
1.34E-09
1.78E-03
3.00E-04
3.77E-05
1.88E-05
1.06E-05
2.03E-06
2.06E-06
1.60E-11
3.13
0.45
0.058
0.029
0.016
0.0030
0.0030
0.00000002
POLLUTANT - FORMALDEHYDE
Background Concentrations 1.43E+02
Aircraft Engines 2.38E-M}!
Road Vehicles l.llE+Ol
Per Capita Area Sources 2.50E+00
Residential Heating . 2.i«E+00
Nonroad Engines 2.2~E-00
Other Industrial Points 2.G5E-rOO
Commercial Heating 2.47E-01
Industrial Heating 1.84E-01
Steel Mills 5.00E-02
Other Hazardous Waste TSDFs 1.77E-04
1.86E-03
3.09E-04
1.45E-04
3.25E-05
3.24E-05
2.95E-05
2.64E-05
3.21E-06
2.39E-06
6.51E-07
2.31E-09
2.72
0.47
0.22
0.051
0.050
0.047
0.037
0.0049
0.0036
0.00098
0.0000034
27
-------
configuration of the study grid around Midway Airport. The reader should note that the lifetime
individual risks and concentrations expressed in the tables excerpted from the SW Chicago study are
not averaged over the receptor grid, but rather, are the sum of the risks or concentrations across all
64 grid cells in the receptor grid. Hence, the values in these tables do not make physical sense, per
se, and should not be construed as representative of any particular grid cell at all or indicative of the
average lifetime cancer risk across the study area.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
28
-------
3.2 lEMP-Baltimore II Study7
Report Title/Date: Baltimore Integrated Environmental Management Project: Phase II,
Ambient Air Toxics Report, EPA/230-R-92-013, February 1992
Conducted by: U.S. EPA, Office of Policy, Planning and Evaluation
Background and Methodology: The Baltimore Integrated Environmental Management
Project (IEMP) Ambient Air Toxics Study was conducted to assist EPA and local officials in
exploring better ways to identify, assess, and manage the human health risks of environmental
pollutants in the area. This study estimated the increases in cancer and noncancer risks resulting from
exposure to ambient air toxics. The air toxics study was designed to test a new approach (using
cancer potency-weighted ambient concentrations) in the screening to set research and control priorities
among sources and pollutants contributing to the complex mixture of ambient air toxics in the
Baltimore area. This study was originally reported in Cancer Risk...', but was further modified and
was thus treated as a new study in this report.
The study area included the City of Baltimore, the southern half of Baltimore County, and the
northern part of Anne Arundel County. The study area was subdivided into 64 (8x8) 5-km squares
grid cells. Another refined grid consisting of 64 (8x8) 2.5-km grid cells overlaid the most densely
populated portion of the study area. The 1980 population of 1.6 million was used for the study area.
The emission inventory covered more than 200 pollutants, but only a small number of
compounds were selected for evaluation in the Baltimore air toxics study. The first set of pollutants
included 12 organic gases, 2 organic particulates, and 3 metals for the cancer effect assessment.
Another set of 22 volatile organics and 1 metal was evaluated for a noncancer effect assessment. A
majority of the pollutants showed up in both the cancer and noncancer effect lists. The emission
inventory included approximately 250 point sources, area sources, and nontraditional sources. The
point source emission inventory was based on a 1985 survey which included a limited number of
pollutants and sources where hazardous materials were used, produced, or handled. The area sources
included motor vehicles, gasoline service stations, solvent usage by commercial and small industrial
facilities, heating (i.e., industrial, commercial, institutional, and residential), waste oil combustion,
agricultural burning, and minor point sources not modeled individually. Nontraditional sources
included cooling towers, trihalomethane volatilization from drinking water distribution, and sewage
treatment plants. The investigators calculated emissions from these area sources using an assortment
of emission factors and algorithms developed by EPA's Regulatory Integration Division and Office of
Air and Radiation. Total hydrocarbon emissions from cars and trucks were estimated using EPA's
MOBILES computer model and then subsequently speciated into specific toxic pollutants.
The Industrial Source Complex - Long Term (ISCLT) dispersion model and the
Climatological Dispersion Model (CDM) were used for air dispersion modeling. These two
dispersion models incorporated selected meteorological and emission inventory data to predict air
pollutant concentrations in the receptor grid network. The ISCLT model was used for predicting
concentrations from point sources and the CDM model was used for area sources. Atmospheric
transformation of formaldehyde and long-term accumulation of carbon tetrachloride were not
considered in this study.
29
-------
In addition to doing a modeling analysis, this study also used available monitoring data from
studies conducted in the Baltimore area from 1983 through 1987.
In the Baltimore IEMP, a relative ranking of pollutants and sources was developed using two
approaches. First, at the recommendation of a risk assessment panel from Johns Hopkins, a ranking
of pollutants and sources was developed by cancer potency-weighted ambient concentrations. This
approach was recommended because it avoids relying on highly uncertain exposure assumptions used
in many screening assessments and does not incorporate population weighting factors. A second
approach consisted of conducting a quantitative risk assessment for the pollutants and sources
examined, considering both cancer and noncancer effects.
Results - Ranking by Cancer Potency-Weighted Ambient Concentrations: In this
approach, the investigators calculated the average areawide modelled concentration of each target
pollutant using the 5 km modelling grid system and then multiplied this value by the pollutant's
corresponding unit cancer risk factor. The product is a measure of each pollutant's potential human
health risk in the Baltimore area. Pollutants were compared and ranked by normalizing against the
cancer potency-weighted concentration for one of the more common pollutants in the ambient air —
chloroform. Table 3-5, adapted directly from the study> shows the relative ranking of target com-
pounds by cancer potency-weighted ambient concentrations determined from dispersion models.
POM ranked highest with a score approximately four times higher than the next highest pollutant,
chromium, and roughly twice as high as the summed scores for all other pollutants. Carbon
tetrachloride ranked lowest of all pollutants.
To rank sources under this approach, the investigators first summed the potency-weighted
average concentrations resulting from each source or source category and then normalized against the
value for gasoline marketing. Again, the potency-weighted concentration is simply a measure of the
potential risk to human health posed by a source or source category in the Baltimore area. Table 3-6
shows the relative significance of point sources and area sources based on this ranking. [Note: area
sources are not defined in this study per the Section 112 10/25 TPY cutoff.]
Results - Based on Population-based Cancer Risk Assessment Screening: In the modeling
analysis, the investigators calculated the average area-wide increased cancer risk by (1) multiplying
the estimated average ambient air concentration by pollutant for each 5 km grid cell by the
appropriate unit cancer risk factor, (2) calculating an arithmetic average across all grid cells and (3)
summing across all pollutants. A similar analysis was performed using monitoring data, substituting
monitoring data for model-projected data.
The study presented both average increased lifetime individual cancer risk and annual excess
cancer incidence based on modeled and monitored ambient concentrations. The average increased
lifetime individual cancer risks using modeled data and monitored data were l.SxlO"4 and 5.2x10^,
respectively. Table 3-7 shows the total and pollutant-specific average increased lifetime individual
cancer risks using the available monitored data for the Baltimore area. Three pollutants account for
most (about 87 percent) of the estimated average increased lifetime individual cancer risk:
perchloroethylene (48 percent), hexavalent chromium (24 percent) and benzene (15 percent).
30
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-5. Relative Ranking of Target Compounds by Cancer
Potency-Weighted Ambient Concentrations
POLLUTANT
UNIT
CANCER RISK
FACTOR
AVERAGE
CONCENTRATION
(ug/m3) (1)
RATIO OF
POTENCY- POTENCY-WEIGHTED
WEIGHTED CONCENTRATIONS
CONCENTRATIONS (Chloroform - 1)
POM
CHROMIUM-6
BENZENE
FORMALDEHYDE
MZTHYLENE
CHLORIDE
ARSENIC
CADMIUM
PERCHLOROETHYLENE
TRICHLOROETHYLENE
CHLOROFORM
ETHYLENE DICHLORIDE
ETHYLENE DIBROMIDE
ETHYLENE OXIDE
CARBON
TETRACHLORIDE
6.50E-OS
1.20E-02
8.00E-06
1.30E-03
4.10E-06
1.4865625
0.0015625
1.115625
0.446875
1.4
4.30E-03
1.80E-03
4.80E-07
1.30E-06
2.30E-05
2.60E-05
2.20E-04
l.OOE-04
3.70E-06
0.0013125
0.0014375
2.33
0.7234375
0.02765625
0.005390625
0.000515625
0.000875
0.00015625
9.66E-05
1.88E-OS
8.93E-06
S.81E-06
5.74E-06
5.64E-06
2.59E-06
1.12E-06
9.40E-07
6.36E-07
1.40E-07
1.13E-07
8.75E-08
5.78E-10
151.9
29.5
14.0
9.1
9.0
8.9
4.1
1.8
1.5
1.0
0.2
0.2
0.1
0.0
(1) Modelled concentration
31
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-6. Relative Ranking of Sources by Contribution to Cancer
Potency-Weighted Ambient Concentrations
SOURCE
UNIT
CANCER
RISK
POLLUTANT FACTOR
AVERAGE SUM OP RATIO OF
CONCEH- POTEHCY- POTENCY- SOURCES
TRATION WEIGHTED WEIGHTED (gas
(ug/m3) (1) CONC. CONG. mrktng-1)
Point sources:
Point Source A
POM 6.50E-05
Chrom.-6 1.20E-02
Benzene 8.00E-06
Point Source B Chrom.-6 1.20E-02
Point Source C Chrom.-6 1.20E-02
•Point Source D Chrom.-6 1.20E-02
0.66406 0.0000362 0.0000905 489
0.00245 0.0000294
0.60623 0.0000049
0.00027 0.0000033 0.0000033 18
0.00026 0.0000031 0.0000031 17
0.00011 0.0000013 0.0000013 7
POM 0.00005
Frmldhyde 1.30E-05
Benzene 8.00E-06
Area sources*
Road vehicles
Solvent usage
Heating
Gas marketing Benzene 8.00E-06
Met. CL
PERC
TCE
POM
Arsenic
Cadisiiuifi
4.10E-06
4.80E-07
1.30E-06
0.00001
4.30E-03
1.80E-03
0.11494 0.0000057
0.27422 0.0000036
0.41250 0.0000033
1.34609 0.0000055
0.68156 0.0000003
0.68000 0.0000009
0.45625 0.0000046
0.00088 0.0000038
0.00100 0.0000018
0.0000126
0.0000067
0.0000101
68
36
55
0.02313 0.0000002 0.0000002
(1) Modelled concentration.
