United States

               Environmental Protection
               Agency
               Research and Development
c/EPA
 Atmospheric Research and Exposure
 Assessment Laboratory
 Research Triangle Park, NC 27711
 EPA/600/R-95/027
™———^™-

 March 1995
               ERA'S Urban Area Source
                    Research  Program
               - A Status Report on Preliminary Research -
                              Principal Authors

                                Larry T. Cupitt
              Atmospheric Research and Exposure Assessment Laboratory
                       Office of Research and Development
                                 X
                                 1la,L Cote
                       Health Effects Research Laboratory
                       Office of Research and Development

                                Joellen Lewtas
                       Health Effects Research Laboratory
                       Office of Research and Development

                               Thomas F. Lahre
                    Air Quality strategies and Standards Division
                    Office of Air Quality Planning  and standards

                                Julian W. Jones
                  Air and Energy Engineering Research Laboratory
                       Office of Research and Development
                       Office of Research and Development
                       U.S. Environmental Protection Agency
                              401 Ml Street, S.W.
                            Washington, D.c. 20460

-------
                                          Disclaimer

   The information in this document has been funded wholly by the United States Environmental
Protection Agency. It has been subjected to the Agency's peer and administrative review, and it has
been approved for publication as an EPA document. Mention of trade names or commercial products
does not constitute endorsement or recommendation for use.
                                              11

-------
                                  Contents
 Tables	,	  iv
 Figures	v
 Acronyms and Abbreviations   	  vi
 Executive Summary	vii
 Acknowledgements  	  xi

 Section 1. Introduction	1

 Section 2. Hazardous Air Pollutant Assessment   	3
    2.1 Overview	3
    2.2 Environmental Health Paradigm	3
       2.2.1 Exposure Assessment	6
          2.2.1.1 Emission Sources	6
          2.2.1.2 Environmental Concentrations  	9
          2.2.1.3 Human Exposures  	10
          2.2.1.4 Complicating Factors	13
       2.2.2 Effects Assessment   	16
          2.2.2.1 Internal Dose   	16
          2.2.2.2 Health Effects	17
          2.2.2.3 Complicating Factors	19

Section 3. Previous Assessments   	23

Section 4. Research Needs	29
   4.1 Research on Exposure Assessment	29
   4.2 Research on Effects Assessment	31

Section 5. Summary of Preliminary Findings	33

References  	37
Appendix	41
Glossary	55
                                        111

-------
                                    Tables
Table E-l. The HAPs with the most extensive available data needed for evaluation
    of the Environmental Health Paradigm	  ix
Table 2-1. Typical median ambient outdoor concentrations of some of the 189
    listed HAPs. Concentrations are in micrograms per cubic meter (^ig/m3) of air. 12
Table 2-2. Occurrence and biological test results indicating carcinogenicity of air-
    borne  chemicals, for the 2,827 chemicals that have been reported to exist in
    the air	   22
Table 5-1. The HAPs with the most extensive available data needed for a risk
    assessment	   36
Table A-l. Availability of data on the 189 listed HAPs	   45
Table A-2. Number of hazardous air pollutants that have been reported to produce
    health effects in humans or animals by inhalation exposure	   55
                                        IV

-------
                                    Figures
 Figure E-l. The components of the Environmental Health Paradigm	  viii
 Figure 2-1. The components of the Environmental Health Paradigm and their
    relationship to Exposure Assessment and Effects Assessment	   4
 Figure 2-2. Summary of the available data on emissions of HAPs from all source
    types. (Table A-l, Appendix A, categorizes the data for each of the  189
    HAPs.)  	   8
 Figure 2-3. Summary of available  data on ambient outdoor concentrations of
    HAPs. (Table A-l, Appendix A,  categorizes the available data for each of the
    189 HAPs.)	   10
 Figure 2-4. Summary of the number  of HAPs that have been measured in a
    variety of U.S. cities or towns	   11
 Figure 2-5. Effects of photochemical reactions on the mutagenicity of wood smoke
    and auto exhaust, two common pollutant sources in populated areas.  (Mutage-
    nicity was measured using two  different bacterial reversion assays.)	   15
 Figure 2-6. Evidence of carcinogenicity of the HAPs. (Table  A-l, Appendix A,
    categorizes the data for each of the 189  HAPs.)	   18
 Figure 2-7. Availability of validated Reference Concentrations (RfCs) for the 189
    listed HAPs. RfCs are available for more chemicals, but several are grouped
    under a single listed HAP. (See Table A-l  for data on each chemical.) ....   19
 Figure 3-1. Relative contribution of various hazardous air pollutants to the
    estimate of nationwide cancer cases (from Cancer Risk From Outdoor Expo-
    sure to Air Toxics)	   24
 Figure 3-2. Relative contribution by source  to the estimate of nationwide cancer
    cases per year caused by all sources, as  reported in Cancer Risk From Outdoor
    Exposure to Air Toxics	   25
Figure 3-3. Results of a screening  study to identify air pollutants with potential
    noncancer  health effects	   26
Figure 5-1. Summary of the available data on the 189 listed HAPs. (Table A-l,
    Appendix A, categorizes the data for each of the 189 HAPs.)  	   35

-------
                     Acronyms and Abbreviations






 BTX     Benzene, toluene, and xylene




 CAA     Clean Air Act




 EPA     Environmental Protection Agency




 FIRE     Factor Information Retrieval system




 HAPs     Hazardous Air Pollutants




 L&E     Locating and Estimating documents




 IARC     International Agency for Research on Cancer




 LOAEL   Lowest-Observed-Adverse-Effect Level




 MACT    Maximum Achievable Control Technology




 PCBs     Polychlorinated biphenyl compounds




 PIC       Products of Incomplete Combustion




 POM     Polycyclic Organic Matter




RfC       Reference Concentrations




TEAM    Total Exposure Assessment Methods




TSDFs    Waste Treatment Storage and Disposal Facilities




VOCs     Volatile Organic Compounds
                                      VI

-------
                                  Executive Summary
    The Clean Air Act (CAA) Amendments of
 1990 require the Environmental Protection Agen-
 cy (EPA) to develop an "Area Source Program"
 that includes both a Research Program and a
 National Strategy to "substantially reduce the
 public health risks posed by the release of hazard-
 ous air pollutants from area sources ...." The
 Research Program is to include three components:
 (a) characterization of the sources of hazardous
 air pollutants (HAPs), especially area sources,  (b)
 characterization of the concentrations of HAPs to
 which people are exposed, and (c) consideration
 of public health risks from the emitted and trans-
 formed HAPs.

    The Research Program is intended to support
 development of the National Strategy. The Na-
 tional Strategy must "identify  not less than 30
 hazardous air pollutants which, as the result of
 emissions from area sources, present the greatest
 threat to  public health...." The National Strategy
 must then propose a strategy to control the sourc-
 es of the identified pollutants.  The strategy must
 also reduce the incidence of cancer attributable to
 exposure to  HAPs by 75% or  more.

    This report deals with the Research Program
 and current research capability to characterize the
 Emission Sources, the Exposure Concentrations,
 and the Health Risks due to area source emissions
 of HAPs. These three areas are discussed in terms
 of the Environmental Health Paradigm. (See
 Figure E-l.) This paradigm provides a conceptual
framework to describe both the three aspects of
the Research Program and the process of risk
assessment - risk management under the National
Strategy.

   There are two primary activities hi the Envi-
ronmental Health Paradigm: exposure assessment
and effects assessment. Exposure Assessment
evaluates how likely people are to come into
contact with HAPs and determines how large their
exposure is likely to be. Effects Assessment iden-
tifies what health effects are likely to occur once
people are exposed to HAPs. In order to under-
stand environmental health issues, it is necessary
to have some knowledge about each component of
the paradigm.

   The current status of information needed for
each of the components in the Environmental
Health Paradigm for HAPs is discussed.  The
availability of data to assess the risks potentially
posed by each of the 189 HAPs listed in the
Clean Air Act was evaluated hi three broad  cate-
gories: (1) characterization of area sources,  (2)
characterization of exposure concentrations,  and
(3) characterization of probable health effects.
The health effects data were characterized for
both non-cancer effects and cancer. In general, a
few HAPs in each category had a great deal of
data, while many chemicals had little or no  data.

   Twenty HAPs were found to have  "Fair or
Better" data available in all three of the catego-
ries. (See Table E-l.) This list of chemicals does
not identify the 30 or more "worst" HAPs;  rath-
er, the list simply identifies those HAPs with
sufficient data to begin a risk assessment of either
                                               Vll

-------
       Environmental  Health  Paradigm
                EXPOSURE ASSESSMENT
   Emission
   Sources
   • Pollutant

   • Amount

   • Location
Environmental
Concentrations
  • Air

  • Water

  • Soil / Dust

  • Food
Human
Exposures
• Route
• Magnitude
• Duration
• Frequency



Internal
Dose
• Absorbed
Dose
• Target
Dose
• Biomarkers

>
Health
Effect(s)
• Cancer
• Noncancer
- Symptoms
- Damage or
Disease
                                                 EFFECTS ASSESSMENT
Figure E-l. The components of the Environmental Health Paradigm.
the cancer or noncancer effects due to exposure to
that chemical. Another 20 HAPs are rated "Fair
or Better" in two of the three required areas.
Targeted research on this second group of HAPs
could readily provide sufficient data to allow a
risk assessment to be initiated. The 40 HAPs with
the most complete available data are listed in the
Table. The remaining 149 HAPs lacked important
data in two or more of the categories. In addition
to the 189 listed HAPs, other chemicals, such as
those produced by atmospheric transformation,
may also be of concern.
                           As a consequence of these data limitations,
                        risk estimates for many of the chemicals known to
                        be present hi urban environments will be very
                        uncertain. Research to overcome or address these
                        data limitations will likely be both expensive  and
                        time-consuming. Data for selected chemicals,
                        however, appear sufficient to assess risks and to
                        develop control strategies as warranted.
                                       Vlll

-------
Table E-l. The HAPs with the most extensive available data needed for evaluation of the Environmen-
tal Health Paradigm.
 HAPs with data rated "Fair or Better" in
 the three areas:
    •  Source Emissions
    •  Ambient Concentrations
 and
    •  Health Effects (Cancer or Noncancer)
HAPs with data rated "Fan- or Better" in
two of the following three areas:
   •  Source Emissions
   •  Ambient Concentrations
and
   •  Health Effects (Cancer or Noncancer)
 Benzene
 1,3-Butadiene
 Carbon tetrachloride
 Chloroform
 Ethylene dibromide
 Ethylene dichloride
 Formaldehyde
 Methylene chloride
 Styrene
 Tetrachloroethylene
 Toluene
 Trichloroethylene
 Vinyl chloride
 Arsenic compounds
 Chromium compounds
 Lead compounds
 Manganese compounds
 Mercury compounds
 Nickel compounds
 Selenium compounds
Acetaldehyde
DDE (p,p'-dichlorodiphenyldichloro-
ethylene)
1,4-Dichlorobenzene
Ethylbenzene
Ethylene oxide
Hexachlorobenzene
Hexane
Methyl bromide
Methyl chloroform
Pentachlorophenol
Polychlorinated biphenyls
Propylene dichloride
2,3,7,8-Tetrachlorodibenzo-p-dioxin
2,4,6-Trichlorophenol
Vinylidene chloride
Xylenes (mixed isomers)
Antimony compounds
Beryllium compounds
Cadmium compounds
Polycyclic Organic Matter
                                             IX

-------

-------
                               Acknowledgements
    The authors would like to acknowledge and
 recognize the efforts of a number of other contri-
 butors to this report. Jeanette Wiltse (OHEA,
 ORD) and Robert Fegley (OSPRE,  ORD) played
 key roles hi designing, drafting, and reviewing
 this report. Cheryl Scott (OHEA, ORD) provided
 important data on the health effects  of the HAPs.
 Anne Pope (OAQPS) provided critical information
 about emission inventories and the FIRE system.
 The authors are grateful to the individuals listed
 below who provided reviews of the  report. We
 would especially like to recognize Blair Martin
 for his insightful comments and suggestions.

 Reviewers
 G.  Blair Martin, AEERL, ORD
 Hal Zenick, HERL, ORD
 Dale Pahl, AREAL, ORD
 John Vandenberg, HERL, ORD
 Susan Perlin, OHR, ORD
 David Klefffman, OHR, ORD
 David Guinnup, OAQPS
 Don Theiler, STAPPA/ALAPCO
 Benjamin Shaw, South Coast Air Quality Manage-
    ment District
Jeffrey Myers, Department of Natural Resources,
    State of Wisconsin
Naydene Maykut, Puget Sound Air Pollution
    Control Agency
                                           XI

-------
Xll

-------
                                          Section  1
                                        introduction
    The purpose of this report is to summarize
 what is currently known about exposures to and
 risks from hazardous air pollutants (HAPs) that
 are emitted by "area" sources. The Clean Air Act
 (CAA) Amendments of 1990 require the Environ-
 mental Protection Agency (EPA) to develop an
 "Area Source Program" that includes both a
 National Strategy and a research program. The
 law also requires EPA to report the results of its
 preliminary research efforts. This report describes
 those preliminary research findings on area source
 emissions.

    Section 112(k) of the CAA1 mandates that
 EPA conduct an area source research program
 "after consultation with state and local air pollu-
 tion control officials." The law specifies that the
 research program should contain at least three
 elements:  (1) "ambient monitoring for a broad
 range of hazardous air pollutants ... in a represen-
 tative number of urban locations;" (2) "analysis to
 characterize the sources" of hazardous air pollut-
 ants (HAPs), with a focus on area sources and
 then* public health risks; and (3)  "consideration of
 atmospheric transformation ... which can elevate
public health risks."

    The mandated research program is intended to
provide the scientific basis for development of a
comprehensive  National Strategy to control emis-
sions of HAPs from area sources. The National
Strategy must be published by November, 1995,
in a report to Congress. It must "identify not less
than 30" HAPs that  "present the greatest threat to
public health in the largest number of urban ar-
eas." The strategy is to be fully implemented by
the year 2000 and must provide guidelines for
controlling the area source emissions of the 30 or
more identified HAPs, while simultaneously
ensuring a reduction  of at least 75% in the "inci-
dence of cancer attributable to exposure to haz-
ardous air pollutants  emitted by stationary sources
..., considering control of emissions of hazardous
air pollutants from all stationary sources and
resulting from measures implemented ... under
[the CAA] or other laws."

    The area source National Strategy is a key
component of the Agency's overall approach to
reducing exposure to and risk from HAPs. It is
especially important because of the variety and
number of sources that might be controlled under
this strategy.

    Traditionally,  scientists and engineers have
associated "area sources" with small, but numer-
ous, sources that are  likely to be found in any
urban area — sources like gas stations, dry clean-
ers, auto repair shops, and even emissions from
cars and trucks. However, the definition of an
area source of HAPs in the CAA is different from
the traditional meaning of the term. The CAA de-
fines an "area source" as "any stationary source
of hazardous  air pollutants that is not a major
source." In the CAA, a  "major" source of HAPs
is  "any stationary source ... that emits or has the
potential to emit considering controls, in the
aggregate, 10 tons per year or more of any haz-

-------
 ardous air pollutant or 25 tons per year or more
 of any combination of hazardous air pollutants."3
 An "area source" of HAPs, as defined in the
 CAA, therefore, is  any stationary source of HAPs
 that emits less than 10 tons per year of any  single
 HAP and less than 25 tons per year of all of the
 HAPs emitted by that source.

     Clearly, the definition of an  "area source" of
 HAPs in the CAA is somewhat different from the
 traditional definition. Specifically, the definition
 in the legislation excludes motor vehicles and
 nonroad mobile sources (which are regulated else-
 where in the Act), while it does include small
 stationary sources, even though they may not be
 "numerous" in an urban area.

     The National Strategy must address area
 sources as they are defined hi the CAA, rather
 than the traditional definition. Throughout the
 remainder of this document, the term "area
 source" refers to the definition found in the  CAA.
 Other documents, some of which are cited in this
 report, however, may use the traditional defini-
 tion. Because the  term  "area source" may nave
 different meanings hi different documents (espe-
 cially  those that date from prior  to the CAA
 Amendments of 1990), readers must be careful to
 understand what is included as an area source
 when evaluating other sources of information.
a Also note that the CAA defines a "major" source differently
when dealing with volatile organic compounds (VOCs),
pollutants that help produce ozone pollution. Throughout this
document, the term "major source" refers to a major source of
HAPs.

-------
                                          Section 2
                       Hazardous Air Pollutant Assessment
 2.1 Overview

    Ambient air pollution can contribute to the
 occurrence and/or aggravation of disease in urban
 and/or industrialized areas. Diseases associated
 with air pollution include respiratory diseases
 (e.g.,  asthma, bronchitis, and emphysema) and
 cancer.2-3> 4 EPA has conducted a number of
 "screening"  studies to begin to  define  the contri-
 bution of HAPs to this problem in the U.S.b The
 "screening"  studies, which are discussed in Sec-
 tion 3, were intended to make broad comparisons
 of risks for program planning purposes.  Such
 studies typically attempted to define exposures
 and risks from as many pollutants and sources as
 possible, although most studies  included only 10
 or fewer of the HAPs listed in the CAA. Because
 many assumptions about emissions, exposures,
 and health effects were commonly made in these
 studies, the results are generally viewed, at best,
 as crude approximations of the comparative risks
 posed to individuals and populations. While the
 results, typically expressed  in terms of cancer
 risks or potential noncancer effects, are not
 viewed as representing absolute risks, they pro-
 vide the best available estimates of the potential
 magnitude of the broad air toxics problem. Con-
 gress clearly  considered the results of such
 screening studies to be relevant  when legislating
the Section 112(k) area source program,  as evi-


b Such studies have been conducted in Philadelphia, Baltimore,
Kanawha Valley (WV). Los Angeles, Chicago, Santa Clara
(CA), Baton Rouge, Phoenix, and a few other locations.
denced by the extensive citations from various
House and Senate Committee Reports containing
the legislative history of the Clean Air Act
Amendments.5
2.2 Environmental Health Paradigm

    In order to assess the risks of HAPs, and to
manage or control those risks, it is often helpful
to consider the interrelated processes of exposure
and effects assessment in a conceptual framework,
or paradigm. Figure 2-1 illustrates one such para-
digm that is especially useful  for describing what
is known about HAPs in urban air.6

    Evaluation of potential health risks from expo-
sure to environmental pollutants is composed of
two primary activities that make up the  Environ-
mental Health Paradigm: exposure assessment  and
effects assessment. Exposure Assessment evaluates
how likely people are to come into contact with
HAPs and determines how large their exposure is
likely to be. Effects Assessment  identifies what
health effects are likely to occur once people are
exposed to HAPs. In order to understand environ-
mental health issues,  it is necessary to have some
knowledge about each component of the paradigm
— from Emission Sources through Health Effects.

    Never will EPA have perfect and complete
data about all aspects of the paradigm, yet critical
decisions about the National Strategy must be

-------
       Environmental  Health  Paradigm
                EXPOSURE ASSESSMENT
   Emission
   Sources
   • Pollutant

   • Amount

   • Location
Environmental
Concentrations
• Air
• Water
• Soil / Dust
• Food

4
>

Human
Exposures
• Route
• Magnitude
• Duration
• Frequency


»

Internal
Dose
• Absorbed
Dose
• Target
Dose
*
• Biomarkers
                          Health
                          Effect(s)
                         » Cancer

                         » Noncancer
                          - Symptoms
                          - Damage or
                             Disease
                                                  EFFECTS ASSESSMENT
Figure 2-1. The components of the Environmental Health Paradigm and their relationship to Exposure
Assessment and Effects Assessment.
made. Often, assumptions about one or more
aspects of the Environmental Health Paradigm
must be made in order to fill in the data gaps. In
some situations, simplifying assumptions might
not significantly affect the risk assessment. For
other chemicals or locations, the need to make
such assumptions might introduce large uncer-
tainties into the assessment. The amount and
quality of information needed to evaluate properly
each component of the Environmental Health
Paradigm will vary from case to case and chemi-
cal to chemical.
   To assess exposure thoroughly, one must
characterize the Emission Sources, Environmental
Concentrations, and Human Exposure factors.
Knowledge of Emission Sources is needed to
determine where, how much, and when HAPs are
emitted. Critical information includes the types
and amounts of pollutants released and the loca-
tions of the sources. Once the HAPs are emitted
into the air, they are transported and transformed
until some of them come into contact with hu-
mans. Information about Environmental Concen-
trations is necessary to determine the pollution
levels to which people might be exposed. For a

-------
 comprehensive assessment, data are needed for all
 media through which exposure might occur,
 including air,  water, soil, or food. The Human
 Exposure factors consider how people and pollut-
 ants come into contact with each other. The goal
 of the human exposure factor is to define the
 route, magnitude, duration, and frequency of the
 contact between humans and HAPs.  Exposure is
 measured as the product of the pollutant concen-
 tration and the time during which people are ex-
 posed.

    Human exposures to HAPs can occur through
 a variety of routes, in addition to the air that
 people breathe. Total exposure assessments in-
 clude estimates for each route. HAPs can deposit
 out of the air to a  variety of surfaces, eventually
 polluting water, soil, food, and objects around us.
 Indirect exposures to HAPs can also occur from
 the food and water people consume, and from the
 objects that humans touch. Although such indirect
 exposures can be extremely important in some
 cases,0 this report will consider primarily expo-
 sures through the air people breathe.

    The  intent  of the final components of the
 Environmental Health Paradigm is to identify the
 health hazards associated with HAPs and to define
 the relationships between exposure, target dose
 (the dose to the affected organs or biological
 systems), and health in human populations. This
 is also known  as the exposure-response relation-
c Other routes of exposure may be very important in many
cases. The Great Waters Program was authorized under
Section 112(m) of the Clean Air Act because of deposition of
toxic air pollutants to lakes and other bodies of water with
subsequent entry into the food chain or drinking water and
human exposure by ingestion. Recent National Academy of
Science reports on lead discuss human exposure by ingestion
of lead-containing particles deposited on food, as well as child
ingestion of lead-containing dust. (See, for example, National
Academy of Sciences, Measuring Lead Exposure in Infants,
Children, and Other Sensitive Populations, National Academy
Press, Washington, DC, 1993.)
ship. The overlap between Exposure Assessment
and Effects Assessment, as shown in Figure 2-1,
reflects the interrelationship of these two assess-
ment activities.

   For a health effect to occur, HAPs in ambient
air first must actually get into the body. Internal
Dose defines how much of the HAPs that one
breathes (or ingests or contacts) actually gets into
the body (absorbed dose), and how much gets to
the specific organ(s) where they might cause
damage (target dose). Significant biologic events
resulting from this target dose can be used as
measures of internal dose (biomarkers). Absorbed
dose, target dose and resulting biomarkers are all
critical links between human exposure and conse-
quent health effects. Improving measures of these
links improves the estimates of risks posed by
HAPs.

   Health Effects are often categorized into can-
cer and noncancer health effects. Historically, one
basis for this categorization of health effects is the
dichotomous nature of cancer (that is, either you
have it or you don't) versus the wider variety of
symptoms, damage, or disease associated with
noncancer effects. For example, respiratory disor-
ders resulting from exposure to HAPs can range
from itching noses, coughing, shortness of breath,
decreased capacity to inhale or exhale, bronchitis,
increased asthma attacks, emphysema,  pulmonary
edema and death.  More than one effect, like those
listed, can often appear together, in varying de-
grees of severity.  Effects in different organs or
biological systems also can occur  simultaneously.
Consequently, Effects Assessment  must often
evaluate a complex set of health effects, with
different patterns of affected organ systems and
with widely different severity of effects. These
patterns are often  chemical-specific and change
with exposure concentrations, durations, frequen-
cy of exposure, and with characteristics unique to
the population that is exposed (for example, ge-
netic or gender or age-related characteristics).

-------
 Based primarily on laboratory animal studies and
 occupational observations, the health effects most
 commonly associated with HAPs exposures are
 cancer, developmental and reproductive disorders
 (for example, retarded development in children or
 birth defects), neurotoxicity, and short-term and
 long-term pulmonary disorders.7'8

    The components of the Environmental Health
 Paradigm also provide a reasonable way to sum-
 marize the current understanding of HAPs in
 urban air. The following discussions will focus on
 the Exposure Assessment and the Effects Assess-
 ment.

 2.2.1 Exposure Assessment

    The first component of the Environmental
 Health Paradigm is Exposure Assessment. In this
 section, we consider each of the components of
 Exposure Assessment:
 • Emission Sources
 • Environmental Concentrations
 • Human Exposure.

 2.2.1.1 Emission Sources

   Reliable data on emissions of HAPs from area
sources are limited. Most previous studies of
emissions in urban areas have focused primarily
on criteria pollutants or their precursors, such as
volatile organic compounds  (VOCs), paniculate
matter, sulfur oxides, and nitrogen oxides, not on
the 189 chemicals listed  as HAPs. Furthermore,
previous studies focused primarily on all types of
sources (major point sources, mobile sources, and
area sources), not just area sources. Emissions
(e.g., tons of pollutant per year) from area sourc-
es may have not been included in such studies,
and even if they were included, the data may not
allow a complete and accurate emissions invento-
ry to be assembled.
    Deficiencies in emissions data might involve
any of the various aspects of Emission Source
characterization, including describing the type of
pollutants, quantify ing how much of the HAP is
released, or locating the sources geographically.
Data available under the Toxic Release Inventory
are very useful in locating potential releases of
HAPs from many major sources, but similar data
are not available for the smaller area sources.
Nonetheless, some area sources of HAPs are well
defined, and a great deal of data  are available for
area sources like residential wood combustion,
dry cleaners, and publicly-owned treatment
works. Aside from such sources, however, emis-
sion inventories have traditionally focused mostly
on major sources of VOC emissions (some of
which are also HAPs) or on sources of criteria
pollutants (for example,  sulfur oxides, paniculate
matter, and nitrogen oxides). In many cases the
exact HAPs and the concentrations that are emit-
ted from small sources are not well known. In
many inventories, emissions from small area
sources  are not located or measured precisely, but
are estimated from indirect measures like the
number of people in an area, the number of cars,
and the  quantity of solvent sold.

   Efforts are underway to reduce the uncertain-
ties in emissions inventories for a number of
important HAPs.  EPA is continuing to develop
improved tools for use in developing HAP emis-
sion inventories. "Locating and Estimating"
("L&E") reports are available for more than 30
HAPs. These reports contain pollutant-specific
information on industrial processes, emission
factors (e.g., pounds of pollutant emitted per ton
of fuel burned), source test methods, and in the
recently updated reports, national inventories,
including emission estimates for point, area and
mobile sources. Thirteen "L&E" reports were
developed or upgraded in fiscal year 1993, and
seven additional updated reports are anticipated
for 1994. In addition, the Factor Information
REtrieval system (FIRE) contains evaluated emis-

-------
 sion factors for both criteria pollutants and HAPs.
 FIRE is updated periodically and now contains
 9700 rated emission factors, of which approxi-
 mately  4000 factors are available for 29 of the
 listed HAPs.

    Even when data on emissions are available,
 there are still uncertainties involved  in extrapolat-
 ing the data to other locations or to other opera-
 tional conditions. To ensure the development of
 reliable emission factors,  one must measure the
 emissions at a variety of sources in a specific
 category and must collect sufficient data on plant
 operations, processes, and conditions. Obtaining
 reliable emission factors is expensive even when
 the source is not difficult  to test and reliable
 measurement techniques are available. Under
 California's Assembly Bill 2588 program (the
 "Hot Spots" Program), many producers of HAPs
 are required  to conduct such tests at their own
 expense. EPA has used the data from California
 to extract more than 1500 HAP emission factors,
 and is implementing a project to obtain source test
 data from other state and  local agencies.

    A number of state or local air pollution con-
 trol agencies have voluntarily developed invento-
 ries9 of HAP emission sources over the last de-
 cade, despite the lack of a federal requirement for
 HAP emission inventories. Without specific guid-
 ance about such inventories, the state and local
 agencies have chosen to include different HAPs hi
 their inventories and to use a wide variety of
 methods to estimate emissions.  Consequently,
 there  is  often very little consistency between the
 available inventories.  Nonetheless, efforts are
underway both at the federal and state levels to
overcome some of the shortcomings  found in the
inventories and to reduce the inconsistencies. As
mentioned above, the California "Hot Spots" Pro-
gram has proven to be very productive in provid-
ing better emissions data.  In addition, the eight
member states of the Great Lakes Commission are
working together to develop a regional emissions
inventory for mobile, area, and point source
emissions of 49 HAPs. Additional data on source
emissions should become available as states im-
plement the permit programs as required by the
Clean Air Act Amendments of 1990.

    On a national scale, EPA has also supported
national HAP emission inventories for fourteen
HAPs (and related species) in 1993. These HAPs
included mercury, alkylated lead,  hexachloroben-
zene, POM, polychlorinated biphenyls (PCBs),
tetrachlorodibenzodioxin, tetrachlorodibenzofuran,
benzene, 1,3-butadiene, carbon tetrachloride,
tetrachloroethylene, trichloroethylene, methylene
chloride, and formaldehyde. These national inven-
tories include estimates for mobile, area and point
sources and are allocated to the county level.
Although such inventories do not precisely locate
all sources of HAPs, they can still provide valu-
able information for estimating urban emissions of
HAPs.

    Efforts to assemble emission inventories have
been identified for a variety of urban sources,
including area sources, for more than 60 HAPs,
but fewer than 20 of the HAPs appear with regu-
larity (that is, in 50% or  more of  studies) hi the
detailed inventories  compiled by state and local
agencies.10 Emissions of other HAPs can be
estimated on the basis of national  inventories, or
might be computed  from  available emission fac-
tors. Figure 2-2 illustrates the availability of emis-
sions data for the  189 listed HAPs. Forty-two
HAPs (seventeen HAPs that appear in 50% or
more of the state and  local inventories, together
with an additional twenty-five HAPs that appear
in the FIRE data base or  that are included in
national inventories) are categorized as "Fair or
Better." HAPs that  appear infrequently (less than
50% of the time) in detailed inventories are listed
as "Occasionally Found." There are little or no
emissions data for more than 120  HAPs.

-------
            Availability of  Source  Emissions Data
                       For the  189  Listed HAPs**
               Little or No Information
               66.7%
                                                       Fair or Better Data
                                                       22.2%
                      Occasionally Found
                      11.1%
   ** Based on frequency of inclusion in state or local inventories, data availability in the FIRE
     data base, or the availability of a national inventory.
Figure 2-2. Summary of the available data on emissions of HAPs from all source types. (Table A-l,
Appendix A, categorizes the data for each of the 189 HAPs.)
   Two major approaches can be used to identify
how much of urban pollution comes from the area
sources: dispersion modeling and source appor-
tionment.

   If the emissions from all sources are well
known, the contribution from area sources to
ambient concentrations of HAPs can be estimated.
The estimates for area sources may then be com-
pared with the contributions from all other types
of sources,  through dispersion modeling. Dis-
persion models describe how the emissions mix in
the atmosphere and are distributed throughout the
urban area. However, there are serious short-
comings in the current emission inventories for
urban areas with regard to area source emissions
of HAPs, as previously noted. These shortcom-
ings bring into question the reliability and accu-
racy of the dispersion modeling approach.