32
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-7. Average Increased Lifetime Individual Cancer Risk
Using Available Monitoring Data (1, 2)
RESULTS INTENDED FOR POLICY DEVELOPMENT ONLY
AVERAGE INCREASED
POLLUTANT LIFETIME INDIVIDUAL
[Weight of Evidence] CANCER RISK
Arsenic [Bl] 7.0E-06
Benzene [A] 7.SE-05
Benzo(a)pyrene [B2] 2.5E-06
Cadmium [Bl] 2.2E-06
Carbon tetrachloride [B2] 3.4E-06
Chloroform [B2] 3.2E-05
Chromium VI [A] 1.2E-04
Ethylene dichloride [B2] 1.6E-05
Propylene dichloride [C] 8.6E-06
Ethyl benzene [N/A] N/A
Lead [N/A] N/A
Methylene chloride [B2] 9.8E-06
Methyl isobutyl ketone [N/A] N/A
Perchloroethylene [B2] 2.4E-04
Toluene [N/A] N/A
Methyl chloroform [N/A] N/A
Trichloroethylene [B2] 1.1E-06
Vinyl chloride [A] N/A
Xylene [N/A] N/A
TOTAL 5.2E-04
N/Ai Not Applicable
(1) Source of monitoring data: See Appendix B.
(2) This study uses conservative estimates of increased cancer
risk from ambient (i.e., outdoor) exposure to establish
priorities ajnong pollutants and sources. The risk estimates
are calculated using modelled or monitored concentrations
and EPA unit cancer risk factors. There is considerable
uncertainty in the estimated concentrations, which could
either overstate or understate the true concentrations (see
Chapter IV). Unit cancer risk factors combine CAG potency
estimates with EPA exposure assumptions. The CAG potency
estimates provide a plausible upper limit to the cancer risk
of a compound {see Appendix A); however, the true value of
the risk is unknown and may be as low as zero. The exposure
assumptions are extremely conservative in that they assume
continuous exposure to outdoor air for 70 years. Because of
the generally conservative bias in the information, it is
highly unlikely that the true risks would be as high as the
estimates and they could be considerably lower.
33
-------
Table 3-8 shows that annual excess cancer incidence using modeled concentrations for the top
14 pollutants was 3.56 cases/year. Almost all of the estimated annual excess cancer incidence (about
94 percent) can be attributed to six pollutants: POM, chromium +6, methylene chloride, benzene,
arsenic, and formaldehyde. Most of the incidence (57 percent) is from POM alone.
Table 3-9 shows that the annual excess cancer incidence using available monitored data for the
top 12 pollutants was 11.79 cases/yr. The difference in the predicted annual excess cancer incidence
from modeled and monitored data was explained by a significantly higher incidence estimated for
perchloroethylene, chromium VI, and benzene based on monitored data.
Results - Based on Noncancer Screening: For noncancer effects, the study divided the
modeled and monitored concentration by the no-effect thresholds of those noncancer effects (e.g.,
liver toxicity, kidney toxicity, reproductive, neurological, fetal, or blood) relevant to the pollutant. If
the resulting ratio exceeded one, the specific pollutant was identified for additional investigation.
Table 3-10 shows the potential for blood effects for benzene at several receptor locations and the
multiple effects for xylene at one receptor location warranting further investigation.
For the noncancer impact of exposure to complex chemical mixtures in the ambient air, the
study used a hazard index that summed individual pollutant ratios by the effect category. If the
hazard index exceeded one, the exposures to the complex chemical mixtures deserved further analysis.
As shown in Table 3-11, this study found that benzene is the primary pollutant of potential noncancer
concern at several discrete receptor locations.
Limitations: This study failed to account for background concentrations imported from
outside the study area or formed by chemical transformation. The emission inventory also potentially
underestimated the point and area source emissions in the study area. Comparing the pollutants
included in this study with those in other studies, it was found that this study covered most of the
pollutants causing the-highest number of cancer risks except for 1,3-butadiene and gasoline vapor.
Gasoline vapor and 1,3-butadiene are primarily emitted from gasoline marketing and mobile sources,
respectively. Also, the study did not include chrome platers or cooling towers as major chromium VI
emitters, but attributed chromium VI emissions to four major point sources.
As shown in Table 3-12, the study compared the predicted average lifetime individual cancer
risk for several pollutants in the Baltimore area with those in other studies. This comparison shows
that the predicted cancer risks for all pollutants are generally in the same range as or lower than those
in other studies. The investigators felt the differences in modeling assumptions regarding release
specifications and receptor placement can significantly affect predictions of lifetime individual cancer
risks. Underestimated emissions also contributed to the lower cancer risk incidence predicted.
Modeled concentrations for 10 pollutants at the locations corresponding to 10 monitoring sites
were compared. For all pollutants considered, except trichloroethylene, the models consistently
underpredicted ambient air concentrations. The investigators felt the bias for the modeling could be
attributed to (1) a potential underestimation of the point and area source emissions, (2) failure to
account for background concentrations imported from outside the study area or by chemical
transformation, or (3) the treatment of atmospheric and meteorological conditions in the dispersion
models. Although the exposure estimates based on modeling were understated; the investigators felt
the estimates of risk using these modeled values were still more likely to overstate than understate
34
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-8. Estimated Annual Excess Cancer Incidence for Selected
Pollutants Modelled in the Baltimore IEMP Air Toxics Study
(total study area, 5 km grid system) '
RESULTS INTENDED FOR POLICY DEVELOPMENT ONLY
ANNUAL EXCESS
CANCER INCIDENCE PERCENTAGE
POLLUTANT (Weight of Evidence) OF TOTAL
Polycyclic organic
matter (POM)r 2.05 (N.A.) 57.4
Chromium (VI) 0.53 (A) 14.8
Methylene chloride 0.20 (B2) 5.6
Benzene 0.20 (A) 5.6
Arsenic 0.20 (A) 5.6
Formaldehyde 0.19 (Bl) 5.3
Cadmium 0.09 (Bl) 2.5
Perchloroethylene (PCE) 0.04 (B2) 1.1
Trichloroethylene (TCE) 0.03 (B2) 0.8
Chloroform 0.02 (B2) 0.6
Ethylene dichloride (EDC) 0.01 (B2) 0.3
Ethylene dibromide (EDB) <0.01 (B2) 0.1
Benzo(a)pyrehe (B(a)P) <0.01 (B2) <0.1
Carbon tetrachloride <0.01 (B2) <0.1
TOTAL1 1756100.0
N.A. « Not Available
1 This study uses conservative estimates of increased cancer
risk from ambient (i.e., outdoor) exposure to establish
priorities among pollutants and sources. The risk estimates
are calculated using modelled or monitored concentrations and
EPA unit cancer risk factors. There is considerable
uncertainty in the estimated concentrations, which could
either overstate or understate the true concentrations (see
Chapter -IV). Unit cancer risk factors combine CAG potency
estimates with EPA exposure assumptions. The CAG potency
estimates provide a plausible upper limit to the cancer risk
of a compound (see Appendix A); however, the true value of the
risk is unknown and may be as low as 'zero. The exposure
assumptions are extremely conservative in that they assume
continuous exposure to outdoor air for 70 years. Because of
the generally conservative bias in the information, it is
highly unlikely that the true risks would be as high as the
estimates, and they could be considerably lower.
2 The POM risk estimates are, in part, based on source-specific
unit cancer potency factors that have not undergone extensive
peer review. Thus, these numbers are subject to change.
J Totals may not sum because of rounding.
35
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-9. Area-Wide Annual Excess Cancer Incidence Using
Available Monitoring Data (1, 2)
RESULTS INTENDED FOR POLICY DEVELOPMENT ONLY
POLLUTANT
(Weight of Evidence)
ANNUAL EXCESS
CANCER INCIDENCE
ANNUAL EXCESS
CANCER INCIDENCE
Arsenic [Bl]
Benzene [A]
Benzo(a)pyrene [B2]
Cadmium [Bl]
Carbon tetrachloride [B2]
Chloroform [B2]
Chromium VI [A]
Ethylene dichloride [B2]
Propylene dichloride [C]
Ethyl benzene [N/A]
Lead [N/A]
Methylene chloride (B2]
Methyl isobutyl ketone [N/A]
Perchloroethylene [B2]
Toluene [N/A]
Methyl chloroform [N/A]
Trichloroethylene [B2]
Vinyl chloride [A]
Xylene [N/A]
0.16
1.72
0.06
0.05
0.08
0.73
2.74
0.37
0.20
N/A
N/A
0.22
N/A
5.44
N/A
N/A
0.03
N/A
N/A
0.20
0.20
<0.01
0.09
<0.01
0.02
0.53
0.01
N.A.
N/A
N/A
0.20
N/A
0.04
N/A
N/A
0.03
N.A.
N/A
TOTAL
11.79
1.32
N/A: Not applicable; not a proven human carcinogen
N.A.t Not available
(1) Source of monitoring data: See Appendix B.
(2) This study uses conservative estimates of increased cancer risk
from ambient (i.e., outdoor) exposure to establish priorities among
pollutants and sources. The risk estimates are calculated using
modelled or monitored concentrations and EPA unit cancer risk
factors. There is considerable uncertainty in the estimated
concentrations, which could either overstate or understate the true
concentrations (see Chapter IV). Unit cancer risk factors combine
CAG potency estimates with EPA exposure assumptions. The CAG
estimates provide a plausible upper limit to the cancer risk of a
compound (see Appendix A); however, the true value of the risk is
(see Appendix A); however, the true value of the risk is unknown
and may be as low as zero. The exposure assumption are extremely
conservative in that they assume continuous exposure to outdoor
air for 70 years. Because of the generally conservative bias
in the information, it is highly unlikely that the true risks
would be as high as the estimates, and they could be
considerably lower.
36
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-10. Receptor Locations Warranting Further Investigation for
Noncancer Effects: Pollutant-Specific (1)
RESULTS INTENDED FOR. POLICY DEVELOPMENT ONLY
RECEPTOR
LOCATION
REFINED GRID
CONCENTRATION
TO THRESHOLD
RATIO (2)
POLLUTANT
Of CONCERN
NONCANCER
EFFECT (3)
4342.23
4342.25
4342.23
4342.75
4342.75
4342.75
364.75
367.25
369.75
364.75
367.25
369.75
1.1
1.6
5.0
1.3
1.1
1.4
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
DISCRETE SITE
4339.85
4340.22
4340.65
4340.85
4341.15
4341.41
4342.00
4342.52
4343.00
4343.15
4350.92
4350.92
4350.92
4350.92
4350.92
4350.92
374.88
374.95
375.10
375.31
375.30
375.30
374.25
374.00
374.12
373.90
370.55
370.55
370.55
370.55
370.55
370.55
1.7
1.6
1.3
1.2
1.2
1.1
1.3
1.2
1.1
1.1
1.2
1.5
6.2
1.5
6.2
1.5
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
XYLENE
XYLENE
XYLENE
XYLENE
XYLENE
XYLENI
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
LIVER
KIDNEY
REPRODUCTIVE
NEUROLOGICAL
FETAL/DEVELOPME*
BLOOD
(1) The noncancer effects analysis did not consider personal exposures
(e.g., indoor and vock).
(2) The threshold values underlying these ratio calculations ace based
on data of uneven quality. Furthermore, exceedance of a threshold
value of greater than one does not necessarily indicate severity
of effect; it simply indicates the need to further explore these
exposure levels. For the purpose of this study, it vae assumed
that a ratio greater than or equal to 0.95 vas equivalent to *1*
through rounding.
(3) The blood effect thresholds for benzene and xylene have not yet
undergone peer review, thus the results are subject to change.
37
-------
Excerpted directly from the Baltimore Integrated
Environmental Management Project: Phase II
Table 3-11. Receptor Locations Warranting Further Investigation for
Noncancer Effects: Complex Pollutant Mixtures (1)
RESULTS INTENDED FOR POLICY DEVELOPMENT ONLY
DISCRETE
RECEPTOR
LOCATION
4342.93
4343.20
4343.45
4343.58
4343.70
4346.27
4346.68
362.98
362.95
362.78
362.59
362.28
367.88
367.77
HAZARD
INDEX (2)
1.0
1.0
1.0
1.0
0.96
1.0
0.95
PRIMARY POLLUTANT
OP CONCERN
(RATIO VALUE)
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
BENZENE
(0.92)
(0.92)
(0.94)
(0.93)
(0.88)
(0.93)
(0.87)
NONCANCER
EFFECT
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
BLOOD
(1) The noncancer effects analysis did not consider personal exposures
(e.g., indoor and work).
(2) The threshold values underlying these hazard indexes are based
on data of uneven quality. Furthermore, exceedance of a threshold
value or a hazard index of greater than one does not necessarily
indicate severity of effect: it simply indicates the need to further
investigate these exposure levels. For the purpose of this study,
it v&s assumed that a ratio greater than or equal to 0.95 was
equivalent to *1* through rounding. The hazard index represents the
sum of all pollutant-specific ratios with the samt systemic effect
at a particular location.
(3) The blood effect threshold for benzene has not yet undergone peer
reviev, thus the results are subject to change.
38
-------
Excerpted directly from the Baltimore Integrateu
Environmental Management Project: Phase II
Table 3-12. A Comparison of Predicted Average Lifetime Individual Cancer Risk in Baltimore with Other Studies
RESULTS INTENDED FOR POLICY DEVELOPMENT ONLY
POLLUTANTS
(WEIGHT OF
EVIDENCE 1
Arsenic (A)
Benzene [A]
Benzo-a-
pyrene [B2]
Cadmium [Bl]
Carbon Tet. (B2]
Chloroform (B2]
Chromium VI [A]
l.ZDichloro-
ethane [B2]
Ethylene
Dibrood.de (B2)
Ethylene
Oxide [Bl]
Formaldehyde [Bl]
Methylene
Chloride [B2]
Perchloro-
ethylene [C]
POMs [N.A.]
Trichloro-
NESUAPS
STUDY
l.AE-06
9.8E-06
4.9E-06
4.2E-06
<7.00E-07
7.7E-06
1.3E-OS
8.4-06
1.5E-05
<7:OOE-07
<7.00E-07
2.8E-06
35
COUNTY
STUDY
1.4E-03
2.7E-05
1.4E-06
1.4E-06
2.8E-07
2.1E-07
2.0E-05
2.8E-06
1.4E-06
1.5E-05
9.8E-06
1.1E-05
AMBIENT
AIR
QUALITY
STUDY
1.8E-05
7.6E-05
1.4E-06
4.2E-06
2.6E-OS
3.2E-OS
7.AE-OS
5.8E-OS
7.7E-06
7.7E-06
SOUTH
COAST
STUDY
l.OE-09
3.9E-OA
6.7E-06
l.OE-08
7.1E-04
5.0E-08
2.4.E-05
3.0E-06
SANTA
CLARA
IEHP
1.5E-05
2.0E-05
4.0E-06
l.OE-05
6.0E-08
2.0E-OS
2.0E-07
2.0E-06
6.0E-07
2.0E-06
l.OE-07
KANAVHA
IEMP
1.3E-05
2.9E-05
4.3E-06
2.0E-06
1.5E-05
1.4E-04
5.9E-OA
2.6E-05
1.5E-06
1.8E-06
PHILA BALT
IEMP STUDY
8.6E-06
1.9E-05 8.7E-06
9.9E-09
3.8E-06
1.5E-06 3.7E-09
4.6E-06 9.0E-07
2.2E-05
2.6E-06 2.1E-07
1.7E-07
8.4E-08
8.1E-06
8.7E-06
1.7E-06 1.7E-0<:
8.9E-OS
1.3E-06 1.5E-06
ethylene [B2]
-------
tnie risk. This conclusion was drawn because of the generally conservative bias in the underlying
data and assumptions used in the risk assessment.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
40
-------
3.3 The Detroit Study12
Report Title/Date: The Transboundary Air Toxics Study: Final Summary Report,
December, 1990
Conducted by: Engineering-Science, Inc. for U.S. EPA Region 5, Air and Radiation
Division
Background and Methodology: The purpose of the Transboundary Air Toxics Study was
twofold. The primary purpose of the study was to evaluate the health risks of non-criteria, toxic air
pollutants on urban populations. The second purpose of the study was to estimate the relative impact
that atmospheric deposition of these toxic substances has on the ecosystem of the Great Lakes area.
The remainder of this summary will not address the deposition assessment but, instead, will only deal
with the risk analysis portion of the study.
This study serves to evaluate the pollutants which contribute to increased cancer risks and the
relative contributions of various source types to that increased risk in the Transboundary area. The
study area focused on the counties that include the industrialized areas within the region. A total of
10 counties (7 in Michigan and 3 in Ontario) were chosen to be included. The study area has a
population of 4,285,000 (499,000 on the Canadian side of the border) and a mix of industrial and
other urban sources. The area was limited to roughly 50 kilometers (km) from Lake St. Clair and the
Detroit and St. Clair Rivers due to the types of modeling techniques planned. The grid system
employed is the same as that developed for the National Acid Precipitation Assessment Program
(NAPAP) inventory. .The grid cells for the NAPAP inventory were 20 km on a side (square) but for
this study were further divided into smaller grid cells (10 km, 5 km, and 2.5 km), for a total of 384 "'
cells, to provide greater detail to the study.
Several criteria were chosen to identify the pollutants to investigate in this study. Generally,
the pollutants chosen were substances known to be atmospherically deposited into the region or
substances known to pose a carcinogenic risk or other significant human health risk. A total of 57
substances was decided on. Based on the types of source categories that produce these 57 substances,
a list of point and area sources was developed. Table 3-13 lists these source categories and breaks
them down between point sources and area sources.
The initial source for emissions data was the volatile organic compound (VOC) and total
suspended paniculate (TSP) emissions inventory from EPA's National Acid Precipitation Assessment
Program (NAPAP). Other data sources that were utilized included databases from EPA (including the
NESHAP's database) and other government agencies from both the U.S. and Canada. The EPA's Air
Emissions Species Manual provided emissions profile data for each source category. From these
sources, specific pollutant emissions were estimated and modeled.