   The second approach, source apportionment,
uses ambient monitoring data to estimate how
much of the pollution came from each source.
This approach works best when each source (or
source category) contributes substantially to the
total pollution in a unique and distinctive way.
                                           8

-------
 Such is not the case, however, for many sources.
 For example, benzene, toluene, and xylene (often
 referred to jointly as BTX) are frequently the
 HAPs with the highest concentrations in urban
 air. It would be very useful to know what fraction
 of BTX in air was due to area sources.  Two
 recent apportionment studies11'12 found that
 85-95% of the BTX in urban air came from
 mobile sources. With such a large and dominant
 source, apportionment of the small remaining
 fraction of BTX from area sources  will prove
 very difficult to assign to specific area sources or
 source categories.

    While both dispersion modeling and source
 apportionment methods have their limitations,
 they can be used together to complement the
 strengths and weaknesses of each approach.

    It is  important to understand the impact of
 area sources on human exposure and risk, even if
 their emissions are small compared to the total
 quantities emitted by all sources. Exposure and
 risk is not necessarily proportional to the magni-
 tude of the emissions.  This is especially true if
 the area  sources (and other sources, like indoor
 sources or sources from personal habits) are much
 more closely linked  with human activities, be-
 cause such sources could still dominate the result-
 ing risk since they could contribute disproportion-
 ally to human exposure.

 2.2.1.2  Environmental Concentrations

    The availability of data on ambient outdoor
 concentrations of the 189 HAPs is highly uneven
 (Figure 2-3). The ambient outdoor concentrations
 result from emissions from all types of sources,
 including point, area, and mobile sources. The
figure plots the total  number of HAPs that have
been measured versus the number of times they
have been measured  in outdoor air in populated
areas.13 There are little or no ambient measure-
ment data (fewer than 100 observations) for near-
 ly two-thirds (112) of the HAPs, while a few
 chemicals — notably benzene, toluene, and the
 three xylene isomers — have each been measured
 many thousands of times. For 71 of the 189
 HAPs (38%). there are no ambient measurements
 at all.  The 43 HAPs with "Fair or Better" data all
 have more than 1000 observations. (An "observa-
 tion" is one or more measurements at the same
 location within a 24 hour period.) Clearly, there
 is little or no information about a large number of
 HAPs, but a great deal of information about a
 smaller number of HAPs.

    The same conclusion (very little data for most
 HAPs; considerable data for some HAPs) also
 extends to the number of cities for which ambient
 outdoor concentration data are available. Nearly
 two-thirds of the listed HAPs have been measured
 at fewer than 5 cities or towns, while BTX data
 are available for more than one hundred cities.
 Figure 2-4 illustrates just how few HAPs have
 been measured at an adequate number of cities.
 Additionally, the data are often available only for
 short periods of time — a few days or weeks —
 while special studies were underway.  Long-term
 collection of data on HAPs is available for only a
 very few cities.

    When two-thirds of the designated HAPs have
 been measured only a few times and at only a few
 cities, the "representativeness" of the ambient
 outdoor data becomes an important issue. Even
 the data that are available are of inconsistent
 quality and duration. When large data gaps  exist,
 either in space or time, it is very difficult to
 estimate human exposures and potential health
 effects reliably,  or to identify trends in order to
 characterize the impacts of regulatory programs.

    Table 2-1 lists typical outdoor concentrations
 of a few HAPs13 that are among the best-studied
 in terms of health effects. As discussed earlier,
the actual ambient outdoor measurements are
often variable; nevertheless, these concentrations

-------
                Availability of Ambient  Outdoor
      Concentration Data For the  189 Listed HAPs
                                                No Data
                                                37.6%
                  Little Data
                  21.7%
               Occasionally Observed
               18.0%
                                                      Fair or Better Data
                                                      22.8%
  The categories are based on the number of reported observations, as described in the text
  HAPs with No data have 0 observations; HAPs with "Little Data" have <100 observations. The
  "occasionally observed" HAPs have 100-1000 observations, and HAPs with "Fair or Better Data"
  have been observed more than 1000 times.
Figure 2-3. Summary of available data on ambient outdoor concentrations of HAPs. (Table A-l,
Appendix A, categorizes the available data for each of the 189 HAPs.)
are typical of the reported data. Median concen-
trations, in micrograms per cubic meter (/ig/m3),
are listed in the table. The median is the middle
of the distribution of observed concentrations: half
of the time, the measured concentrations were
larger than those listed, and half of die time, die
concentrations were reported to be smaller. "Av-
erage" concentrations are not given since an
arithmetic average can sometimes be misleading,
especially if there are a few very large concentra-
tion measurements or if there are many observa-
tions with concentrations too small to measure
accurately. There are major differences between
the number of times and number of locations in
which the various chemicals have been measured.

2.2.1.3 Human Exposures

   To develop the National Strategy to minimize
adverse health effects from area source emissions
of HAPs, it is necessary to consider the actual hu-
man exposure to die HAPs, not merely the ambi-
ent concentrations. The following text describes
what is currently known about the distribution of
HAPs across urban areas and about the impact of
outdoor air on indoor air and personal  exposures.
                                          10

-------
      Cities  and Towns with Ambient Outdoor  Data
                        For the 189 Listed HAPs
                        Fewer than 5 Cities
                        60.3%
                                                          More than 50 Cities
                                                          12.7%
                                5 to 50 Cities
                                27.0%
 Figure 2-4. Summary of the number of HAPs that have been measured in a variety of U.S. cities or
 towns.
Distribution

   To estimate human exposure to HAPs one
must know how widespread are the concentrations
of urban air pollutants. If area sources are uni-
formly and widely distributed across an urban
area, one would expect the concentrations of the
emissions to be relatively consistent across the
community, although this is not always true. Only
a few studies have included simultaneous mea-
surements of pollutants at different sites across an
urban area. One recent study, focusing on the
particle-bound pollutants from residential wood
burning and from automobiles, found the concen-
tration of fine particles from these sources to be
relatively consistent across an urban area.14 Oth-
er studies measuring gaseous HAPs and other -
vapor-phase pollutants have found that a few gas-
eous pollutants appear to have relatively constant
concentrations15 across distances as large as  10
km, implying that the  sources of those pollutants
are widely and uniformly distributed throughout
the community.11 However, in the same study, a
larger group of gaseous pollutants were reason-
                                          11

-------
 Table 2-1. Typical median ambient outdoor concentrations of some of the 189 listed HAPS.
 concentrations are in micrograms per cubic meter fc/g/m ) of air.
Chemical
Abstract
System
Number
71-43-2
108-88-8
95-47-6
108-38-3
106-42-3
75-09-2
67-66-3
106-99-0
71-55-6
56-23-5
50-00-0
Not applicable
57-74-9
Chemical Name
Benzene
Toluene
xylenes (o-, m-, p- Isomers)
Methyiene chloride
Chloroform
1,3-Butadiene
Methyl chloroform
Carbon Tetrachlorlde
Formaldehyde
Chromium compounds
Chlordane
Median
Concentration
fjg/ml
5
9
2 to 4
0.5
0.2
0.4
2
0.8
3
0.003
0.02
Number
of
Observations'
>8600
>6500
>5700
> 3400
> 4900
> 1900
> 4900
> 6300
> 2500
> 1800
> 345
Number of
Cities
with
Ambient
Data
172
159
> 130
86
135
66
155
149
75
> 192
> 8
 An observation is one or more measurements taken within the same 24-hour day.
ably constant only across distances of about 1 km,
while still other pollutants were even more vari-
able.

Impact of Outdoor Air

   People typically spend more than 80% of their
time indoors,16 so any analysis of the health ef-
fects from exposure to area sources must assess
the penetration of the area-source pollutants from
outdoors to indoors. Many of the volatile HAPs
are stable chemicals that do not react o^iickly with
other chemicals in the environment. Such stable
gaseous pollutants can easily penetrate indoors
with little or no loss of concentration. The  instan-
taneous indoor and outdoor concentrations can be
different, however, due to delays caused by the
rate at which outdoor air enters the building —
the air exchange rate.d
 For time periods  longer than a few hours, the
average indoor concentration of stable gaseous
pollutants generated by outdoor sources (including
area sources) is identical to the outdoor concentra-
tion adjacent to the house (for example, "on the
front porch.")17

    Some HAPs (for example, most POMs) are
not volatile vapors;  instead,  these HAPs are at-
tached to small particles in the air that people
breathe.  Non-volatile HAPs  that are emitted by
chemical or combustion processes are often bound
d Conversely, reactive pollutants, for example ozone, are
readily destroyed as they penetrate indoors resulting in indoor
concentrations that are generally less than outdoor concentra-
tions. Few of titie listed HAPs are expected to be so reactive.
                                                 12

-------
 to "fine" particles (less than 2.5 micrometers in
 diameter). Such particles are only partially re-
 moved as the air penetrates indoors. The number
 of particles that successfully penetrate indoors is
 roughly proportional to the air exchange rate.
 Thus, the more air brought indoors, the more the
 concentration of particles in the indoor air is like
 the concentration  of particles outdoors. For many
 buildings, the air  exchange rate is large enough to
 permit about 50%-90%  of the outdoor fine parti-
 cles to penetrate indoors successfully.  Non-vola-
 tile HAPs found on particles that are generated by
 mechanical processes (like dust kicked-up by
 automotive traffic, wind-blown dust, and con-
 struction projects) are usually bound to larger
 particles that are much less likely to penetrate in-
 doors.17

    Finally,  indoor sources, workplace sources,
 and personal activities can provide additional
 exposures to HAPs, beyond those due to the out-
 door sources. The outdoor sources provide a
 baseline of exposure to HAPs, on top of which
 indoor sources, workplace sources, and personal
 activities add additional exposures. If such indoor,
 workplace, or personal sources are large, they can
 dominate the total exposure calculation for those
 exposed individuals.  These sources must be taken
 into  account when determining the total human
 exposure to HAPs.

 2.2.1.4 Complicating Factors

   There are a number of factors that make
 Exposure Assessment a difficult and complex
 task. Two factors  that make identifying and char-
 acterizing the urban area sources of HAPs diffi-
 cult are: the  complexity of urban air pollution,
 and uncertainty in defining area sources.

   Urban air is a  complex mixture of thousands
of chemicals. These chemicals come from a wide
variety of sources, including major point sources,
area sources, mobile sources, and natural sources.
Examples of natural sources of HAPs include
forest fires, plant decay, and weathering of miner-
als containing heavy metals. The objective of the
urban area source research program is to charac-
terize the exposures and health risks due to area
sources in support of the mandated National
Strategy. But once the pollutants from the area
sources have mixed with those from major point
sources, mobile sources, and natural sources, it is
extremely difficult to identify how much of a
specific pollutant came from just the area sources.

    Even the definition of an area source under
Section 112 adds a  complicating factor.  For pur-
poses of the HAP National Strategy, area sources
also include point sources that do not meet the
requirements to be classified as major sources.
These "non-major point" sources have not tradi-
tionally been considered as area sources, and
were not previously included in efforts to charac-
terize area sources. "Major" sources are defined
as part of the Maximum Achievable Control
Technology (MACT) standard setting process
under Section 112(d): sources that do not meet the
requirements for MACT standards are by default
"non-major point" sources, or area sources.
(Some source categories that include individual
sources that are likely not to meet the definition
of a major source are:  bulk liquid (e.g., gasoline)
terminals, electric arc furnaces/stainless  steel
mini-mills, wood furniture manufacturing, second-
ary lead smelters, etc.) The final MACT stan-
dards are not scheduled for promulgation until
November 2000. Additional area sources might be
added for consideration, long after the National
Strategy has had to go into effect.

    Other factors make characterization of ambient
outdoor concentrations of HAPs a difficult under-
taking. For example, measurement methods are
not available for many HAPs, and natural reac-
tions in the atmosphere can either destroy or
produce HAPs.
                                                13

-------
 Measurement Methods

    One reason for the lack of data on both emis-
 sions and environmental concentrations of many
 of the HAPs is that there are often no reliable
 methods to collect and measure these chemicals.
 Measurements at the source and in ambient air are
 often made under distinctive conditions that make
 such measurements difficult. For example, source
 measurements often have high concentrations of
 contaminants and harsh conditions that make
 sampling and analysis difficult: ambient samples
 contain very small amounts of the species of
 interest and must be concentrated to be detected
 reliably. Validated source sampling methods exist
 for only 87  of the HAPs. In ambient air,  there is
 one group of HAPs where there is a particularly
 noteworthy lack of data. These compounds, nitro-
 genated or oxygenated organics, are often referred
 to as "polar" organics, and they comprise 89 of
 the 189 HAPs.  Only about one-third of these
 polar organics have actually been measured in
 ambient ah-.

 Atmospheric Transformation

   Another  difficulty with evaluating HAPs in
urban air is atmospheric transformation. Natural
atmospheric  events cause chemical reactions that
can both destroy and create HAPs.  These trans-
formation processes will eventually break down
and remove some of the HAPs from the ah".
Conversely,  transformation processes might con-
vert non-hazardous pollutants into dangerous
products  (or even transform HAPs into products
that are more hazardous than the original  HAPs.)
The HAPs formaldehyde, acetaldehyde, acetone,
and acrolein, for example, are all produced hi
significant quantities in urban air18 by the atmo-
spheric transformation of many organic com-
pounds — including many compounds not on the
list of HAPs. In other words, transformation pro-
cesses can produce a HAP even when one was not
emitted. This is similar to the situation with
ground-level ozone, which is produced primarily
through transformation of other pollutants, even
though it is not directly emitted.

   Many of the most important of these atmo-
spheric transformation processes involve sunlight.
Sunlight, shining on polluted urban air, sets  into
motion a complex series of chemical reactions
that convert the directly emitted pollutants into an
even more complex "soup." It is not possible to
identify all of the chemicals in the resulting prod-
uct mixture, but studies over the last decade
suggest that the sunlight-transformed mixture
might be even more hazardous than the originally
emitted pollutants. As an indicator of this poten-
tial for increased hazard, the bacterial mutagenici-
ty — the ability to cause changes hi the genetic
material of bacteria — of the transformed mixture
is often much greater  than that of the original
pollutants. This increase in mutagenicity is espe-
cially true for the gaseous products, which are
likely to be the partially-oxygenated or -nitrogen-
ated transformation products of the emitted chemi-
cals. Figure 2-5 shows the dramatic increases in
bacterial mutagenicity brought about by sunlight
in two complex pollutant mixtures that are often
found hi urban air, namely wood smoke and auto-
mobile exhaust.19'20

   The data hi Figure 2-5 are from smog cham-
ber simulations of atmospheric reactions, but at
concentrations higher  than those normally found
in the environment: such simulations are neces-
sary, since the mutagenicity tests are not suffi-
ciently sensitive to measure  such changes in  actual
urban air. Indirect evidence, however, suggests
such transformation effects do occur in ambient
outdoor urban air. A variety of simulations by re-
searchers around the world, involving many of
the pollutants commonly found in urban air, have
demonstrated several important facts about the
mutagenic products of atmospheric transforma-
tion:
                                                14

-------
    o
    5
    UJ
    o
    o
    o
               Effect of Atmospheric Transformation
           Increases in Bacterial Mutagenicity During Chamber Studies
b     20000
    o
    LU
    <3
                                              Legend
                                           After Transformation
                                           Before Transformation
                  Gas Phase  Particulate Phase
                      Wood Smoke
                                                   Gas Phase  Particulate Phase
                                                   Auto Exhaust
Figure 2-5. Effects of photochemical reactions on the mutagenicity of wood smoke and auto exhaust,
two common pollutant sources in populated areas. (Mutagenicity was measured using two different
bacterial reversion assays.)
•  Sunlight transforms many, but not all, urban
   pollutants into both gaseous and particle-bound
   mutagenic' products.

•  The gaseous mutagenic transformation prod-
   ucts are persistent: in the laboratory simula-
   tions, they are stable in the air for hours after
   they are produced. If they are produced and
   are stable under ambient conditions, then
   exposures can occur over large areas and for
   long times.
                                         •  About 90% (by mass) of organic chemicals in
                                            urban air are gaseous, with only about 10%
                                            bound to particles. In the laboratory simula-
                                            tions, the total mutagenicity of the gaseous
                                            transformation products in the air greatly
                                            exceeded the total mutagenicity of the particle-
                                            bound products in the same volume of air.
                                            The relative risk from gaseous mutagens
                                            versus particle-bound mutagens is unknown.