Once the emissions inventory was prepared, air dispersion modeling proceeded. The
Industrial Source Complex-Long Term (ISCLT) Model was used for air dispersion modeling of both
point sources and area sources for this study. Meteorological data was taken from Detroit
Metropolitan Airport for the years 1982-1986. Background concentrations were assumed to be zero
except for formaldehyde (assumed to be 2.23 /zg/m3) and carbon tetrachloride (assumed to be 0.76
/tg/m3). These background concentrations were adopted from the Southeast Chicago study.
41
-------
Excerpted directly from the Detroit Transboundary Air Toxics Study
Table 3-13. Summary of Source Categories
Point Sources
Chemical production
Coke and charcoal combustion
Coke ovens/iron and steel
Fuel combustion
Metals production
Motor vehicle manufacturing
Refineries
Waste incineration
Area Sources
Additional miscellaneous NAPAP categories
Architectural coatings
Auto refinishing
Cold degreasing
Cooling towers
Dry cleaning
Gasoline marketing
Mobile sources
Pesticides
Residential oil combustion
Residential wood combustion
42
-------
Unit risk factors for the specific pollutants were obtained from several different EPA offices
and 16 of these risk factors have received agency-wide review. The unit risk factors used in this
study reflect the best judgements of the EPA scientists, however, the uncertainties in the unit risk
factors are probably the greatest uncertainties in the study.
Cancer risk at a given location was estimated by multiplying, for each pollutant, the modeled
concentration times the unit risk factor of that pollutant, and then summing for all pollutants. Excess
cancer incidence is a population-oriented measure. By multiplying the risk in a given area times the
population of that area, an estimate of the number of excess cancer cases is obtained.
Results: Of the 57 pollutants initially considered, this study found atmospheric emissions of
27 pollutants which EPA considers carcinogenic and 15 pollutants which EPA does not consider
carcinogenic. This study suggests that for the study area, roughly 373 cancer cases over 70 years (or
slightly over 5 cases per year) can be attributed to air pollution. The average lifetime risk across the
entire study area is estimated at roughly 9xlO~3.
Table 3-14 provides a summary of additional cancer incidence, by pollutant, for a 70 year
exposure period. Figure 3-1 shows the percentage of total excess cancer incidence can be attributed
to the individual pollutants. As can be seen from Figure 3-1, over 90 percent of the predicted excess
cancer incidence results from 7 specific pollutants: formaldehyde, coke oven emissions, 1,3-
butadiene, carbon tetrachloride, hexavalent chromium, total polycyclic organic matter, and dioxins.
The largest single contributor is formaldehyde (36.1 percent). Most of this formaldehyde is due to
background concentrations from photochemical generation.
The next largest contributors to total predicted incidence are coke oven emissions, 1,3-
butadiene, and carbon tetrachloride. Coke oven emissions are from one particular source (steel and
coke manufacturing). Motor vehicles are the major contributor for 1,3-butadiene, while carbon
tetrachloride is primarily a background pollutant, due to historical emissions that have accumulated in
the atmospheric over the years because of carbon tetrachloride's low reactivity.
The next most important contributors to increased cancer incidence are hexavalent chromium,
polycyclic organic matter (POM), and dioxins. While hexavalent chromium is emitted from many
source categories, almost half of the hexavalent chromium in this study came from chrome plating
operations. Dioxins and POM come from a variety of fuel combustion sources. The remaining
cancer incidence (roughly eight percent) results from the other 20 carcinogenic pollutants identified in
the study and from the various source categories.
All emissions were divided among eight broad source groupings: photochemical generation,
vehicle manufacturing, steel/coke manufacturing, other manufacturing, highway vehicles, fuel
combustion, background concentrations, and residential/miscellaneous. Figure 3-2 presents the cancer
incidence attributable to each of these source groupings.
Cancer incidence and individual risk results were also presented for the grid cell of maximum
risk on both the U.S. and Canadian sides of the border and for the entire study area. A lifetime
cancer risk of 1.2x10"* is estimated for this particular grid cell. Tables 3-15 and 3-16 present the
contributions by pollutant and source groupings to the risks in the peak risk area. These tables show
the relative importance of coke oven emissions over other pollutants and source groupings at the peak
risk area.
43
-------
Excerpted directly from the Detroit Transboundary Air Toxics Study
Table 3-14. Summary of Estimated Excess Cancer Cases by
Pollutant Across the Study Area
Substance Total
Formaldehyde 134.7
Coke oven emissions 61.0
1,3 butadiene 56.5
Carbon tetrachloride 52.1
Chromium 13.5
POM 123
Dioxins 12.0
Arsenic 7.4
Beryllium 5.8
Asbestos 5.7
Benzene 5.0
Gasoline vapors 3.0
Cadmium 1.4
Beiuo(a)pyrene 0.7
Ethylene dibromide 0.6
Vinyl chloride 0.4
Trichloroethylene 0.3
Perchloroethylene 0.2
PCBs 0.1
Styrene 0.1
All others* 0.1
TOTAL 372.9
Chloroform, ctbyleoc oxide, acrylonitrile, methylene chloride, chlordane, heptachlor, and epichlorohydrin.
44
-------
Excerpted directly from the Detroit Transboundary Air Toxics Study
Figure 3-1. Estimated Excess 70-Year Incidence, Area Wide by Pollutant Contribution
CllROMIUM+6-3.6%
COKL OVENS-16.4%
OIOXIN-3.2%
FORMALOEIIYDE-36. 1%
CARBON TET-14.0%
ALL OTHER-8.3%
1.3-BUTADIKNE-I!).I1
•OM-3.3\
-------
Excerpted directly from the Detroit Transboundary Air Toxics Study
Figure 3-2. Estimated Excess 70-Year Incidence, Area Wide by Source Grouping Contribution
HIGH HAY VEHICLES-11.4%
OTHER MFC-8.8%
FUEL COMBUSTN-6.5\
.P-
CT>
IIOTOCHEM '-38.71
BACKGRND CCL«-15.2»
VEHICLE MFC-1.2t
STEEL/COKE MfG-18.1
RESIDENT1AL/HISC<0. It
*PHOTOCHEMICAL GENERATION OF FORMALDEHYDE FROM EMISSIONS
FROM VEHICLES, INDUSTRIAL FACILITIES, AND MISCELLANEOUS AREA SOURCES
-------
Excerpted directly from lite Detroit Transboundary Air Toxics Study
Table 3-15. Pollutant Contribution to Risk at Grid Cell with
Highest Individual Risk*
Pollutant %
Coke oven emissions 37^
Formaldehyde 24.6
1.3 butadiene 119
Carbon tetrachloride 9,4
Arsenic 2.8
Chromium 2.6
POM 2.6
Dioxins 2.6
Beryllium 2.0
Asbestos U
Benzene 1_2
Gasoline vapors 0.6
Cadmium 0.5
Benzo(a)pyrene 0.1
Ethylene dibromide 0.1
Vinyl chloride 0.1
Trichloroethylene 0.1
Others'" QJ.
TOTAL
id CcU 136.
PcrchJorocthylene, PCBs, chloroform, styreoe, acryloaJtiUc, ethylene oride, methylene chloride, chlordane,
heptachJor, and epichlorobydria.
47
-------
Excerpted directly from the Detroit Transboundary Air Toxics Study
Table 3-16. Source Grouping Contribution to Grid Cell with
Highest Individual Risk
Source Grouping %
Steel and coke manufacturing 40.5
Photochemically generated formaldehyde 26.8
Background carbontetrachloride 10.6
Highway vehicles 9_1
Other industry . 5 g
Fuel combustion 5^
Vehicle manufacturing j^j
TOTAL 10o.o
d cell 136.
Residential and miscellaneous sources contribute less than 0.1%.
48
-------
Limitations: One strength of this study is the number of pollutants that were analyzed (27
carcinogenic and 15 non-carcinogenic) relative to the other studies. The pollutants that were analyzed
also cover those pollutants that cause the highest increased cancer incidence in the other studies. A
weakness of the study was that it did not break down the sources associated with each pollutant.
What is presented are two breakdowns: (1) a breakdown of total risk by individual pollutants, and (2)
a breakdown of total risk by source groupings. The other studies segregated individual pollutants by
their sources. While the study does account for background levels of carbon tetrachloride and
formaldehyde, it does not account for background levels of any other pollutants [most other studies
don't either] or for in-migration of pollutants from other areas. The study also does not segregate
formaldehyde concentrations between background levels and those levels produced by the sources.
Another weakness of the study involves the use of the ISCLT model for modeling area
sources. Ambient concentration estimates from ISCLT may be less than the estimates of the
Climatological Dispersion Model (CDM) by a factor of two or more under some assumptions.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
3.4 The Houston Study"
Report Title/Date: A Risk Assessment Based Air Enforcement Strategy for Harris
County, Texas, 1988
Conducted by: U.S. EPA Region 6
Background and Methodology: The purpose of this study was to provide insight into the
process of assessing the health impacts of chemical compounds in developing effective
compliance/enforcement efforts using existing regulations. Harris county was selected because it
contains Houston, Texas, the fifth most populated city in the U.S. Immediately adjacent to downtown
Houston is the Houston Ship Channel which contains hundreds of companies that comprise what has
been termed the largest petrochemical complex in the world. Thus, the study area has the potential
for the greatest impact of chemical compounds on the largest population within EPA Region 6 and
possibly the entire U.S.
The emissions inventory of air pollutants for Harris County was examined with respect to five
groups of source categories. The five categories are Texas Air Control Board (TACB) permitted
point (stack) sources, TACB permitted fugitive (area) sources, small non-permitted and area sources,
mobile sources, and accidental releases. The TACB emissions inventory has generally been
developed and maintained in terms of non-methane volatile organic compounds (VOC) emissions for
permitted sources. VOC emissions were speciated based on speciation factors associated with the
Source Classification Code (SCC) of the individual emission point. The individual chemical
compounds examined for this study are the chemical compounds defined as hazardous, extremely
hazardous, or toxic chemical substances in Title III Sections 302 and 313 of the Superfund
Amendments and Reauthorization Act (SARA), the Resource Conservation and Recovery Act
(RCRA), and the Comprehensive Environmental Response, Compensation, and Liability Act
(CERCLA), or any amendments to these acts. The compounds considered as having the greatest
potential for putting the population at risk were then selected for dispersion modeling and a risk
assessment study.
49
-------
All permitted (both stack and fugitive) sources within a 20 mile radius of the center of the
ship channel were evaluated. The speciation of the permitted point sources resulted in 35.1 percent of
the VOC emissions being classified into individual chemical compounds. The speciation of the
permitted fugitive sources resulted in 37.2 percent of the VOC emissions being classified into
individual chemical compounds.
The categories examined for non-permitted area sources included: emissions from
architectural coatings, automobile refinishing, consumer solvents, cutback asphalt operations,
degreasing operations, drycleaning, fuel combustion, gasoline transfer, graphic arts operations, and
solid waste disposal. The speciation of the non-permitted area sources resulted in 37.2 percent of the
VOC emissions being classified into individual chemical compounds.
The mobile emissions inventory included categories of aircraft, automobiles, railways, and
vessels. The EPA National Emissions Data System (NEDS) which utilizes the Mobile 3 model was
used to calculate the mobile emissions. Standard emission factors were then applied to the mobile
emissions which resulted in 28.9 percent of the VOC emissions being classified into individual
chemical compounds.
The TACB requires that all permitted facilities report air pollutants emitted outside of the
normal course of operations. These emissions are generally associated with plant start-ups, shut-
downs, maintenance operations, and upsets. These upset reports for TACB were examined for the
period June 1985 through May 1986 to determine the magnitude and frequency of the accidental
releases from permitted facilities in the area. There were 1,372 upsets reported for this time period
but only 307 of the reports contained data sufficient to quantify emissions. It appears that emissions
from upsets may be greatly underestimated for this study.
The compounds selected for dispersion modeling were butadiene, methyl chloride, benzene,
formaldehyde, toluene, and xylene. Toluene and xylene are not carcinogenic and have no unit risk
factors associated with them so they are omitted from further discussion. The model employed was
the Industrial Source Complex Model in its long term mode (ISCLT). The results represent the
annual average concentrations. Receptors were selected at census tract centroids so that the number
of people exposed to the annual concentrations were known. Unit risk factors were applied to each
predicted annual concentration for each chemical modeled to yield the upper bound number of cancer
cases for each of the 182 receptor areas. The unit risk factors employed for benzene, butadiene,
formaldehyde, and methyl chloride were 8.3E-6; 2.8E-4; 1.3E-5; and 3.6E-6 lifetime risk per
ug/m*3, respectively. The modeling was conducted for each source category so that the predicted
cancer cases could be compared by source category for each chemical.
Results: Table 3-17 presents the predicted excess cancer cases for the study assuming a
population of 2,055,066. As can be seen from the table, butadiene is associated with the highest
predicted excess cancer incidence. Fugitive sources are associated with the highest predicted excess
cancer incidence.
Limitations: The most obvious limitation to this study is the limited number (4) of pollutants
analyzed. Of the 11 studies whose risk analyses were reviewed, only one study examined fewer
pollutants. The grouping of the source categories also limits the inferences that can be made towards
the effects that specific source categories have on increased cancer incidence.
50
-------
Excerpted directly from the Houston Risk Assessment Based
Air Enforcement Strategy for Harris County, Texas Study
Table 3-17. Upper Bound Annual Predicted Cancer Cases in
Harris County for a Population of 2,055,066
Form- Methyl
Source Category Benzene Butadiene aldehyde Chloride' Total
Point Sources
TACB Permits
Point Sources
Speciated VOC
Fugitive Sources
TACB Permits
Fugitive Sources
Speciated VOC
Small Area Sources
Non-Permitted
Mobile Sources
.031 .139
.017
.013 1.682
.015 .047
.00021
.0023
.060
.072
.002
.569
.00013
.0047
.004 .234
.00038 .089
.020 1.716
.003 .634
.00033
.0071
TOTAL .078 1.867 .708 .027 2.681
Actual Cancer Mortality in Harris County (1985) is 3,422
51
-------
The use of SCC profiles and VOC speciation profiles has several weaknesses in that they
under-represent many chemical manufacturing operations; they do not include many less common but
potent substances; they do not address metals; and speciation profiles are not considered the best
approach for estimating air toxics emissions. It also appears that emissions from spills and accidental
releases may be under-represented due to the limited data that were good enough to estimate
emissions from these episodes.
It was emphasized throughout this report that the findings, and the analysis in general, were
strictly preliminary. It was concluded that the emissions inventory of the individual chemical
compounds, as used in the study, was only of a preliminary nature. The compilation of a more
accurate inventory by both the reporting facilities and the TACB is needed in order to perform a more
definitive risk assessment.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
3.5 The Twin-City Study17
Report Title/Date: Estimation and Evaluation of Cancer Risks from Air Pollution in the
Minneapolis/St. Paul Metropolitan Area, March 1992
Conducted by: Minnesota Pollution Control Agency, Air Quality Division
Background and Methodology: The purpose of this study was to analyze sources of air
pollutants suspected or known to cause cancer, and to estimate the health risk from exposure to these
pollutants. The study was designed to estimate and characterize emissions of these selected
pollutants, and to model the resulting human health risk.
The study area consisted of two overlapping sections - the source area and the receptor area.
In an attempt to include all major sources of toxic air pollutants, a source area much larger than the
receptor area was defined. The receptor area population was estimated at 1,227,584. In order to use
air dispersion modeling the study area was divided into a 1 km by 1 km square grid system.
Table 3-18 lists the carcinogenic pollutants inventoried and the unit risk factors associated
with each pollutant. These are air pollutants for which dose-response relationships for carcinogenicity
have been estimated by the U.S. EPA. It should be noted that because there were no monitoring data
available for this study, background concentrations of pollutants (e.g., carbon tetrachloride) and
secondary formation of pollutants (e.g., formaldehyde) are not considered in this study.
The emissions inventory consisted of point sources, area sources, and mobile sources. For
the purposes of this study, only larger sources with adequate emissions point information were treated
as point sources. Specifically, facilities with data from EPA's National Emissions Data System
(NEDS) or that reported emissions to EPA's Toxic Release Inventory (TRI) were treated as point
sources. One hundred eighty one (181) facilities were included in the database.
Emissions data from TRI was considered superior to NEDS emissions data. If a facility had
TRI emissions data, these data were used as reported (because the emissions are compound specific).
The NEDS database consists of criteria emissions data. Use of NEDS entailed speciating the reported
52
-------
Excerpted directly from the Twin-City Estimation and Evaluation of Cancer Risks
from Air Pollution in the Minneapolis/SL Paul Metropolitan Area Study
Table 3-18. Pollutants Included in Inventory and Their Unit Risk Estimates
I. CARCINOGENS
ORE* CAS
COMPOUND
WEIGHT OF
EVIDENCE
ACETALDEHYDE B2
ACRYLONITRILE 81
ARSENIC A
BENZENE A
BENZO(a)PYRENE (BaP) B2
BERYLLIUM B2
1,3-BUTADIENE B1
CADMIUM 81
CARBON TETRACHLORIDE B2
CHLOROFORM B2
CHROME-6 A
COKE OVEN EMISSIONS A
1,2-DICHLOROETHANE (EtMen* Oiehiorid.) B2
1,1 -DICHLOROETHYLENE tvmyiiden. Chiond.) C
EPICHLOROHYDRIN 82
ETHYLENE DIBROMIDE B2
ETHYLENE OXIDE B1
FORMALDEHYDE 81
GAS VAPORS (MARKETING) B2
HEXACHLOROBENZENE B2
METHYLENE CHLORIDE B2
NICKEL *• A
PERCHLOROETHYLENE rr.uechioro.th.n«) B2
POLYCYCLIC ORGANIC MATTER (POM)
PROPYLENE OXIDE B2
STYRENE B2
2,3,7,8-TETRACHLORODIBENZO-P-DIOXIN B2
TRICHLOROETHYLENE B2
VINYL CHLORIDE A
II. CARCINOGENS USING COMPARATIVE POTENCY FOR POM
URE* SUBSTANCE
2.2E-6
6.8E-5
4.3E-3
8.3E-6
1.7E-3
2.4E-3
2.8E-4
1.8E-3
1.5E-5
2.3E-5
1.2E-2
6.2E-4
2.6E-5
5.0E-5
1.2E-6
2.2E-4
1.0E-4
1.3E-5
6.6E-7
4.9E-4
4.7E-7
O.OE-0
5.8E-7
1.7E-3
3.7E-6
5.7E-7
3.3E + 1
1.7E-6
4.1E-6
75-07-0
107-13-1
7440-38-2
71-43-2
50-32-8
7440-41-7
106-99-0
7440-43-9
56-23-5
67-66-3
7440-47-3
107-06-2
75-35-4
106-89-8
106-93-4
75-21-8
50-00-0
118-74-1
75-09-2
127-18-4
75-56-9
100-42-5
1746-01-6
79-01-6
75-01-4
3.0E-5***
2.9E-4***
1.0E-5***
- DIESEL PM
- GASOLINE PM
- WOOD STOVE PM
•URE - UNIT RISK ESTIMATE (LIFETIME RISKAJG/CUBIC METER) (th» probability of contracting c«nc»r *i lh«
rttult of coniunt txpotur* ovtr 70 y«*r« to an ambient concentration of 1 microQfun per cubic rrwttr)
"URE (or Nick*! U fitted M 0.0 E-0 b*e*ut« It It «ttwn*d in ttvt itudy th*t non* of th« nicktl •mintd is in th«
cueirvogenic riTintry (tutt or (ubtutfid* form*.