                                            These data on mutagenicity taken together
                                         cause concern about the potential impact of atmo-
                                           15

-------
 spheric transformation on cancer risks in urban
 areas. If transformation of non-hazardous air
 pollutants can cause a substantial cancer risk hi
 urban areas, it will make it difficult to develop a
 National Strategy that can reduce cancer risks by
 75%, as required by law.

 Exposure Variabilities

    One factor that complicates efforts to estimate
 human exposure is the fact that people and air
 pollutants move around throughout the day. What
 people do, and where they are, and when they are
 at a specific location all affect their exposure.
 Available exposure or concentration data often do
 not describe well the extremes in exposure, either
 very  large or very small exposures. People who
 live very close to a source (for example, in an
 apartment above a dry cleaning business, or near
 industrial or gasoline-handling facilities) can be
 exposed to abnormally high concentrations of
 specific HAPs. In addition, both people  and air
 pollutants move about during the day. As people
 move in and out of polluted areas,  their  exposures
 can change. Time-activity patterns  are descrip-
 tions of: 1) where people are  throughout the day,
 2) how long they remain in each location, and 3)
 what activities they are doing that can influence
 exposure (for example, jogging in a park will
 cause  a person to inhale more air and more pol-
 lutants than will reading a book on a bench in the
 same park).  Clearly, where a  person is during the
day and what be or she is doing can significantly
affect that person's exposure to HAPs. Only with
 information on the time-activity patterns of the
population relative to the sources of HAPs is it
possible to characterize accurately the exposures
 of people at the high end of the range of expo-
 sures — the very people who  are most likely to be
at risk. Some studies, like the Total Exposure
Assessment Methodology (TEAM)  studies21-22
 or the planned National Human Exposure Assess-
ment Survey  (NHEXAS),23 have a  statistical
 approach that is designed to measure a wide range
of exposures. Such studies are extending the
understanding of the range of potential human
exposures, but such statistically based studies are
very expensive to conduct and difficult to analyze.

2.2.2 Effects Assessment

   The second major aspect of the Environmental
Health Paradigm is Effects Assessment. Effects
Assessment is concerned with what happens to
human health once someone is exposed to HAPs.
There are three components of Effects Assess-
ment: Human Exposure, Internal Dose, and
Health Effects. Since Human Exposure is also a
part of Exposure Assessment and was described
earlier, the following describes the remaining two
components of Effects Assessment:
• Internal Dose
• Health Effect(s).

2.2.2.1 internal Dose

   The term  "Internal Dose" is often used to
convey a variety of concepts. In the current con-
text it means  the estimation of the amount of HAP
that enters the body and reaches an organ or
system where it might cause damage to human
health. Ambient air concentrations of HAPs have
often been used as surrogates for Internal Dose.
However, this practice can result in either over-
or tinder-estimations of  risk. Ambient concentra-
tions are not always  reliable indicators  of internal
dose because  biological  and biochemical process-
es, such as absorption into the body, distribution
in the body, metabolism, and excretion, all affect
how much of the HAP concentration in the air
actually reaches the organs or physiological sys-
tems where the pollutants might cause damage.
For particle-bound HAPs, even the physical char-
acteristics of the pollution may be important.
Particle size and the  nature of the particles on
which the HAPs are carried may strongly influ-
ence the location in the  body where the HAPs are
deposited, the mechanism by which adverse ef-
                                                16

-------
 fects may occur, the distribution of the pollutant
 within the body, and the internal persistence of
 the pollutants. It is important, therefore, to esti-
 mate Internal Dose as precisely as possible. The
 more accurate this estimation, the more accurate
 will be the assessment of potential HAP risks.

    The use of Internal Dose is particularly valu-
 able  when human risk estimates are derived from
 animal laboratory experiments or occupational
 studies. (HAP risk assessments are almost always
 derived from these types of data [see discussion
 of extrapolation of health effects data in the dis-
 cussion of complicating factors that follows]).
 New techniques are now being developed that
 allow for better estimates of Internal Dose. Some
 of these techniques are: measurements of biologi-
 cal and biochemical processes (pharmacokinetics);
 use of alternative and more relevant surrogates
 (biomarkers)  of Internal Dose; and actual mea-
 surement of the HAPs at the affected tissue (mo-
 lecular dosimetry). Scientific groups such as the
 National  Academy of Sciences and EPA's Science
 Advisory Board have encouraged the use of im-
 proved estimates of Internal Dose hi risk assess-
 ments. Unfortunately, reliable information on
 Internal Dose is currently available for only a few
 HAPs, and development of such information is
 currently  expensive, slow, and laborious. Through
 experience with available methods, and through
 research to  improve methodology, the costs to
 obtain better estimates of Internal Dose will,
 undoubtedly, decline over tune, and unproved
 estimates  will become more and more available.
           •
 2.2.2.2 Health Effects

    There are  some toxicity data available for
each of the  189 HAPs. In almost no case,  howev-
er,  are data available on all of the most important
health effects: cancer, developmental and repro-
ductive disorders (birth defects), neurotoxicity,
and acute (short-term) and chronic (long-term)
pulmonary effects. Moreover, the quality of the
available studies is highly variable. Some studies
are barely adequate, others are excellent. Another
problem is the lack of data on toxicity associated
with exposure by inhalation. Much of data on
health effects comes from tests involving only
ingestion of the HAP  (commonly called oral
exposure).  However, it is known that differences
hi the route of exposure can produce major differ-
ences hi the character and extent of toxicity.
Relying on only ingestion data alone generally
results hi large uncertainties hi the prediction of
health effects.

Cancer

    A serious possible health effect of HAPs is
their potential to cause cancer. More than 100 of
the 189 HAPs have sufficient data to assess their
ability to cause cancer qualitatively:24 even for
these chemicals, however, a quantitative estimate
of the dose-response relationship (potency) is not
always possible. Chemicals are classified based on
a variety of factors such as the quality of the
studies, the number of studies, and the species
reported to have chemically induced cancer.  Both
human and animal data are considered. Eighty-
three of the listed HAPs are considered to be
"probable" or known  human carcinogens. An-
other 25 HAPs are considered "possible" human
carcinogens. (See the classification definitions in
the glossary. N.B., EPA is revising its guidelines
for carcinogen risk assessment and the definitions
are expected to change.) Twenty-two of the HAPs
lack sufficient data for a classification, while the
remaining 59 of the HAPs have not been evaluat-
ed for carcinogenicity.e The carcinogenicity data
are illustrated hi Figure 2-6.

Noncancer Effects

    For  noncancer health effects, a Reference
Concentration (RfC) is used to estimate an expo-
sure concentration that is not harmful. An RfC is
an estimate, based on a single critical effect, of
                                                17

-------
                 Availability of Carcinogenicity Data
                          For the 189 Listed HAPs
             Unknown Carcinogenicity
             11.6%
            Possible Carcinogen
            13.2%
         Unevaluated HAPs
         31.2%
                                                   Probable or Known Carcinogen
                                                   43.9%

  (ARC carcinogens and Class A & B carcinogens are listed as "Probable or Known." Class C
  chemicals are listed as "Possible." Class D chemicals are classified as "Unknown." The remaining
  chemicals have not been evaluated.
Figure 2-6. Evidence of Carcinogenicity of the HAPs. (Table A-l, Appendix A, categorizes the data
for each of the 189 HAPs.)
the concentration (with uncertainty spanning a
factor of 10) that could be inhaled for a lifetime
c Some of the listed HAPs are actually groups of compounds,
and data may exist on several different chemicals within a
single HAP definition. For example, nickel compounds are
generally considered to be "possible" carcinogens, but nickel
subsulfide and nickel refinery dust are "known" human carcin-
ogens. In such case, the HAP (nickel compounds) was classi-
fied at the higher risk level ("Probable or Known") for tabula-
tion in this report. Similarly, Polycyclic Organic Matter
(POM) was classified as a "probable" carcinogen on the basis
of some specific compounds (for example, benzo(a)pyrene)
that often occur in POM.
with no adverse health effects. Only 40 of the 189
HAPs have sufficient data to support estimation of
an RfC. Confidence levels for an RfC vary from
unknown to low to high as illustrated in Figure
2-7.  Only five of the RfCs have "high" confi-
dence.  Most of the 149 HAPs without a validated
RfC  have not been studied for chronic inhalation
effects at all.

   Some of the chemicals on the list of 189
HAPs are also of concern to EPA because of their
potential  to cause serious, immediate health ef-
fects if people are exposed to very large concen-
                                              18

-------
       Availability of Noncancer Effects Data
                   For the  189 Listed  HAPs
            No validated RfC
            78.8%
                                                      Moderate or High Confidence in RfC
                                                      14.3%
                                              Low Confidence in RfC
                                              6.9%

   • Based on the level of confidence placed on the calculated Reference Concentration (RfC), the
    concentration that could safely be inhaled for a lifetime with no harmful noncancer health effect

Figure 2-7. Availability of validated Reference Concentrations (RfCs) for the 189 listed HAPs. RfCs
are available for more chemicals, but several are grouped under a single listed HAP. (See Table A-l
for data on each chemical.)
trations. Exposures to large concentrations might
occur, for example, after an industrial accident
that releases large quantities of a chemical. An
accepted method for describing the relationship
between dose of pollutants  and the biological
effects (the dose-response)  due to large, short-
term exposures to these chemicals is currently
under development by EPA. Data on short-term
(acute) effects are critical to EPA's Accidental
Release Program which is also mandated in the
Clean Air Act.
2.2.2.3 Complicating Factors

   Several factors make Effects Assessment —
evaluation of health effects from exposures to
HAPs in urban air — a very difficult task. Three
complicating factors are discussed below: (1)
extrapolation of health effects data, (2) exposure
to complex mixtures of environmental pollutants,
and (3) the fact that chemicals other than those on
the list of 189 HAPs can pose hazards to human
health.
                                           19

-------
 Extrapolation

    Most of the HAPs data available must be
 interpreted in some manner in order to assess
 public health risks. The most common types of
 interpretations involve the following: 1) using
 animal effects  data to predict effects that might
 occur in humans; 2) using effects data collected at
 relatively high exposure concentrations to predict
 effects that might occur at lower exposure concen-
 trations; and 3) using effects data collected with
 certain exposure durations and patterns to predict
 the effects that might occur with different expo-
 sure durations  and patterns. These interpretations
 are often called animal-to-human, high-to-low
 dose, and across-exposure-scenario extrapolations.
 These types of extrapolations are often difficult to
 perform with a great degree  of certainty. Limited
 data on which  to base extrapolations increase the
 uncertainty.

    Biological or biochemical processes might
 differ between laboratory animals and humans.
 Consequently,  responses to the same ambient
 exposures can also differ. Similarly, biological
 and biochemical processes in healthy adult male
 workers can differ from important segments of
 the general population,  such  as children and the
 elderly. Also, exposure concentrations in animal
 experiments and occupational studies are likely to
 be higher than  environmental exposures.  Exposure
 durations and patterns are also often different.
 These disparities can differentially affect biolog-
 ical and biochemical processes, and consequently,
Internal Dose. With careful study and  estimation
 of Internal Dose, many of these differences  can
be understood and quantified.

   Many of the uncertainties in risk assessment
 are unavoidable, given the current state of knowl-
 edge and the  need to assess public health risks
 from HAPs. Particular types of scientific informa-
 tion, however,  can improve analyses and reduce
 some uncertainties. In particular, reduction in
uncertainties can occur via better estimation of
dose to the affected organ (through such methods
as evaluation of pharmacokinetics, biomarkers,
and molecular dosimetry), and understanding  what
causes HAPs to have a toxic effect (the mecha-
nisms of action). The size of the effort that will
be required to gather these types of data, for even
just the most important HAPs, is substantial.

Complex Mixtures

   Complex mixtures confound the evaluation of
the HAP problem in urban air. Urban air is a
mixture of many pollutants, and little is known
about the effects of exposure to mixtures of chem-
icals. Usually, effects assessments deal with only
one chemical at a time. Sometimes, however, the
effects of simple mixtures are assessed by adding
together the anticipated effects from exposure to
each individual compound. This additivity ap-
proach is normally only used when the anticipated
effects are  similar for the various chemicals in the
mixture. When dealing with complex mixtures
(like those  found hi urban air)  and with many
different potential health effects, scientists are
reluctant simply to add together all of the antici-
pated individual health effects. They are reluctant
because the interactions of mixtures on health are
not well understood. Because of the complexity of
the interactions, the total  effect of the mixture
might be very different than the simple sum of the
individual effects. Additional research is currently
being conducted to develop methods that will
allow assessment of the effects of exposure to
complex mixtures.

Chemicals  Not on the List of 189 HAPs

   Another important uncertainty in evaluating
urban air is that chemicals, other than the 189
listed HAPS, might be shown to be more impor-
tant air pollutants in the future. Thousands of
individual chemicals, representing almost every
known chemical class, are expected to be present
                                                20

-------
 in urban air. There is recent evidence from stud-
 ies of complex mixtures of urban air particles and
 gases that the major contributors to the mutage-
 nicity of urban air are chemicals that have not yet
 been identified. These as yet unidentified chemi-
 cals might be produced by atmospheric transfor-
 mation of organic pollutants emitted by a variety
 of sources. Specifically, new bioassay-directed
 chemical identification techniques have identified
 polar organic chemicals  (for example,  hydroxyla-
 ted- and nitrated-aromatic hydrocarbons) in urban
 air that appear to arise from atmospheric transfor-
consider the wide range of toxic effects needed
for a full evaluation of the potential hazard. The
development of data on this broad range of sub-
stances is almost certainly not warranted. Further
analysis is needed to target specific chemicals for
further evaluation.
 mation
         25
     Table 2-2 categorizes the almost 3000 chemi-
 cals that have been detected in ambient air: it also
 notes the number of chemicals in each category
 that have been evaluated in cancer biological
 assays and the number that have been found to be
 carcinogenic.26 Several important points can be
 seen from this table: 1) only 10% of the chemi-
 cals detected hi ah" have been screened in short-
 term genotoxic  tests for their ability to cause
 cancer; 2) of the approximately 300 chemicals
 that have been screened, roughly 22% were found
 to be carcinogenic hi the laboratory animal stud-
 ies^ and 3)  most evaluation has been focused  on a
 few pollutant categories.8 Consequently the
 contribution of many categories of chemicals as
 airborne carcinogens cannot be estimated. Fur-
 ther, it should be noted that this analysis does not
f This percentage(of positive cancer results must be interpreted
with caution. Candidates for carcinogenicity testing often can
be identified based on short-term mutagenic assays or other
assays that detect genetic changes. Consequently, the chemicals
selected for long-term cancer bioassays are more likely to be
positive than randomly selected chemicals.