• "PARTICLE UNIT RISK ESTIMATE (LIFETIME RISKAJG/CU8IC METER)
53
-------
VOC emissions data into specific pollutants. The species profiles employed were obtained from the
U.S. EPA's Air Emissions Species Manual, Volume I and Volume II.
Three other sources of data were used in the analysis of point sources. These included:
emissions data from the Ford Motor Company Twin Cities Assembly Plant; survey data on ethylene
oxide from local hospitals; and emissions test data from the Hennepin Energy Resource Corporation.
These source specific data were considered superior over both the NEDS and TRI data.
The area source inventory also consisted of estimating emissions and identifying locations.
Emissions were estimated based on information from a variety of sources (e.g., emission factors and
employment data for specific Standard Industrial Classification [SIC] codes). The emissions were
then assigned a location. Generally the emissions estimates resulted in area wide or county wide
emissions values, which were subsequently allocated to census tracts.
Mobile source emissions were estimated using MOBILE4 and then speciating the results.
MOBILE4 calculates the emissions of hydrocarbons, carbon monoxide, and nitrogen oxides from
gasoline fueled and diesel highway motor vehicles. The emissions estimates are dependent upon a
variety of area-specific conditions such as ambient temperature, speed, and mileage accrual rates.
The emissions inventory process resulted in approximately 37 million pounds/year of
carcinogens. Figures 3-3 and 3-4 show the breakdown of the total emission inventory by source type
and pollutant respectively.
Two dispersion models were used in this study — the Industrial Source Complex-Long Term
(ISCLT) for point sources, and Climatological Dispersion Model (CDM) for area and mobile sources.
The receptors in this study were defined as the centroids of the census tracts within the modeling area
for all source types. There were 366 census tracts in the source area and 284 in the receptor area.
The dispersion modeling resulted in area-wide, upper-bound risk estimates.
The risk assessment for this study was limited to cancer effects. Risk at a particular census
tract was estimated by multiplying the modeled concentration of individual pollutants times the unit
risk factor for that pollutant and summing for all pollutants. Excess cancer incidence at a census tract
was then calculated by multiplying the individual risk by the population of the census tract. It follows
that excess cancer incidence for the entire study is the sum of the excess cancer incidence for each of
the census tracts.
Results: This study estimates 222 excess cancer cases over a 70 year period in the receptor
area for the pollutants and sources studied. This translates to slightly over three excess cancer cases
per year for the receptor area. The average population risk is the total excess cancer incidence (222)
divided by the population (1.2 million) and divided by the 70 years of assumed exposure. This
results in an population risk of 2.64 excess cancer cases per year per million residents. The area
wide individual lifetime risk is the total excess cancer incidence (222) divided by the population (1.2
million). This results in 1.84E-04 chance of an individual developing cancer in their lifetime from
continuous exposure to the air pollutants studied.
Figures 3-5 and 3-6 show the breakdown of total cancer incidence by source category and
specific pollutant, respectively. As can be seen in Figure 3-5, road vehicles are attributable for 61
54
-------
Excerpted directly from the Twin-City Estimation and Evaluation of Cancer Risks
from Air Pollution in the Minneapolis/St. Paul Metropolitan Area Study
Figure 3-3. Sources of Carcinogenic Pollutant Emissions in the 7-County
Twin Cities Metropolitan Area
Smafl Commercial
Incinerators
18* Mi»c. Area Sources
< 1% Chrome Plater*
Point Sources
< 1% Heating
Wood Stoves & 39%
Fkeplaces
Road Vehicles
33%
55
-------
Excerpted directly from the Twin-City Estimation and Evaluation of Cancer Risks
from Air Pollution in the Minneapolis/St. Paul Metropolitan Area Study
Figure 3-4. Emissions of Carcinogenic Pollutants in the 7-County
Twin Cities Metropolitan Area
23%
Others
1%
1,3-Butadiene
<1% Chrome + 6
Benzene
POM
<1% Formaldehyde
9%
Diesel PM
18%
Gasoline PM
1%
Wood Stove PM
37%
56
-------
Excerpted directly from the Twin-City Estimation and Evaluation of Cancer Risks
from Air Pollution in the MinneapolisJSt. Paul Metropolitan Area Study
Figure 3-5. Estimated Excess Cancer Incidence by Source Category
Total Estimated Excess 70 Year Cancer Incidence = 222 = 2.64
Excess Cancer Cases/Year/Million Population in Receptor Area
10%
Chrome Platers
1% Point Sources
7% Heating
1%
Misc. Area Sources
3% Small Commercial
Incinerators
17%
Wood Stoves &
Fireplaces
57
Road Vehicles 61%
Misc. Area Sources Includes:
o Comfort Cooling Towers 37%
o Degreaslng 26%
0 Small Hospitals 16%
o Gas Marketing 6%
o Industrial Cooling Towers 6%
o Research Labs 5%
o Dry Cleaners 1%
o Solvent Usage 1%
o Surface Coating 1%
o Waste Oil Burning 1%
-------
Excerpted directly from the Twin-City Estimation and Evaluation of Cancer Risks
from Air Pollution in the Minneapolis/St. Paul Metropolitan Area Study
Figure 3-6. Estimated Excess Cancer Incidence by Pollutant
Total Estimated Excess 70 Year Cancer Incidence = 222 = 2.64
Excess Cancer Cases/Year/Million Population in Receptor Area
4%
Benzene
Formaldehyde
POM
3%
Other*
Chroma+ 6
15%
Wood Stove PM
99.7% of Thlt Excew
Incidence U from Rw
Vehicle*
58
-------
percent of excess cancer incidence. The next most important source is woodstoves/fireplaces with 17
percent of excess cancer incidence.
The pollutants contributing the most to excess cancer incidence are diesel particulate, gasoline
particulate, and woodstove particulate, contributing 27 percent, 15 percent, and 15 percent
respectively. This cancer incidence is associated with the polycyclic organic matter (POM) fraction of
the particulate.
Limitations: The strengths of this study include the number, and type, of pollutants and
source categories that were analyzed. The results of the study fall within the range of results obtained
from other, similar studies.
One weakness of the study was the fact that secondary formaldehyde and background carbon
tetrachloride were not considered. Inclusion of these pollutants could increase the estimated excess
cancer incidence for the study. Another weakness is the use of speciation factors for extrapolating air
toxics emissions from VOC or particulate emissions.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
3.6 The Texas Study16
Report Title/Date: An Assessment of the Urban Air Toxics Problem in Texas; June 1989
(second draft)
Conducted by: Texas Air Control Board (Now called the Texas Natural Resources
Conservation Commission), Research Division
Background and Methodology: This project was designed to be a screening effort using
available data sources to evaluate impacts from both point and area sources in two areas: Dallas-Fort
Worth and the Houston area, in Texas. It used computer models to predict ground level
concentrations of and assess public exposure to a selected list of compounds the potential health
effects of which were of concern. Results were analyzed to identify areas with predicted excessive
exposures. Ambient data for air toxics was also analyzed to identify any areas with monitored levels
considered excessive. In some cases, it possible to compare modeled and monitored levels to estimate
the relative contribution of point sources to the urban exposure. The adequacy of the available data
bases and the effectiveness of the agency's air toxics strategy were also reviewed.
The Houston area is highly industrialized, including one of the world's largest concentration
of petrochemical and related chemical process industries. Along the Gulf Coast a large fraction of the
nation's basic petrochemicals is produced. The Houston area has a strong thermal mixing of its
atmosphere and almost continuous low level winds, providing effective dilution.
Dallas-Fort Worth is less industrialized than Houston with approximately 3000 manufacturing
plants making apparel, building material, food, oil field supplies, electronic and other products. Fort
Worth has a diversified urban economy with aircraft, foods, mobile homes, electronics, chemicals and
plastics among the products of its over 1100 plants.
59
-------
Ambient monitoring data in Houston (extensive) and Dallas-Fort Worth (sparse) were
reviewed and compared with TACB's effects screening levels for each compound detected. TACB
screening levels for chronic exposure are typically based on 0.1 percent of the most appropriate
occupational exposure limit or on other toxicity information. Review of the monitoring data revealed
that several compounds were detected at concentrations that significantly exceeded the applicable
screening levels.
Assessing the impact of emissions from point sources consisted of 3 steps: describe the
population distribution in each county, estimate annual average ambient concentrations using
dispersion modeling, and use the calculated concentrations with the population data to calculate
population exposure.
Results - Dallas/Fort Worth: Of the four compounds assessed in Dallas County, 3 were
modeled or measured at concentrations of potential concern: asbestos, chromium and benzene.
Modeled asbestos levels of up to 0.009 ug/m3 were predicted, a 3-fold exceedence of the TACB
annual effects screening level for asbestos. Benzene levels of 2.5 ppb (annual average) were model-
predicted, a 3-fold exceedence of the effects screening level, representing the highest potential risk of
the compounds studied. Area sources (motor vehicles, solvent use, etc.) were found to be mostly
responsible for ambient benzene levels. Modeling of chromium levels in residential areas adjacent to
two sources predicted maximum annual concentrations in the range of 0.1 to 0.5 ug/m3, which is up
to a 20 fold exceedence above screening levels if all is assumed to be hexavalent in form. The
screening level is still exceeded, but by only 4 fold, if the chromium is assumed all trivalent. The
report authors state that most ambient chromium should be assumed to exist in the trivalent state.
Styrene was also considered, but model predictions indicate no possibility of levels exceeding
screening levels, and styrene is not detected in NMOC ambient sampling done in Dallas-Fort Worth.
The results are in Table 3-19 and summarized in Table 1-5 in Chapter 1.
Results - Houston/Harris County: The results of human exposure modeling were examined
to evaluate maximum predicted concentrations due to point sources and total inhaled dose. In
addition, dispersion modeling was conducted and the grid output examined to compare the
concentrations predicted to occur at the receptor nearest each of seven ambient monitoring stations.
The predicted versus measured concentrations are compared in Table 3-20 to give an idea of the
relative contribution of point sources to ambient concentrations in Harris County.
The results are summarized in Table 3-19. Of the pollutants considered, benzene,
formaldehyde, allyl chloride, acrylonitrile and ethylene dichloride exceed or equal the respective
significance levels, in order of extent of exceedence. For benzene, as in Dallas-Fort Worth, mobile
and other area sources are implicated as major contributors to ambient benzene levels. Regarding
formaldehyde, the study authors surmise that area sources and atmospheric reactions are perhaps more
important contributors to ambient formaldehyde than point sources, but this can not be adequately
assessed because of errors in the data base. Model-predicted allyl chloride levels were up to three
times the screening level, whereas ambient allyl chloride was detected at only one site at a level 4
times lower than model-predicted. No clear conclusions were drawn concerning allyl chloride other
than improved emissions data are needed. Dispersion and population modeling of acrylonitrile
predicted annual averages in residential areas adjacent to point sources at a level four times the effects
screening level. A comparison of predicted versus measured concentrations for three monitoring
studies is shown in Table 3-20. The report recommended that sampling be done near the major
60
-------
Excerpted directly from the Texas Assessment of the Urban Air Toxics Problem Study
Table 3-19. Results of the Texas Urban Air Toxics Assessment
Location Compound
Evaluated
Dallas County asbestos
benzene
chromium
styrene
ON Harris County acrylonitrile
»— «
benzene
carbon tetrachloride
chloroprene
allyl chloride
•thylen« dichloride
formaldehyde
epichlorohydrin
ethylene oxide
Monitored or
Predicted Levels
of Concern
(ug/ra3)
.009 (P)
8 (M)
0.1-0.5 (P)
-------
Excerpted directly from the Texas Assessment of the Urban Air Toxics Problem Study
Table 3-20. Predicted Versus Measured Concentrations of Compounds Evaluated
in Harris County (ug/m3)
HRM 1
COMPOUND
acrylonitrile
benzene
carbon tetrachlorid
chloroprene
allyl chloride
ethylBie dichlocide .6 <2
forwaJdehyde
SAMPLING STATION
HRM 3 HRM 4
HRM 7
HRM 8
TAMS
UATMP
P
.1
.4
.3
.4
.2
.6
./.
M
-------
acrylonitrile source to determine the actual public exposure. For ethylene dichloride, human exposure
modeling predicted maximum exposures in the range of 3 to 10 ug/m3, whereas the highest measured
concentration was 4.7 0 ug/m3, a slight exceedence of the screening level. Measured concentrations,
as shown in Table 3-20, are up to an order of magnitude higher than predicted by modeling at the
sampling site. Either other sources are implicated or point source emissions were under-reported. In
either case, the concentrations were not high enough to be of concern to the report authors.
Modeling/monitoring results show levels of carbon tetrachloride, chloropene, epichlorohydrin,
and ethylene oxide below screening levels, as shown in Table 3-19. The results from these studies
are summarized in Table 1-5.
Limitations: The emission inventory data bases used in this study were acknowledged to
very weak for area sources. Assessment of the area source contribution could only be inferred
through examination of ambient air monitoring data. Point/area source cutoffs were not specified in
the report.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
3.7 The EPA Noncancer Screening Study0"15
Report Title/Date: Toxic Air Pollutants and Noncancer Health Risks: Screening Studies.
External Review Draft.
Conducted by: EPA's Office of Air Quality Planning and Standards, Research
Triangle Park, NC, 27711. September 1990.
Background and Methodology: Two screening studies were conducted within the context of
this study to characterize the potential noncancer risks associated with exposure to toxic air pollutants,
each looking at slightly different aspects of the question. Each was conducted by comparing modeled
and/or monitored ambient concentrations to health reference (HRLs) levels and lowest-observed-
adverse-effect levels (LOAELs). [In this context, a LOAEL is the lowest dose or exposure level at
which a statistically or biologically significant effect is observed in the exposed population. A HRL is
the LOAEL divided by appropriate uncertainty factors to account for inter- and intra-species
variability and the use of a LOAEL rather than a no-observed-adverse-effect level (NOAEL).]
The first study was a national "broad screening study" which examined exposure to single or
multiple pollutants in ambient air based on exposure data inferred from ambient monitored pollutant
levels or emissions-modeled data from many areas of the country. Exposure data was based on
measured ambient concentrations for approximately 325 pollutants monitored throughout the U.S. and
modeled annual average ambient concentrations for approximately 40 pollutants emitted from more
than 3,500 facilities. Health data and quantitative exposure data were available for approximately 150
pollutants.
The second study involved a more detailed evaluation of a midwestern industrialized urban
county. This analysis expanded the number of chemicals evaluated in the broad screening study and
assessed the combined impact of multiple emission sources versus the impact of sources, singly.
Approximately 200 chemicals emitted from 122 point sources plus 9 area sources were evaluated.
63
-------
Health reference levels and LOAELs were compared to modeled pollutant concentrations in three
independent modeling exercises.
Results - Broad Screening Study: For those few chemicals with both health and exposure
data, noncancer health risks appeared to be of concern. For approximately half of these chemicals,
modeled and/or monitored levels exceeded health reference levels at numerous sites throughout the
country. Ambient levels for approximately one-third of these chemicals exceeded the health reference
level at more than 25 percent of the sites studies. These exceedences were associated with both short-
term and long-term ambient monitored and long-term modeled concentrations. Modeling of short-
term emissions was not performed because of data limitations. Less than 5 percent of the sites and
chemicals indicated ambient concentrations exceeding LOAELs.
The results of the broad screening study are summarized in Table 1-5 in Chapter 1 of this
report. The HAPs ranking of highest potential concern based on potential exceedences of HRLs at 25
percent or more of observed sites are (in no relative order or ranking): acetaldehyde, acrolein,
arsenic, benzene, beryllium, carbon disulfide, carbon tetrachloride, chloroform, ethylene oxide,
formaldehyde, hydrogen sulfide, methyl ethyl ketone, methyl methacrylate, methyl isocyanate,
nitrobenzene, perchloroethylene, phenol, phthalic anhydride, styrene, tetramethyl lead, toluene
diisocyanate, and vinyl chloride.
Because of the considerable uncertainty in the data sets used in this noncancer analysis,
and because of the difficulty in comparing noncancer endpoints, the pollutants are not ranked in
this study. In addition, it is unclear how to interpret the effects of a pollutant concentration
slightly above a specific health effects level.
Results - Urban County: Results suggested that a larger number of pollutants exceeded
health reference levels for short-term modeled concentrations than for long-term modeled
concentrations. Ambient concentrations were estimated to exceed health reference levels for long-
term concentrations predicted by EPA's Industrial Source Complex Long-Term Model (ISCLT) and
the Human Exposure Model (4 and 8 percent of pollutants, respectively) and short-term (24 hour)
concentrations predicted by the SCREEN Model (22 percent of pollutants). Estimated long-term
concentrations did not exceed any pollutant LOAELs. Estimated short-term (24 hour) concentrations
exceeded LOAELs for approximately 2 percent of the pollutants assessed. In general, proximity to
individual sources was a significant factor in determining degree of potential exposure. Another
important finding of this study was that the additive contribution for a single pollutant emitted from a
variety of sources resulted in health reference exceedences over a broad geographic area.
The urban county study concludes that for certain pollutants, the combined impact of multiple
sources may result in substantial exposure to many people. This finding suggests that the problem
may not be limited to large point sources, but that smaller point sources and area sources mat are
numerous in populated areas cannot be ignored. Similarly, exposure to chemical mixtures may result
in noncancer health impacts that might not be predicted if only impacts of individual pollutants are
considered.
A listing of HAPs of highest potential concern based on the urban county modeling study are
shown in Table 1-5 in Chapter 1 of this report. Two columns of pollutants are shown in Table 1-5,
based on long-term and short-term modelled exposures, based on data from Figure 3-7 and Table 3-
21, respectively. Those pollutants (not all of which are HAPs under Section 112) of potential concern
64
-------
Excerpted directly from the EPA Noncancer Screening Study
Figure 3-7. Results of Long-term (HEM) Modeling Analysis (Exposure from Individual Facility Emissions)
Number of Facilities
10
8
6
(13x)
5,200 people
(5.9x)
7,100 people
Emissions from
Facilities > HRL*
(30x)
(4.0x) 400,000 people
6,000 people
0
Arsenic Cadmium Copper Manganese
Benzene Chlorine 2-Furfural Mercury
Pollutant
*HRL=Hea!th Reference Level
Note: Factor in parenthesis denotes highest modeled estimate expressed as a multiple
of the highest reference level.