8 Some categories of chemicals (for example, hydrocarbons,
nitrogen-containing organics and halogenated organics) are
relatively well tested. Other categories of chemicals, like
ketones and carboxylic acids and their derivatives, are com-
monly detected in ambient air but have not been extensively
evaluated.
                                                   21

-------
Table 2-2. occurrence and biological test results indicating carcinogenicity of airborne
for the 2,827 chemicals that have been reported to exist in the air.
Category
inorganics
Hydrocarbons
Ethers
Alcohols
Ketones
Aldehydes
Carboxylic Acid Derivatives
Carboxyllc Acids
Heterocycllc Oxygen compounds
Nitrogen-containing Organics
Sulfur-
Containing Organics
Halogen-Containing Organics
Organometallic Compounds
GRAND TOTALS
Number of
Air Pollutants
identified
in Each
Category
260
729
44
233
227
108
219
174
93
384
99
216
41
2,827
Number of
Pollutants
that have
Been Screened
For Genotoxic
Effects3
30
51
3
28
11
6
6
5
16
59
4
71
13
503
Number of
Pollutants that
have been
Found Positive
in Genotoxic
Tests
5
12
1
1
0
4
0
0
4
22
1
16
6
72
chemicals,
Number of
Chemicals
Found to
Cause
Cancer In
Laboratory
Animals 6
4
19
0
0
0
1
2
0
7
12
1
21
0
67
a Short-term mutagenic or other genotoxic tests.
b Does not Include all human carcinogens.
Data are compiled from craedel, Hawkins and Claxton. Atmospheric Chemical
Compounds: Sources. Occurrence, and Bioassay, Academic Press, inc.. Orlando,
FL 1986.
                22

-------
                                            Section 3
                                  Previous Assessments
    The results from many previous screening
 studies have been compiled and presented in a
 1990 report entitled Cancer Risk From Outdoor
 Exposure to Air Toxics27. This report provides a
 "snapshot" of the current understanding of the air
 toxics problem.  It included emissions from all
 source types, not just area sources, including
 motor vehicle emissions. The report estimates that
 exposure to hazardous air pollutants from all
 source types accounts for as many as 1000-3000
 cancer deaths each year in the U.S.h

    Figures 3-1 and 3-2, adapted from the  Cancer
 Risk From Outdoor Exposure  to Air Toxics report,
 show the HAPs  and source categories  associated
 with the estimated HAPs-related cancer risks,
 respectively. Figure 3-1 shows that products of
 incomplete combustion (PIC), 1,3-butadiene,
 hexavalent chromium,  and benzene account for
 more than half of the overall cancer risk among
 the pollutants evaluated.'•28 ("PIC" refers  to a
 group of chemicals generated  when fuels are only
 partially burned. PIC includes the HAP listed as
 polycyclic organic matter,  or POM.) Results from
 such screening studies  suggest that a handful of
 source categories — such as motor vehicles,
 chrome elecrroplaters,  waste treatment storage
b Please note the use of the term "as many as." The risk
factors used to derive the estimates of possible cancer deaths in
the cited report are "upper bound estimates." Such estimates
are highly uncertain. The actual human risks are not known
and are expected to be lower than the "upper bound" estimates
used in the report.
and disposal facilities (TSDFs), woodstoves and
fireplaces, asbestos demolition, and gasoline mar-
keting — account for a majority of HAPs-related
cancer risks (see Figure 3-2) in these screening
studies. Lifetime cancer risk to individuals  living
in urban areas, aside from those risks obviously
associated with major sources of HAPs, typically
range from 1 in 100,000 (10'5) to 1 in  1000
(10"3). These figures demonstrate the relative
importance of controlling non-major sources of
HAPs in urban areas.

   Other cancer screening studies not covered hi
Cancer Risk From Outdoor Exposure to Air Toxics
generally suggest similar results; while data are
available for only a small number of sources and
pollutants, a relatively small subset  of these gen-
erally account for most of the currently estimated
HAPs-related cancer risk.  The comparative rank-
ings of sources and pollutants in each study vary,
depending on what cities, sources, and pollutants
are included in the  analysis,  and on methodologi-
cal differences in the risk assessments.
1 The actual risk estimates will change as new and better data
are obtained. Indeed, a recent update (Motor Vehicle-Related
Air Toxics Study, EPA 420-R-93-005, April, 1993) of mobile
source risks suggests that the relative roles of PIC and 1,3-
butadiene may be reversed. This assessment found the risk
from PIC from all urban sources may be less than that shown
and the risk from 1,3-butadiene may be greater than that found
in the Cancer Risk From Outdoor Exposure to Air Toxics
report and discussed in this section.
                                                  23

-------
                    Relative Contribution by Pollutant
                   To Total Nationwide Cancer Cases
               I Others
        [Vinyl chloride
          Carbon Tetrachloride
           Ethylene dichloride
          [Gasoline vapors
           Dioxins
            Ethylene dibromide
            Arsenic
              Asbestos
               I Chloroform
                 Formaldehyde
                      Benzene
                        Chromium, hexavalent
                              1,3-Butadiene
                                 15        20       25
                                  Percent Contribution (%)
                Other Pollutants
             Acrylonitrile
             Cadmium
             Vinylidene chloride
             Hexachlorobutadiene
             Trichloroethylene
             Coke oven emissions
             Perchloroethylene
             Hydrazine
             Ethylene oxide
             Methylene chloride
             Radon
             Other Radionuclides
             56 other pollutants
                                                                           PIC
                30
35
40
Figure 3-1. Relative contribution of various hazardous air pollutants to the estimate of nationwide
cancer cases (from Cancer Risk From Outdoor Exposure to Air Toxics).
   Very few screening studies have examined
health effects other than cancer. One such effort,
however, found that noncancer effects8'29 would
likely be expected to occur in exposed urban
populations.

   The study attempted to estimate the potential
noncancer effects of urban air pollutants, not just
the listed HAPs. The study considered pollutants
from all types of sources (not just area sources).
Outdoor air monitoring data or computer-modeled
estimates of ambient outdoor concentrations were
used to examine potential exposures to air pollut-
ants. Monitoring or modeling estimates of ambi-
ent outdoor concentrations were available for 334
air pollutants.J
j The average annual concentrations of 40 chemicals were
modeled, based on estimated emissions data that were provided
by more man 3500 individual commercial and industrial facili-
ties across the U.S. Measured outdoor concentrations, of
varying reliability and completeness, were available for more
man 300 volatile organic chemicals at more than 1000 sites in
310 cities, and for 6 trace metals in more man two million
samples from more than 1500 U.S. cities.

-------
                       Relative Contribution by  Source Categories
                     To Total  Estimated Nationwide Cancer Cases
        Treatment, Storage & Disposal Facilities
                         Woodsmoke	
               Asbestos, demolition
           Solvent use/degreasing—
             Gasoline marketing-
            Other area sources	
 Secondary formaldehyde, area-

  Secondary formaldehyde, point
       Chemical users/producers
                  Iron and steel
                  Cooling towers
             Coal & oil combustion
                       Electroplating
                            Motor vehicles
                        Other point sources
Figure 3-2. Relative contribution by source to the estimate of nationwide cancer cases per year caused
by all sources, as reported in Cancer Risk From Outdoor Exposure to Air Toxics.
   Of the 334 air pollutants with estimated ambi-
ent outdoor concentrations, information on poten-
tial noncancer health effects were available for
143 chemicals. For these pollutants, the estimated
outdoor concentrations were compared to the
lowest-observed-adverse-effect level (LOAEL)
and to a health reference level.k Concentrations of
54 of the 143 pollutants exceeded the health refer-
ence level at one or more sites; more than 20
pollutants exceeded the health reference levels at
more than 25% of the sites studied. Figure 3-3
shows the number of chemicals that exceeded
these levels. The data are grouped according to
whether the noncancer health effect was acute or
chronic, and whether the estimated concentration
k A LOAEL is the lowest dose or exposure level at which an
adverse effect has been reported in the health literature,
typically from studies conducted in laboratory animals A
health reference level is die LOAEL divided by appropriate
uncertainty factors to account for intra- and inter-species
variability. The goal is to establish an exposure level below
which die population is not expected to be affected at some
unspecified level of frequency (risk). Health reference levels
differ from Reference Concentrations (RfCs), which will be
discussed later in this document, in that health reference levels
receive much less review and validation than do RfCs.
                                                 25

-------
             Nationwide  Screening  Study
       Number of Individual Air Pollutants Exceeding Noncancer Health Levels
      Chronic Modeling

     (Includes thirty one
     chemicals with chronic
     health effects data and
     modeled concentrations.)
     Chronic Monitoring
    (Includes seventy five
    chemicals with chronic
    health effects data,
    combined with median
    values of the monitored
    concentrations.)
       Acute Monitoring
   (Includes one hundred
   nine chemicals with
   acute health effects data,
   combined with
   maritniim values of the
   monitored
                                                          Above HRL & LOAEL
                                                          Above Health Reference Level (HRL)
                                                          Below Health Reference Levels
       60
Number of Air Pollutants
Figure 3-3. Results of a screening study to identify air pollutants with potential noncancer health
effects.
was modeled or measured. An estimated 50 mil-
lion persons lived within 10 km of monitored sites
or within 2 km of facilities where modeled con-
centrations of one or more chemicals exceeded the
health reference level. For the LOAEL, the com-
parable population estimate was 19 million per-
sons. The data in Figure 3-3 are for individual air
pollutants: typically, however, several pollutants
were present in each area studied, but the effects
of simultaneous  exposure to multiple pollutants
were not considered. This screening study con-
cluded that exposure to air pollutants may pose
risks of respiratory, neurologic, and reproductive
   systems effects and a risk for adverse develop-
   mental effects, for both individual chemicals,  and
   chemical mixtures.

       None of the screening studies performed to
   date claim to demonstrate actual cause-and-effect
   relationships between routine emissions of HAPs
   (or their resulting exposures) and an observed
   disease or other health effect. As noted previ-
   ously, the risk factors used in such screening
   studies are "upper bound estimates" and are
   highly uncertain. The actual human risks are not
   known and are expected to be lower than the
                                             26

-------
"upper bound" estimates derived in such screen-
ing studies. The screening studies are useful,
however, for comparing the relative ranking of
the potential risks due to different pollutants and
sources.
                                                 27

-------
28

-------
                                         Section 4
                                     Research Needs
    The limited information currently available
 suggests that area sources contribute to air pollu-
 tion that can potentially damage public health. In
 addition, atmospheric transformation products
 formed from area source emissions might also
 contribute  to health risks. Only a limited amount
 of credible data is available with which to charac-
 terize the risks posed by exposures to urban air.
 Determining exposures, identifying the sources of
 those exposures,  estimating the likely resulting
 health impacts, and identifying needed controls
 are very complex tasks. The scope and the com-
 plexity of these tasks make it necessary to identify
 the most critical research needs. Identification of
 the critical research needs provides a framework
 for systematically gathering information about
 urban HAPs over the coming decade to support
 development and implementation of the National
 Strategy for area sources.

   To assess and manage efficiently the risks
 associated with HAPs from area sources, data
 from each compartment of the Environmental
 Health Paradigm  are needed. Consequently, the
 research needs are organized and presented using
 this paradigm.

 4.1 Research on Exposure Assess-
 ment

   The discussion of Exposure Assessment re-
search needs will address the key research ques-
tions related to Emission Sources, Environmental
Concentrations, and Human Exposures.
Emission Sources

   The key research needs for characterizing
emission sources of HAPs are organized around
the following questions:
   •  Which area sources emit HAPs, and how
       much do they emit?
   •  What are the most important sources and
       pollutants for which detailed emissions
       data must be developed?
   •  What are the most reasonable approaches
       for reducing emissions?
To assist with implementation of the National
Strategy, data on the feasibility of pollution pre-
vention or of adding emission controls must also
be addressed.

Which area sources emit HAPs, and how much
do they emit?

   Given the limited availability of high-quality
   data on emissions of HAPs from many area
   sources, research is needed to identify the
   specific types of stationary sources that meet
   the definition of an "area source" and to  char-
   acterize which HAPs they emit and in what
   quantities. Such data are critical to identifying
   the 30  or more "worst" HAPs, as required by
   the CAA. Research into methods to measure
   the emitted HAPs is fundamental to increasing
   our knowledge of area source emissions.
                                              29

-------
 What are the most important sources and pollut-
 ants for which detailed emissions data must be
 developed?

    The 30 or more "worst" HAPs have not yet
    been identified. Once the chemicals are speci-
    fied, the task of identify ing the area sources
    accounting for 90%  of the area source  emis-
    sions  of each of the  identified compounds
    becomes critical. In order to identify those
    area sources, detailed emission factors  and
    emission estimation  techniques will need to be
    developed.

 What are the most logical approaches for reducing
 emissions from area sources, in terms of potential
 benefit, technical feasibility, costs, and impacts of
 other control programs?

    Currently, the best approaches have not been
    determined. Pollution prevention approaches
    must be explored, while taking into account
    current data to define the achievable level  of
    control and the costs of control. Other  emis-
    sion control programs, notably efforts to limit
    precursors of ozone  (some of which are also
    HAPs; others of which might produce HAPs
    through transformation processes, etc.), can
    indirectly benefit the National Strategy, and
    the benefits from those programs must  also be
    considered.

Environmental Concentrations

    The key research issues under the Environ-
mental Concentrations component deal with col-
lecting the ambient data, considering the impacts
of atmospheric transformation, and developing
methods to make use of the ambient monitoring
data. The  key research questions are:
    •   What are  the concentrations of HAPs from
       area sources?
    •   How does atmospheric transformation
       increase public risks?
    •  How can monitoring and modeling best be
       used to assess the effectiveness of the
       National Strategy?

What are the concentrations of HAPs from area
sources, both from direct emissions and as sec-
ondary products, to which people are exposed?

    Research is needed to develop methods to
    measure not only the listed HAPs, but the
    myriad potentially harmful chemicals present
    in urban air.  Data are also needed to assess
    just how much monitoring is needed (for
    example, number of cities needed to provide a
    "representative" sample, the number of sites
    per city and the  distances between sites, and
    the frequency of sample collection)  to charac-
    terize the urban  levels to which people are
    exposed.

How does atmospheric transformation increase
public risks?

    Research is needed to determine if the muta-
    genic transformation products  formed in urban
    air are actually a hazard to human health, and
    if so, to identify the specific transformation
    products and any other  necessary precursors
    that are responsible for  the potential elevated
    risks. Only then can reasonable  steps be taken
    to mitigate or prevent the exposure  to and risk
    from these transformation products.

How can ambient monitoring best be used with
available modeling methods (including  emissions
modeling, dispersion modeling, and source appor-
tionment modeling) to demonstrate the  effective-
ness of the National Strategy (as required  in the
CAA)?

    Critical components of  this research are: 1)
    defining how to use ambient outdoor monitor-
    ing data to establish a "baseline" (the concen-
    trations existing before  the National Strategy
                                               30

-------
    is implemented) and to determine concentra-
    tion trends to measure the effectiveness of the
    National Strategy; 2) identify ing the other
    factors (like wind speed, wind direction and
    the mixing depth, source emission profiles,
    and the distribution of sources throughout the
    urban area) that must be measured in order to
    derive an estimate of total area source emis-
    sions from the measured ambient outdoor
    concentrations; 3) developing data analysis
    methods to  allow the trend in area source
    emissions to be determined despite  "noise"
    from natural variations (like those caused by
    year to year changes  hi weather) and from the
    trends of point sources and mobile sources;
    and 4) determining if ambient outdoor data
    indicate that all area sources of the  controlled
    HAPs have been recognized (that is, do the
    ambient concentrations reconcile with EPA's
    understanding of the emission sources?)

Human  Exposures

    The key research questions for Human Expo-
sures are:
    •  What are the human exposures to HAPs?
    •  What are the routes of exposure?

What is the distribution of human exposures to
the various HAPs? By what route, and  how effec-
tively, do the HAPs reach humans?

   Data are needed to define how people's activi-
   ties and the  concentration of the HAPs vary
   with time and to characterize how that varia-
   tion will affect the distribution of exposures.
   Research is also needed to define those cir-
   cumstances that will lead to high exposures
   and high potential risks, including research to
   identify the chemicals and circumstances  that
   make indirect exposures important.
4.2 Research on Effects Assessment

    As with Exposure Assessment, there is a need
for more research into Effects Assessment. Two
areas that need additional research are Internal
Dose and Health Effects.

internal Dose and Health Effects

    Critical issues facing health effects researchers
in trying to define the potential human health
effects of hazardous air pollutant emissions from
area sources are:
    •  How can the most substantial hazards from
      HAPs be identified?
    •  How can health risks be estimated reli-
      ably?

How can the most substantial hazards from HAPs
be identified?

    Hazard identification research is needed to
    develop, refine, and validate methods for
    identifying chemicals and agents that pose
    potential human hazards. Faster, more accu-
    rate, less expensive, and more reliable tech-
    niques are needed to determine cause and
    effect relationships between environmental
    pollutants and adverse health outcomes than
    the methods that are currently available. Bat-
    teries of test methods designed to  evaluate
    potential hazards comprehensively also need to
    be validated. A comprehensive program to
    collect toxicity data also is needed. Efforts
    should include evaluation of realistic scenarios
    for concentrations and exposures.

    Additionally, field studies that evaluate the
    biological effects of exposure to urban air
    pollution are needed. These field studies
    should combine short-term methods developed
    in the laboratory to screen for problem chemi-
    cals, mixtures, and/or sources, and longer-
                                               31

-------
   term studies to describe in more detail the
   hazards of urban air pollutant exposures.

How can health risks be estimated reliably?

   Improved methods are needed to link ambient
   exposures to internal dose. Efforts in this area
   should include development and validation of
   biological markers for exposure, effects, and
   susceptibility in  human populations; and im-
   provement hi pharmacokinetic models. These
   models use physiological and biochemical data
   to estimate internal doses resulting from exter-
   nal exposures. These efforts improve the
   confidence in extrapolation of animal data to
   humans and from the high doses used in labo-
   ratory studies  to the lower doses more typical
   of human exposures.

   Dose-response research is needed to develop
   biologically based dose-response models that
   elucidate: 1) the relationship between exposure
   concentration (or, the applied dose) and the
   dose at the site of toxics action  (that is,  the
   target dose) and  2) the basic biological mecha-
   nisms responsible for the observed effects.
   Understanding of underlying biological mecha-
   nisms  is crucial to the accurate extrapolation
   of research results (for example, extrapolation
   of results from animals to humans, from high-
   to low-dose, and from  "across-exposure sce-
   nario" effects.) These models estimate the
   type and extent of biological damage resulting
   from doses to the affected tissues, which,
   when coupled with exposure  data, provides
   estimates of public health risks.

   In addition, because HAPs in the environment
   never occur alone, predictive models for risk
   assessment of complex mixtures of HAPs are
   needed: the most urgent needs include tech-
   niques to compare potencies of various mix-
   tures, to understand the mechanisms of chemi-
   cal interactions hi complex mixtures, to  identi-
fy the most critical components leading to
biological activity in complex mixtures, to
determine the quantity of the biologically
active components that reach susceptible or-
gans or tissues hi exposed people, and to
develop biological markers of exposure and
effects. These efforts are necessary to enable
evaluation of the risks from environmental
mixtures of pollutants.

Lastly, research hi environmental epidemiolo-
gy is  needed to assess the impact of exposure
to HAPs on the general population and to
establish the link between environmental
exposures and human health effects. Identifi-
cation of appropriate biomarkers of exposure
and effects are likely  to be necessary to make
such studies feasible for  many pollutants.
                                               32

-------
                                         section 5
                        Summary of Preliminary Findings
    Much of the discussion in this report has been
 framed around the Environmental Health Para-
 digm.  Without at least some understanding of
 each component hi the paradigm, it is impossible
 to develop reliable risk assessments. Adequate
 data exist only for a few HAPs. Screening stud-
 ies, like those referred to in Section 3,  are helpful
for outlining the potential dimensions of the urban
 HAP problem, but these studies are often based
 on incomplete, inadequate, and unreliable data.
From a strictly scientific perspective, such studies
are suggestive; however, they might not be suffi-
ciently comprehensive or reliable to use for identi-
fying the "worst" HAPs from area sources, or to
use as the basis for the National Strategy.  In the
following discussion, the summary of preliminary
findings on what is currently known about HAPs
from area sources in urban areas  is organized
according to the components of the Environmental
Health Paradigm.

Emission Sources

•  A total of 42 HAPs appear to have  "Fair or
   Better" emissions data for all (not just area)
   sources. (Seventeen HAPs are regularly in-
   cluded in the available urban area emission
   inventories; an additional twenty-five HAPs
   either have national inventories or have vali-
   dated emission factors in the FIRE data base.)
   Detailed area source  information in most
   urban area  HAP emission inventories is limit-
   ed.  Much of the data (including data available
   under Title m of the Superfund Amendments
   and Reauthorization Act) is considered to be
   incomplete, out-of-date, or limited in scope
   and application. More than 120 HAPs have
   little or no validated source emissions data.

•  Emission factors, source activity data, and
   other emission estimation techniques are of
   questionable quality or are currently unavail-
   able for a number of area sources of HAPs.

Environmental Concentrations

•  There are no measurements of the air concen-
   trations of almost 40% of the listed HAPs.
   Another 20%  of the HAPs have very little
   monitoring data. For a few compounds, there
   are considerable monitoring data collected at  a
   variety of locations. The ability to measure the
   HAPs is severely limited by the lack of meth-
   ods to collect and  analyze many of the listed
   chemicals.

•  Atmospheric transformations complicate expo-
   sure assessment because they can increase or
   decrease the environmental concentrations of
   the listed HAPs. In addition, sunlight causes
   reactions  among pollutants in urban air that
   can produce a variety of products, some of
   which are potentially even more harmful than
   the original pollutants. HAPs might be formed
   from non-hazardous precursors, some of
   which are emitted in large amounts into urban
   air.
                                              33

-------
 Human Exposures

 •  Outdoor sources of HAPs form the baseline
    for human exposure, on top of which HAPs
    from indoor, workplace, and personal use
    sources add additional exposures. For some of
    the HAPs, such interior sources  may be very
    commonplace and may frequently increase
    interior concentrations substantially above
    outdoor concentrations. For many of the
    gaseous HAPs,  the indoor concentrations due
    to outdoor sources are equal to the outdoor
    concentrations.  For other HAPs, the indoor
    concentrations attributable to outdoor sources
    are expected to  be somewhat less because of
    physical or chemical losses as the HAPs are
    transported indoors. For HAPs attached to
    fine particles hi the ah", the indoor concentra-
    tions from outdoor sources are expected to  be
    50-90% of the outdoor concentrations.

 •  Available human exposure  data often do not
    describe well those situations that can lead to
    very high exposures to area source emissions
    (for example,  living above a dry cleaning
    establishment or adjacent to a  gas station).

Internal Dose

•   Estimating the amount of HAP that reaches
    affected or susceptible  organ(s) and causes
    damage to health is important  in understanding
    the relationship between exposures to HAPs
    and the nature and magnitude  of potential
    public  health effects. This is particularly true
    when risk estimates are based  on extrapolated
    information. Current methods  and data for
    estimating Internal dose are often crude. Good
    information exists only for a few HAPs.

Health Effects

•   There are some health  effects data available
    for each of the 189 HAPs. In  almost no case,
   however, are there data available on all of the
   most important health effects: cancer, devel-
   opmental and reproductive effects, neurotoxic-
   ity, and short-term and long-term pulmonary
   effects. The quality of the available data var-
   ies, ranging from inadequate to excellent.

•  The evaluation of the cancer-causing potential
   of the HAPs is more complete than for other
   health effects. Also,  reference concentrations
   (RfCs) for noncancer health effects have been
   developed for 40 of the listed HAPs. Values
   of the cancer risk estimates and of the RfCs
   are likely to change as new and better data be-
   come available.

•  Only 10% of nearly 3,000 chemicals that can
   exist as air pollutants have been tested for
   genotoxicity or carcinogenicity. The number
   of chemicals tested for noncancer effects is
   even smaller. The  development of data on this
   broad  range of substances is almost certainly
   not warranted. Further analysis is needed to
   target  specific chemicals for further evalua-
   tion.

•  People are exposed to mixtures of many pol-
   lutants simultaneously, not just one pollutant
   at a time. Yet,  how these mixtures of pollut-
   ants interact to affect human health is only
   poorly understood.

Synopsis

   The availability of data on the 189 HAPs that
are needed to do a complete environmental health
assessment is illustrated  in Figure 5-1. It reveals
that very little is known about many of the HAPs,
while significant amounts of information exist for
a few chemicals. The  same data are given for
each of the 189 listed HAPs in Table  A-l,  found
in the Appendix. A review of Table A-l reveals
that 20 chemicals have enough data to merit "Fair
or Better" classifications in Source Emissions
                                               34

-------
           Availability of Data for Various Categories
                          For the 189  Listed HAPs
              Source Emissions Data
        Uttie or No Information
        Ambient Concentration Data
                       No Data
                                                     Little Data
                               Fair or Better Data                  ^••f^' Fair or Better Data
            Occasionally Found                       Occasionally Observed
              Noncancer Effects Data

          No validated RfC
            Carcinogenicity Data
Unknown Carcinogenicity ^t	> Devaluated HAPs
                     59
              ASsKV
 Possible Carcinogen
                                Moderate or High Confidence in RfC

                             Low Confidence in RfC
                                                                   Probable or Known Carcinogen
 Figure 5-1. Summary of the available data on the 189 listed HAPs. (Table A-l, Appendix A,
 categorizes the data for each of the 189 HAPs.)
 Data, Ambient Concentration Data, and in one of
 the Health Effects areas, either Noncancer Health
 Effects or Cancer Health Effects. This list of
 chemicals does not identify the 30 or more
 "worst" HAPs; rather, the list simply identifies
 those HAPs with sufficient data to begin a risk
 assessment of either the cancer or noncancer
 effects due to exposure to that chemical. Another
 20 HAPs are rated "Fair or Better" in two of the
 three required areas. Targeted research on this
 second group of HAPs could readily provide
 sufficient data to allow a risk assessment to be
initiated. The 40 HAPs with the most complete
  available data are listed in Table 5-1.

     Continuing research will undoubtedly improve
  the scientific understanding of human exposures
  and health effects from increasing numbers of
  HAPs.
                                            35

-------
Table 5-1. The HAPs with the most extensive available data needed for a risk assessment.
 HAPs with data rated "Fair or Better" in
 the three areas:
    • Source Emissions
    • Ambient Concentrations
 and
    • Health Effects (Cancer or Noncancer)
HAPs with data rated "Fair or Better" in
two of the following three areas:
   • Source Emissions
   • Ambient Concentrations
and
   • Health Effects (Cancer or Noncancer)
 Benzene
 1,3-Butadiene
 Carbon tetrachloride
 Chloroform
 Ethylene dibromide
 Ethylene dichloride
 Formaldehyde
 Methylene chloride
 Styrene
 Tetrachloroethylene
 Toluene
 Trichloroethylene
 Vinyl chloride
 Arsenic compounds
 Chromium compounds
 Lead compounds
 Manganese compounds
 Mercury compounds
 Nickel compounds
 Selenium compounds
Acetaldehyde
DDE (p,p'-dichlorodiphenyldichloro-
ethylene)
1,4-Dichlorobenzene
Ethylbenzene
Ethylene oxide
Hexachlorobenzene
Hexane
Methyl bromide
Methyl chloroform
Pentachlorophenol
Polychlorinated biphenyls
Propylene dichloride
2,3,7,8-Tetrachlorodibenzo-p-dioxin
2,4,6-Trichlorophenol
Vinylidene chloride
Xylenes (mixed isomers)
Antimony compounds
Beryllium compounds
Cadmium compounds
Polycyclic Organic Matter
                                             36

-------
References
     37

-------
 1.     The Clean Air Act (42 U.S.C. 7401-7626) consists of Public Law 159 (July
       14, 1955; 69 Stat. 322) and the amendments made by subsequent enactments,
       including'Public Law 101-549 (The Clean Air Act Amendments of 1990)
       approved November 15, 1990.

 2.     Andur, M. O. "Air Pollutants"  In: Toxicology:  Basic Science of Poisons. Eds.
       M. O. Andur, J.  Doull, C. Klaassen, Pergamon Press, NY, NY. 1994, p.
       856.

 3.     World Health Organization and  United Nations Environment Programme.
       Urban Air Pollution in Megacities of the World. Blackwell Publishers, Oxford,
       UK. p. 7, 1992.

 4.     World Health Organization. Air Quality Guidelines for Europe.  WHO Regional
       Publishing, Copenhagen,  DK. 1987.

 5.     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. 8468-
       8471 and pp. 8489-8492.

 6.     Sexton, K., S.G.  Selevan, D.K. Wagener, and J.A. Lybarger, Estimating
       Human Exposures to Environmental Pollutants:  Availability  and Utility of
       Existing Databases. Archives of Environmental Health. 47: 398,  1992.

 7.     Cote, I. L. and J. J. Vandenberg, "Overview of Health Effects and Risk
       Assessment Issues Associated with Air Pollution,"  in The Vulnerable Brain,
       Vol. Ill, Isaacson and Jensen, editors, Plenum Press,  in press.

8.     U.S. Environmental Protection Agency, Toxic Air Pollutants and Noncancer
       Risks:  Screening  Studies, External Review Draft, September, 1990.

9.     Jones, J. W., D.  Campbell, D.L. Jones, S. Kersteter and M. Saeger,
       "AEERL's Hazardous Air Pollutant Emissions Research Under JEIOG:  A
       Status Report" Proceedings VIP-27, Emission Inventory Issues,  Specialty
       Conference, Durham, NC. October 19, 1992.

10.     Sullivan, D.,  T. Lahre and M. Alford, Assessing Multiple Pollutant Multiple
       Source Cancer Risks from Urban Air Toxics, Office of Air Quality Planning
       and Standards, Research Triangle Park, NC. EPA-450/2-89-010. (NTIS PB89-
       197222), April, 1989.

11.     Ramamurthi,  M. T.J. Kelly and C. W.  Spicer, Temporal and Spatial Variabil-
       ity  of Toxic VOC  Sources in Columbus,  Ohio, Measurement of Toxic and
       Related Air Pollutants: Proceedings of the 1993 EPA/A&WMA International
       Specialty Conference, Durham,  NC. May, 1993. EPA/600/A93/024.
                                        38

-------
 12.    Lewis, C. L., T.L. Conner, R.K. Stevens, J.F. Collins and R.C. Henry,
       "Receptor Modeling of Volatile Hydrocarbons Measured in the 1990 Atlanta
       Ozone Precursor Study," Proceedings of the 86th Annual Meeting of the Air &
       Waste Management Association, Volume II, 93-TP-58.04, Denver, CO.  June,
       1993.

 13.    Kelly, T.J., M. Ramamurthi, A.J. Pollack, C.W. Spicer, J. Shah, D. Joseph
       and L.T. Cupitt, "Surveys of the 189 CAAA Hazardous Air Pollutants:  I
       Atmospheric Concentrations in the U.S.", Measurement of Toxic and Related
       Air Pollutants: Proceedings of the 1993 EPA/A&WMA International Specialty
       Conference, Durham, NC, May, 1993.  EPA/600/A93/024.

 14.    U.S. Environmental Protection Agency, Integrated Air Cancer Project:
       Summary Report, Office of Research and Development, Research Triangle
       Park, NC. 1993.

 15.    Buxton, B.E.  and A.D. Pate, "Joint Temporal-Spatial Modeling of Concentra-
       tions of Hazardous Pollutants in Urban Air," presented at Forum on Geo-
       statistics for the Next Century, Battelle, Columbus, OH.  1993.

 16.    Robinson, J. and J. Holland,  "Trends in Time Use,"  in Technology and the
       American Economic Transition,  OTA-TET-283, Office of Technology Assess-
       ment, 1988.

 17.    Lewis, C. L.  "Sources of Air Pollutants Indoors: VOC and Fine Paniculate
       Species," J. Exposure Analysis and Environ., 1:31-44, 1991.

 18.    Harley,  R. A. and G. R. Cass, "Modeling the Concentrations of Gas-Phase
       Toxic Organic Air Pollutants: Direct Emissions and Atmospheric Formation,"
       Environ. Sci. Technol., 28, 88-98, 1994.

 19.    Kleindienst, T.E., P.B. Shepson, E.G. Edney, L.D. Claxton, and L.T. Cupitt,
       "Wood smoke: measurement of the mutagenic activities of its gas- and parti-
       cle-phase photooxidation products," Environmental Science & Technology,
       (20), 493-501, 1986.