-------
Excerpted directly from the UFA Noncancer Screening study
Table 3-21. Urban County Study Results of Short Term (Screen) Analysis
Pollutant
Ethyl benzene
Formaldehyde
Heptane
Hexamethylenediamine
Hexane
Hexylene gtycol
Lead
Methanol
Methyl cellosolve
Methyl ethyl ketone
Methyl methacrylate
o-Dichlorobenzene
o-Xylene
Phenol
Phthalic anhydride
Styrene
Sulfur
Toluene
Vinyl chloride
Xylene
Exceedance Factor
forHRL*
2.0
210
14.0
2.2
1.7
1.0
4.9
1.7
2.6
85.0
1.5
1.1
4.2
1.8
1.700.0
170.0
1.1
5.7
22.0
50.0
Facilities
3
20
6
2
1
1
2
3
1
36
1
1
4
1
32
16
1
9
12
32
Exceedance Factor
forLOAEL*
2
4.9
1.7
Facilities
1
2
3
Uncertainty
Factor
1,000
100
100
1,000
100
100
1
100
1,000
100
100
1,000
1.000
1.000
1.000
1,000
1,000
1,000
1,000
1,000
*HRUHealth Reference Level, LOAEL=Lowest-Observed-Adverse-Effect Level
continl
-------
Excerpted directly from the EPA Noncancer Screening Study
Table 3-21. Urban County Study Results of Short Term (Screen) Analysis (continued)
Pollutant
1.1.1Trichloroethane
2-Furfural
a-Pinene
Aectaldehyde
Acrolein
Acrvlon'rtrile
Benzene
Butvl Cellosolve
Cadmium
Carbon sulfide
Cellosolve
Chlorine
Chlorobenzene
Chloroform
Chloroprene
Cvclohexanone
Cvclopentane
Dichlorotetrafluoroethane
Dimethvl Formamide
Ethyl acetate
Exceed ance Factor
for HRL*
5.5
2.2
35.0
1.2
97.0
8.4
120.0
17.0
13.0
450.0
1.8
440.0
2.4
26.0
110.0
1.4
36.0
820.0
24.0
180.0
Facilities
2
2
3
1
12
3
6
16
6
22
1
31
2
4
14
1
8
22
1
42
Uncertainty
1.000
1.000
1.000
100
100
100
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
*HRL=Hea!th Reference Level
-------
based on long-term exposures include: arsenic, benzene, cadmium, chlorine, copper, 2-furfural,
manganese, and mercury. As is suggested by the projected HRL exceedance levels and number of
persons exposed in Figure 3-7, benzene exceeds the respective HRL by the highest ratio (x30) and
has the maximum number of people exposed (400,000). Some of the pollutants of greatest potential
concern in the urban county study based on short-term exposures include: phthalic anhydride, carbon
disulfide, chlorine, formaldehyde, styrene, benzene, chloroprene, acrolein and methyl ethyl ketone.
All of the pollutants exceeding an HRL in this study because of projected short-term exposures are
shown in Table 3-21 and Table 1-5 in Chapter 1.
Because of the considerable uncertainty in the data sets used in this noncancer analysis,
and because of the difficulty in comparing noncancer endpoints, the pollutants are not ranked in
this study. In addition, it is unclear how to interpret the effects of a pollutant concentration
slightly above a specific health effects level.
Limitations: The sparseness of the available data represents the principal limitation of the
screening studies. Few data were available to aid in the prediction of ambient concentrations and the
derivation of health reference levels. Also, the contribution of mobile sources was omitted from the
modeling analysis and the contribution of smaller point and area sources needed to be improved.
Many of the general assumptions and limitations itemized in Section 2.5 apply to this study.
68
-------
REFERENCES
1, U.S. EPA, Office of Air Quality Planning and Standards. Cancer Risk from Outdoor
Exposure to Air Toxics. Vol. I and II. EPA-450/l-90-004a,b. Research Triangle Park, NC.
September 1990.
2 See, for example: A Legislative History of the Clean Air Act Amendments of 1990. volume
V, S.Prt. 103-38, 103rd congress, November 1993, pp. 8487-8532.
3. U.S. EPA, Office of Air Quality Planning and Standards. Analysis of Air Toxics Emissions.
Exposures. Cancer Risks and Controllability in Five Urban Areas. EPA-450/2-89-012a.
Research Triangle Park, NC. July 1989. (Five-city study)
4. U.S. EPA, Region III - Environmental Services Division and Office of Policy Analysis.
Kanawha Valley Toxics Screening Study. Final Report. Philadelphia, PA. July 1987.
(lEMP-Kanawha Valley study)
5. U.S. EPA, Office Policy, Planning, and Evaluation. Final Report of the Philadelphia
Integrated Environmental Management Project. December 1986. (lEMP-Philadelphia study)
6. U.S. EPA, Office Policy, Planning, and Evaluation. Santa Clara Valley Inteerated
Environmental Management Project. Stage Two Report. September 1987. (lEMP-Santa
Clara study)
7. U.S. EPA, Office of Policy, Planning, and Evaluation. Baltimore Integrated Environmental
Management Project: Phase II. Ambient Air Toxics Report. February 1992. (IEMP-
Baltimore, Phase II study) EPA-230-R-92-013
8. South Coast Air Quality Management District. The Multiple Air Toxics Exposure Study.
1987 Air Quality Management Plan Revision Working Paper No. 3. June 1987. (South Coast
study) Reprinted by U.S. EPA in November, 1988. EPA/4-88-013.
9. U.S. EPA, Region V. Estimation and Evaluation of Cancer Risks Attributed to Air Pollution
in Southeast Chicago. Chicago, IL. September 1989. (Southeast Chicago study)
10. U.S. EPA, Region V. Estimation and Evaluation of Cancer Risks Attributed to Air Pollution
in Southwest Chicago. Chicago, IL. April 1993. (Southwest Chicago study)
11. U.S. EPA, Region VI. A Risk Assessment Based Air Enforcement Strategy for Harris
County. Texas. Dallas, TX. July 1988. (Houston study)
12. U.E. EPA, Region V. The Transboundarv Air Toxics Study. Final Summary Report.
Chicago, IL. December 1990. (Detroit study)
13. U.S. EPA, Office of Air Quality Planning and Standards. Toxic Air Pollutants and Noncancer
Health risks: Screening Studies. Research Triangle Park, NC. September 1990. External
review draft. (Noncancer health study)
69
-------
14. Cote, Ila, Larry Cupitt and Beth Hassett, "Toxic Air Pollutants and Noncancer Health Risks,"
In Advances in Risk Analysis. Vol 7, 1988
15. Vorhees, Beth hassett, and Ila Cote, "Analysis of the Potential for Noncancer Health Risks
Associated With Exposure to Toxic Air Pollutants," Paper No. 89-91.1 presented at Air and
Waste Management Association annual meeting, Anaheim, California, June 25-30, 1989.
16. Texas Air Control Board. An Assessment of the Urban Air Toxics Problem in Texas.
Austin, TX. June 1989. (Texas study)
17. Minnesota Pollution Control Agency. Estimation and Evaluation of Cancer Risks from Air
Pollution in the Minneapolis/St. Paul Metropolitan Area. March 1992. (Twin city study)
18. Environmental Protection Agency. Interim Procedures for Estimating Risks Associated with
Exposures to Mixtures of Chlorinated Dibenzo-p-Dioxins and Dibenzofurans (CDDs and
CDFs). Risk Assessment Forum. Prepared by Bellin, J. S. and D. G. Barnes. Washington,
DC. March 1989.
19. National Academy of Sciences, National Research Council. Science and Judgement in Risk
Assessment. National Academy Press: Washington, DC. 1994.
70
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA -452/R-95-001
3. RECIPIENT'S ACCESSION NO.
TITLE AND SUBTITLE
Summary of Urban Air Toxics Risk Assessment Screening
Studies to Support the Urban Area Source Program
5. REPORT DATE
March 1995
6. PERFORMING ORGANIZATION CODE
OAQPS
AUTHOR(S)
Tom Lahre, U.S. Environmental Protection Agency
Emory Kong, Research Triangle Institute
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORGANIZATION NAME AND ADDRESS
Re-search Triangle Institute
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
2. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
5. SUPPLEMENTARY NOTES
6. ABSTRACT
A number of screening studies are compiled and compared that assess exposures .and
risks from air toxics. Generally, these studies were based on dispersion modeling of '
air toxics emissions data. Cancer is the principal endpoint, although some noncancer'
endpoints are evaluated in the studies. This report is intended to provide background
for development of EPA's national strategy to control urban area sources.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air toxics
Cancer incidence, air toxics
Hazardous air pollutants
Risk assessment
Screening studies, air toxics
Urban air toxics (urban soup)
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report/
70
20. SECURITY CLASS (Tillspage)
22. PRICE
EPA Form 2220-1 (Rov. 4-77) PREVIOUS EDITION is OBSOLETE
-------
INSTRUCTIONS
1. REPORT NUMBER
Insert (he LPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Sol subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat Hie primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g.. Jate ttf issue, dale of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doc. J. Robert Doe. etc.). List author's at'filiaiion if it differs from tlu- performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code. List no more than two levels of an organisational liircarchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers nuiy be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGtNCY CODE
Insert appropriate code.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with. Translation of. Presented at eonlereiuv <>f.
To be published in. Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less/ factual summary of the most significant information contained in the report. II I IK- report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authori/.ed terms that identify the major
concept of the research and are sufficiently specific and precise to be used us index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI HELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, urea of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary I ickl/(>ruur> assignments Ihul will lollow
the primary posting(s).
18. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." file any availability to
the public, with address and price.
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, it any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (Rev. 4-77) (Reverse)
------- |