20.    Cupitt, L. T., L. D. Claxton, T. E. Kleindienst, D. F. Smith and P. B.
       Shepson, "Transformation of Boise Sources: The Production and Distribution
       of Mutagenic Compounds in Wood Smoke and Auto Exhaust,"  Proceedings of
       the 1988 EPA/APCA International Symposium on Measurement  of Toxic and
       Related Air Pollutants, EPA-600/9-88-015 (NTIS PB90-225863), pp. 885-889,
       1988.
                                       39

-------
 21.    Ott, W. R.  "Total Human Exposure: Basic Concepts, EPA Field Studies, and
       Future Research Needs," Journal of the Air & Waste Management Association,
       Vol. 40, No. 7, 966-975 (1990).

 22.    Wallace, L. "The Team Studies," EPA Journal, Volume 19, Number 4, EPA
       175-N-93-027, October-December 1993.

 23.    See, for example: Sexton, K.  "National human exposure assessment survey,"
       U.S. EPA Position Paper, Office of Health Research, US EPA, Washington,
       DC, 1991. or
       Clickner, R.,  G. Kalton and A. Chu, "Statistical Design Issues in Human
       Exposure Assessment Surveys: Sample  Design Issues and Options", Delivery
       Order 23 Report, EPA Contract 68-W1-0019. Westat, Inc., Rockville, MD,
       July, 1993.

 24.    U.S. Environmental Protection Agency. Guidelines for Carcinogen Risk
       Assessment. Carcinogen Assessment Group, Office of Health and Environmen-
       tal Assessment. Washington, D.C. 51 FR 33992. September 24,  1986.

 25.    Schuetzle D, and J. Lewtas. "Bioassay-directed Chemical Analysis in Envi-
       ronmental Research," Anal. Chem. (58):1060A-1075A,  1986.

 26.    Graedel, T.E., D.  T. Hawkins and L.D. Claxton,  Atmospheric Chemical Com-
       pounds: Sources, Occurrence,  andBioassay, Academic Press, Inc., Orlando,
       FL. 1986.

 27.    Meardon, K.,  Cancer Risk from Outdoor Exposure to Air Toxics, Vols.  1 and
       2, Office of Air Quality Planning and Standards, Research Triangle Park, NC.
       EPA-450/l-90-004a,b (NTIS PB91-159624, -159632), September, 1990.

28.    U.S. Environmental Protection Agency, Motor Vehicle-Related Air Toxics
       Study,  Office of Mobile Sources, Ann Arbor, MI.  EPA 420-R-93-005.  April,
       1993.

29.    Cote, I., L. T. Cupitt and B. M. Hassett, "Toxic Air Pollutants and Noncan-
       cer Health Risks" in Risk Analysis: Prospects and  Opportunities, C.  Zervos,
       editor, Plenum Press, New York, NY,  pp.  697-706, 1991.
                                       40

-------
Appendix
     41

-------
       Table A-l provides a listing of the available data for each of the 189 listed
HAPs. The characterizations are consistent with those used in the text of the report.
The shaded chemicals are the 20 HAPs identified in Table 5-1 as having "Fair or
Better" data in three categories: Emission Sources, Environmental Concentrations,
and Health Effects (either Cancer Effects or Noncancer Effects).
Meaning of the Symbols Used hi the Table:
       Source Emissions Data
Blank
                                        /
      Ambient Concentration Data
       Seldom included in emissions in-
       ventories (Little or no data)
       Occasionally included in emissions
       inventories (i.e., included in 10%
       or more, but less than half, of
       emission inventories studied)
       Routinely included in emissions
       inventories (i.e., in 50% or more
       of the case studies), or national
       inventory is available, or emission
       factors included in FIRE data base
Blank  Little or no (71 HAPs) ambient
       data available
/     Between 100 and 1000 observations
       More than 1000 observations
      Noncancer Health Effects Data     Blank No validated RfC available
/
      Cancer Health Effects Data
                                              RfC available, but "low" confi-
                                              dence
                                              RfC available, "moderate" or
                                              "high" confidence
Blank  Unclassified (59 HAPs) or Class D
       (22 HAPs)
/     Class C carcinogen or LARC Class
       2B chemical
       Class A or B carcinogen or IARC
       Class 1 or Class 2A chemical
                                        42

-------
Table A-l. Availability of data on the 189 listed HAPs.
NO.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
/*
16
17
18
19
20
21
Chemical Name
Acetaldehyde
Acetamide
Acetonltrlle
Acetophenone
2-Acetylamlnofluorene
Acroleln
Acrylamide
Acrylic acid
Acrylonltrlle
Ally! chloride
4-Amlnobiphenyl
Aniline
o-Anlsldlne
Asbestos
- ; , - x - - - \ .:!,; ' - "

Benzldine
Benzotrichlorlde
Benzyl chloride
Blphenyl
Bls(2-ethylhexyl)phthalate
Bls(chloromethyOether
Frequency of Occur-
rence
Source
Emis-
sions
Data






/

/
/



/



'

/

Ambient
Concen-
tration
Data
ss







s
s



'
f f


^



Health Effects Data
Noncan-
cer
s




SS

ss
ss
s

s


; - -I






Cancer
ss
s


ss
/
//

ss
s
ss
ss
s
ss
-'**'
ss
SJ
ss

ss
ss
                                        43

-------
No.
22
23
24
25
26
27
28
20
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
Chemical Name
Bromoform
1T5«Butatfieft&
Calcium cyanamlde
Caprolactam
Captan
Carbaryl
carbon disulflde
carboB tetrachiorwe
Carbonyl sulflde
Catechol
Chloramben
Chlordane
Chlorine
Chloroacetlc acid
2-Chloroacetophenone
Chlorobenzene
Chlorobenzllate
ctttomfdrm
Chloromethyl methyl ether
Chloroprene
Cresols (Isomers and mixture)
o-Cresol
m-Cresol
p-Cresol
Frequency of Occur-
rence
Source
Emis-
sions
Data
/
vV





SS









JJ






Ambient
Concen-
tration
Data
/
//


/
/

ss



s



ss

ss

/




Health Effects Data
Noncan-
cer














s




SS




Cancer
//
SS





ss



//




//
ss
//

/
/•
/
/
44

-------
NO.
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
Chemical Name
Cumene
2,4-D, salts & esters
DDE (72-55-9: p.p'-dlchlorodiphenyldichloroethylene)
Diazomethane
Dibenzofuran
1,2-Dlbromo-3-chloropropane
Dlbutyl phthalate
1,4-Dlchlorobenzene
3,3'-Dlchlorobenzldlne
Dlchloroethyl ether
1,3-Dlchloropropene
Dichlorvos
Dlethanolamlne
N,N-Dimethyl aniline (also, dlethyl)
Dlethyl sulfate
3.3'-Dlmethoxybenzldlne
4-Dlmethyiamlnoazobenzene
3,3'-Dlmethylbenzldlne
Dlmethylcarbamoyl chloride
Dlmethylformamlde
1,1-Dlmethylhydrazlne
Dimethyl phthalate
Dimethyl sulfate
Frequency of Occur-
rence
Source
Emis-
sions
Data




//

//
/




/










Ambient
concen-
tration
Data
//
//
//




//


/•
/











Health Effects Data
Noncan-
cer





SS

ss


ss
ss







ss



Cancer


ss


ss

ss
ss
ss
ss
ss


ss
ss
ss
ss
ss
s
ss

ss
45

-------
NO
69
70
71
72
73
74
75
76
77
78
79
80
»1
82
83
84
85
86
,«M
88
89
90
91
Chemical Name
4,6-Dinltro-o-cresol & salts
2,4-Dinltrophenol
2,4-Dlnltrotoluene
1,4-Dloxane
1 ,2-Dlphenylhydrazlne
Eplchlorohydrln
1,2-Epoxybutane
Ethyl acrylate
Ethylbenzene
Ethyl carbamate
Ethyl chloride
£awleaedl&ram&i»
Sttiylene dlchtoridsi
Ethylene glycol
Ethylenimlne
Ethylene oxide
Ethylenethlourea
Ethylldene dlchlorlde
i^m»sktei*y
-------
NO.
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
Chemical Name
Hexachloroethane
Hexamethylene-l.e-dilsocyanate
Hexamethylphosphoramlde
Hexane
Hydrazine
Hydrochloric acid
Hydrogen fluoride
Hydroguinone
isophorone
Llndane
Malelc anhydride
Methanol
Methoxychlor
Methyl bromide
Methyl chloride
Methyl chloroform
Methyl ethyl ketone
Methyl hydrazlne
Methyl iodide
Methyl Isobutyl ketone
Methyl isocyanate
Methyl methacrylate
Methyl tert-butyi ether
4^'-Methylenebis(2-chloroanlllne)
Frequency of Occur-
rence
Source
Emis-
sions
Data
SS




ss
ss






/
/
//
//







Ambient
Concen-
tration
Data



//





/•


^
SS
ss
ss
/

/



^

Health Effects Data
Noncan-
cer

^/
/
//

/







SS


s
ss
J



ss

Cancer
/

/

^/



/
SS




V


ss
/




ss
47

-------
No.
-m
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
Chemical Name
»ea»$et»<&lortflfr
Methylene dipnenyl dilsocyanate
4.4'-Methylenedianillne
Naphthalene
Nitrobenzene
4-Nltrobiphenyl
4-Nitrophenol
2-Nltropropane
N-Nltroso-N-methylurea
N-Nltrosodlmethylamlne
N-Nltrosomorpholine
Parathlon
Pentachloronltrobenzene
Pentachlorophenol
Phenol
p-Phenylenedlamine
Phosgene
Phosphlne
Phosphorus
Phthallc anhydride
Polychlorinated blphenyls
1,3-Propane sultone
beta-Propiolactone
Proplonaldehyde
Frequency of Occur-
rence
Source
Emis-
sions
Data
SS

/

/




/
J


ss
ss





ss



Ambient
Concen-
tration
Data
SS



s






/"








j


s
Health Effects Data
Noncan-
cer

SS





j
















Cancer
//






//
^/
//
/
s
s
SS






ss
SS
SS

48

-------
NO.
140
141
142
143
144
145
*4$
147
148
149
•m*
151
*&"
153
154
155
156
157
158
'JsVSV
160
161
162
163
Chemical Name
Propoxur
Propylene dichlorlde
Propylene oxide
1,2-Propylenimlne
Qulnollne
Qulnone
$tY**n*
Styrene oxide
2,3,7,8-Tetrachlorodlbenzo-p-dloxln
1 ,1 ,2,2-Tetrachloroethane
t^*s^|j?ro0toy(^i$ - ' ^ : -- • •• - *% ,
Titanium tetrachlorlde
Yfr(uenVx - m - ,/ "" ^'-;- c"- -
2,4-Toluenedlamlne
Toluene-2,4-dllsocyanate
o-Toluldlne
Toxaphene
1.2,4-Trlchlorobenzene
1,1,2-Trlchloroethane
\t,--^ X- \ y ' 0x*xV ° «0 ; , - >^oN% I - '- - ' *
*3$ftto&$fr\&t& * <^1V ,;H; i 4\\ ^ -,- '
2,4,5-Trlchlorophenol
2,4,6-Trichlorophenol
Trlethylamlne
Trifluralln
Frequency of Occur-
rence
Source
Emis-
sions
Data

/
/



M

ss

JS

\.-ff"






JS
ss
ss


Ambient
Concen-
tration
Data
/
//




ss

/
//
Jf

JS



/
/
//
"-v;r;:'l



/
Health Effects Data
Noncan-
cer

//
//



/y



•.

/y






zff^sj
<„, \'f^\-
{', ,,,\ ' * '


s

Cancer
^/
^/
SS



s
ss
ss
/
ss


ss

ss
ss

/
''-"ss -

ss

/
49

-------
NO.
164
165
166
«7
168
169
170
171
172
173
174
175
176
177
178
179
180,
181
^182:
18$
184'
185
xW
Chemical Name
2,2,4-Trlmethylpentane
Vinyl acetate
Vinyl bromide
Vlrtyl tfcloritie
Vinylldene chloride
xylenes (mixed isomers)
o-Xylene
m-xylene
p-Xylene
Antimony compounds
Arsftrtfp OOmpOUntfe  /'.
*&$'^
50

-------
NO.
187
188
m
Chemical Name
Polycycllc organic Matter (various PAHs)
Radlonuclldes
Selenium compounds butf fde, dfeuif Itte ar otfw
Frequency of Occur-
rence
source
Emis-
sions
Data
SS

//
Ambient
Concen-
tration
Data
/

vV
Health Effects Data
Noncan-
cer


Cancer
^S^^^^^^^^^^^E

//
51

-------
       Table A-2 identifies the types of health effects, other than cancer (referred to
as noncancer effects), that have been reported for the listed HAPs. The table presents
data for only those HAPs that have produced effects in humans or animals by
inhalation exposure.
                                        52

-------
SYSTEM
or HEALTH EFFECT
Bone
Cardiovascular
Death
Dermal
Reproductive/
Developmental
Endocrlne/Exocrlne
Ocular
Gastrointestinal
Hematopoletlc
Hepatic
Immunologic
Multiple
Neurologic/Behavioral
Olfactory
Pancreatic
Renal
Respiratory
spleen
Systemic
LD50
Total Number of HAPs
Showing an Effect
EXPOSURE DURATION
ACUTE
0
52
56
33
7
24
96
63
38
51
14
1
107
8
1
47
114
3
48
65
142
SUBCHRONIC
6
30
15
21
54
24
44
31
49
69
24
0
74
15
0
49
- 78
18
71
0
122
CHRONIC
2
6
1
3
13
8
7
7
16
27
5
1
20
6
1
23
30
11
27
0
57
Table A-2. Number of hazardous air pollutants that have been reported to produce health effects In humans
or animals by Inhalation exposure.
                                                  53

-------
54

-------
Glossary
    55

-------
 Term
 Definition
 Accidental re-
 lease

 Accuracy
 Acute effects
 Adverse health
 effects
 Air quality mod-
 eling
 Air toxics
Ambient air
Ambient concen-
tration

Ambient mea-
surement

Ambient monitor-
ing
 Emissions resulting from an unpredicted failure of a system due
 to which  some harm results.
 The quality of being free from error. The degree of accuracy is a
 measure of the uncertainty in identifying the true measure of a
 quantity at the level of precision of the scale used for quantity.

 Toxic effects of a substance which become manifest after only a
 short period of exposure of a duration measured in  minutes,
 hours, or days.
 An undesirable antagonistic consequence to human health due to
 some causative agent.
 A mathematical representation of pollutant concentrations and
 their distribution in the atmosphere based upon assumptions or
 simulations of pollutant emissions, meteorological dispersion and
 transport, chemical and physical reactions, etc.

 An expression commonly used to refer to hazardous air pollutants
 — often used  interchangeably with "hazardous air pollutants."
 Any air pollutant (excluding those pollutants for which ambient
 criteria do exist,  namely ozone, sulfur dioxide, carbon monoxide,
 nitrogen oxides, lead, and particulate matter) that may cause any
 of a wide range of potential harmful effects.

 The surrounding  or encompassing atmosphere. In the context of
 pollution  monitoring,  ambient air is often erroneously used to
 refer only to "outdoor" air, even though indoor air is "ambient"
 to a person who is indoors.

 The concentration of a chemical (usually, a  pollutant) in the
 atmosphere surrounding humans or other potentially affected
 receptors.

Measurement of a chemical (pollutant) found in the atmosphere
 surrounding humans or other receptors, any potentially affected
species or ecosystem.

Measuring the concentrations of pollutants or other species in
ambient air.
                                         56

-------
 Term
 Area source
 of hazardous air
 pollutants
 Area Source Na-
 tional Strategy
Atmospheric
transformation
Bacterial mutage-
nicity

Bioassay
Definition

A stationary source which annually releases to the atmosphere, or
has the potential to release considering controls,  less than 10 tons
of a single hazardous air pollutant listed hi the Clean Air Act or
less than 25 tons of a mixture of these pollutants. The term  "area
source" shall not include motor vehicles or nonroad vehicles
subject to regulation under Title II of the Clean Air Act.

The National Strategy mandated hi Section 112(k) of the Clean
Air Act. By November 1995, EPA must "prepare and transmit to
Congress a comprehensive strategy to control emissions of haz-
ardous air pollutants from  area sources hi urban areas." The
strategy shall "identify not less than 30 hazardous air pollutants
which, as the result of emissions from area sources, present the
greatest threat to public health hi the largest number of urban
areas." The strategy also shall "identify the source categories or
subcategories" emitting the 30 or more hazardous air pollutants
and "shall assure that sources accounting for 90 per centum  or
more of the aggregate emissions of each of the 30 identified
hazardous air pollutants are subject to [emission] standards."
"The strategy shall achieve a reduction hi the incidence of cancer
attributable to exposure to hazardous ah- pollutants emitted by
stationary sources of not less than 75 per centum, considering
control of emissions of hazardous air pollutants from all station-
ary sources and resulting from measures implemented ... under
[the Clean Air Act] or other laws."

The chemical reactions hi  the atmosphere,  many  of which occur
naturally and are unavoidable, that change (transform) one sub-
stance hi the air into a different chemical or chemicals;  or the
physical processes (like washout into rain water or adsorption
onto particles) that change the form of the chemical hi the atmo-
sphere and affect its distribution hi the environment.

Refers to the use of bacteria to assess the mutagenic potential of
pollutants.

Determination of the biological  activity or potency of a substance
by testing its effect on an organism. As used in this report,  a test
for carcinogenicity in laboratory animals (generally, rats and
mice) that includes near-lifelong exposure to the  agent (pollutant)
under test. The term is used interchangeably with "animal test."
                                          57

-------
 Term
 Definition
 Bioassay-directed
 chemical identifi-
 cation techniques


 Biomarkers
 Cancer risk

 Carcinogen


 Carcinogenicity

 Characterizing
 emission sources
 Chronic effects
Data base

Developmental
disorders or
effects

Direct emissions
Distribution of
human exposures

Dose
Dose-response
relationship
 A combination of chemical and physical separation and identifi-
 cation methods with short-term biological tests in order to identi-
 fy the chemicals in a complex mixture of pollutants that have a
 potential biological effect.
 Surrogates or indicators of biological exposure, dose, or effect.
 Part of the considerations under the Internal Dose component of
 the Environmental Health Paradigm.

 The risk of developing cancer.

 A substance or agent that tends to produce cancer in living
 organisms.
 The ability of a substance or agent to produce cancer.

 Describing the pollutants emitted by a source, including the
 chemical composition, the quantity emitted as a function of time,
 and the location and relevant operational parameters of the
 source.

 Toxic effects of a substance that become manifest after prolonged
 or repeated exposures of a duration measured in weeks, months,
 or years.

 Available, relevant raw information about the subject of concern.

 One type  of noncancer health effects of concern;  impairment of
 the normal development of a fetus,  infant, or child, including
 developmental retardation and birth defects.

 Emissions of a pollutant that come directly from the source,
 without having to be produced by transformations.

 A mathematical representation or other characterization of the
 range of exposures that people have to a pollutant.

 The amount of a  substance administered to an animal or human,
 usually measured in mg/kg of body weight, mg/m2 of body
 surface area, or parts per million in the food, drinking water, or
 inhaled air. Dose, or target dose, is often used to refer to the
 quantity of the agent that reaches an affected organ of interest.

 The functional relationship between the amount of a substance at
the affected organ and the lethality, morbidity, or level of health
 effect produced.
                                         58

-------
Term

Effects Assess-
ment


Emission


Emission estima-
tion techniques
Emission factor
                   Definition

                   Identification of the health effects that are likely to occur once
                   humans (or ecosystems) are exposed to HAPs (or other pollut-
                   ants).

                   The releasing of pollutant(s) to the atmosphere by  a source or
                   source category.

                   A method of estimating pollutant emissions from a particular
                   source or category of sources. Such methods include the use of
                   emission factors and activity data for the source  or source catego-
                   ry, as well as statistical approaches using surrogate data (e.g.,
                   census information) to estimate emissions hi a specific geographic
                   area.

                   An emission factor is a measure of the quantity of HAP that is
                   emitted per unit quantity of a source activity  (for example,
                   pounds of HAPs per barrel of crude  oil processed). Ideally, the
                   source "activity" will represent the operations that lead to emis-
                   sions (for example, how many barrels of crude oil are processed
                   in a day). The product of the emission factor and the  source
                   activity is used to estimate the mass of HAP emitted.  An emis-
                   sion factor is an average value which relates the quantity of a
                   pollutant released to the atmosphere by a source (e.g., chemical
                   process, fuel combustion) to the activity associated with release
                   of that pollutant. It is usually expressed as the weight of pollutant
                   per unit weight, volume, distance, or duration of the activity that
                   emits the pollutant (e.g., kg of paniculate matter per  Mg of coal
                   burned).  To estimate emissions of a pollutant from a source, the
                   emission factor for that source/pollutant is typically multiplied by
                   the corresponding source activity level.

                   A commercial,  individual/residential, industrial,  or institutional
                   activity or process that releases pollutants to the atmosphere.
                   These can be stationary (at a fixed geographic location) or mobile
                   (e.g.,  automobiles).

Emission Sources  One of the components of the Environmental Health Paradigm: it
                   includes evaluation of the pollutants emitted by the sources,
                   including identification of the chemical emitted,  the amount
                   emitted, and the location of the source and the emission points
                   and their characteristics.
Emission source
                                          59

-------
 Term
Definition
 Environmental
 Concentrations
 Environmental
 fate


 Environmental
 Health Paradigm
 Epidemiology
Estimation
Exposure
Exposure Assess-
ment
Extrapolation
(e.g., across-
exposure-scenar-
io, animal-to-
human, high-to-
low dose)
A component of the Environmental Health Paradigm:  it includes
evaluation of the concentrations of the pollutants in all environ-
mental compartments  and media, as appropriate, including indoor
and outdoor air, water, soil, and food.
The disposition of substance hi the environment, including a
description of the distribution between various media (air, water,
soil).
A conceptual framework with which  to organize and relate all of
the aspects or considerations needed to characterize how pollut-
ants from a source reach a human (or other receptor) and cause
an effect. Understanding the linkages between the components of
the paradigm also helps with evaluation of environmental man-
agement options. The paradigm includes evaluation of Emission
Sources, Environmental Concentrations, Human Exposures,
Internal Dose, and Health Effect(s).
The study of the causes of diseases by identifying personal and
environmental characteristics common to those contracting the
disease. The  sum of the factors controlling the presence or
absence of a  disease or pathogen.

The assignment or derivation of outcome values and/or probabili-
ty measures to a postulated event; a rough or approximate calcu-
lation; a numerical value obtained from a statistical sample and
assigned to a population parameter.

The coming into contact of humans (or ecosystems) with pollut-
ants; exposure is measured  as the product of concentration of the
pollutant and the time of the exposure.

Evaluation of how people are likely to come into contact with
HAPs (or other pollutants)  and the determination of how large
the exposure  is likely  to be; the measurement or estimation of the
magnitude, frequency, duration, and  route of exposure to a haz-
ardous substance or situation, and the size,  nature, and classes of
the exposed population.

To project, extend, or expand known or observed data to an area
not known or observed. In  the context of hazardous air pollut-
ants, extrapolation is used to predict  the following: responses in
humans from animal data; low-dose responses from high-dose
responses; and responses from one specific hazardous air pollut-
ant exposure  scenario to another different exposure scenario.
                                         60

-------
 Term
Definition
 Factor Informa-
 tion Retrieval
 (FIRE)

 Genotoxic
 Great Waters
 Program


 Hazard
 Hazardous air
 pollutant
Health Effect(s)
Human Carcino-
gen


Human Expo-
sures
Incidence
 An EPA supported and published data base of information on
 emission factors of various sources.


 Possessing the ability to produce harmful effects in the genetic
 makeup of an organism.

 A research and assessment program being conducted by EPA hi
 response to Section 112(m) of the Clean Air Act, entitled "Atmo-
 spheric Deposition to Great Lakes and Coastal Waters."

 A source of risk (danger,  peril, threat) that does not necessarily
 imply potential for occurrence. A hazard produces risk only  if an
 exposure pathway exists and if exposures create the possibility of
 adverse consequences.

 An airborne substance whose effect on man or animals is poten-
 tially large but undefined since an exposure pathway may or may
 not exist; the  189  chemicals,  or groups of chemicals, in the
 initial list of hazardous air pollutants found in Section 112(b) of
 the Clean Air Act; an air toxic.

 One of the components of the Environmental Health Paradigm: it
 includes characterization of the potential health effects due to
 exposure, including cancer effects, noncancer effects, any ob-
 servable damage of disease or symptoms of adverse  effects.

 A classification given to a chemical when there is sufficient
 evidence from epidemiologic  studies to support a causal associa-
 tion between exposure to the  agents and cancer.

 One of the components of the Environmental Health Paradigm: it
 involves evaluation of the route, magnitude, duration and fre-
 quency  of exposure; the interaction of humans with a pollutant or
 other physical parameter. Exposure is measured as the product of
 concentration and time.

 The number of new cases of a disease, usually expressed as  a
 rate; typically, the number of new cases of a disease occurring in
 a population during a specified period of time divided by the
 number of persons exposed to risk of developing the disease
during that period of time. The incidence rate is a direct estimate
of the probability of developing a disease during a specified
period of time.
                                         61

-------
 Term

 Internal dose
 LD50
 (lethal dose for a
 50% death rate)

 Locating and
 Estimating Re-
 ports

 Major source
 of hazardous air
 pollutants
 Margin of safety
Mechanism(s) of
action
Mechanistic data

Median
Metabolism
Mobile sources
 Definition
 One of the components of the Environmental Health Paradigm: it
 involves identification of the quantity (dose) of pollutant that is
 absorbed (the absorbed dose), the quantity that reaches the affect-
 ed organ where it may have an effect (the target dose), and
 biological indicators (biomarkers) of exposure and effects.

 A calculated dose of a substance that is expected to cause the
 death of 50% of an entire defined experimental population within
 a specified length of time.
 A series of documents issued by EPA to compile available infor-
 mation on sources and emissions of substances which may  be
 toxic at certain concentrations in the ambient air.

 A stationary  source or group of stationary sources  located within
 a contiguous area and under common control that annually releas-
 es to the atmosphere, or has the potential to release considering
 controls, 10 tons or  more of a single hazardous ah- pollutant
 listed hi the Clean Air Act or 25 tons or more of a mixture of
 these pollutants.

 A factor added to an estimated risk level for purposes of increas-
 ing the probability that a standard based on the resultant level
 will provide increased protection to the general population  and
 individual members  from harmful effects of a given substance.

 The underlying cause of disorder or disease; the specific physi-
 cal,  chemical, and/or biological events caused by HAP exposure
 that are necessary for development of the resulting symptoms,
 disorder, or disease.

 Data describing or pertaining to mechanisms of action.

 The value in an ordered set of values (that is, ambient concentra-
 tion measurements arranged from lowest to highest) hi the  mid-
 dle,  with the number of values (measurements) that are larger
 than the median being equal to the number of values (measure-
ments) that are smaller than the median.

The  sum of the physical and chemical process hi an organism by
which its material substance is produced, maintained, and de-
stroyed, and by which energy is made available.

Sources of emissions that can move, like automobiles, trucks,
planes,  boats, and trams.
                                         62

-------
 Term
                   Definition
 Model
 Molecular dosim-
 etry

 Monitoring
 (of pollutants)

 Mutagen


 Mutagenic

 Mutagenicity
 Mutagenic prod-
 ucts
 Mutation
                   A simplified representation of a system or phenomenon, as in the
                   sciences or economics, with any hypotheses required to describe
                   the system or explain the phenomenon, often mathematically; a
                   system of postulates, data, and inferences presented as a mathe-
                   matical description of an entity or state of affairs; a represen-
                   tation of reality; a description or analogy used to help visualize
                   something (e.g., air pollution patterns across  a city) that can not
                   be directly observed.

                   Characterization of the quantity of a chemical reaching an affect-
                   ed organ at a molecular level.

                   Periodic or continuous sampling  and analysis  to determine the
                   level of pollution or other characteristics.

                   A substance possessing the ability to induce heritable mutations
                   in living organisms.

                   Having the characteristic of being a mutagen.

                   The quality of being mutagenic. Mutagenicity is often measured
                   using short-term bioassays in which changes to the genetic code
                   of bacteria are identified.

                   Products of atmospheric transformation that are mutagenic.
                   A departure from being like the parent in one or more heritable
                   characteristics, due to a change in a gene or chromosome.
                   The degree to which a substance is toxic to nerve tissues; one of
                   the noncancer health effects of concern in development of the
                   Area Source National Strategy.
                   A health effect other than the development of cancer.  Section
                   112(k) of the Clean Air Act lists a number of noncancer health
                   effects to be considered under the Area Source program, includ-
                   ing "mutagenicity, teratogenicity, neurotoxicity,  reproductive
                   dysfunction and other acute and chronic effects including the role
                   of such pollutants as precursors of ozone or  acid aerosol forma-
                   tion."
Noncancer risks    The risk of developing a noncancer health effect.
Neurotoxicity
Noncancer health
effect
Oral exposure
data
                   Data on health effects developed from animal tests in which the
                   exposure to the pollutant is through ingestion.
                                          63

-------
 Term
                  Definition
 Particulate matter  Solid or liquid particles suspended in the atmosphere; a form of
                   pollution for which maximum allowable concentrations m the air
                   have been established through legislation and regulation.

                   Models that describe the fate of pharmacological substances hi
                   the body, including absorption, distribution, metabolism, and
                   elimination; dose-response models based on the principle that
                   biological effects are the result of biochemical  interaction be-
                   tween foreign substances or their metabolites and parts of the
                   body.
Pharmacokinetic
models
 Photochemical
 process
 Point source


 Possible human
 carcinogen


 Potency


 Probable human
 carcinogen


Products of
incomplete com-
bustion (PIC)
Pulmonary ef-
fects, acute and
chronic
                  Chemical reactions initiated by the absorption of light. Formation
                  of ozone and other manifestations of "smog" are the result of a
                  long series of atmospheric reactions that are started by the ab-
                  sorption of light by chemicals in the air and the resultant produc-
                  tion of highly reactive molecular fragments.

                  A stationary source of pollutants, where the location of the
                  source and its emissions of pollutants can be specified.

                  A classification given to a chemical when there is limited evi-
                  dence of carcinogenicity in animals in the absence of human
                  data.

                  The efficacy, effectiveness, or strength of a chemical to cause a
                  toxicologic response.

                  A classification given to a chemical when there is limited evi-
                  dence of human carcinogenicity based on epidemiologic studies
                  or sufficient evidence of carcinogenicity based on annual studies.

                  All of the products other than water and carbon dioxide that are
                  produced when an organic fuel, like gasoline, fuel oil, or wood,
                  is burned;  commonly, PIC is used  to refer to a complex mixture
                  of non-volatile and semi-volatile organic chemicals, many of
                  which are polycyclic  organic compounds, associated with panicu-
                  late emissions that occur whenever a fuel is burned incompletely.

                  Adverse health effects involving the lungs and due to short-term
                  exposures to high concentrations of pollutants (acute) or to long-
                  term exposures to lower concentrations of pollutants (chronic).
                  One of the noncancer health effects listed in Section 112(k) of the
                  Clean Air Act.
                                          64

-------
 Term

 Quantification



 Range of values


 Reference con-
 centrations
 (RfCs)

 Reliability


 Revertants
 Risk


 Risk, absolute
Risk, compara-
tive

Risk assessment
Risk assessment
method

Risk estimation
Superfund
Amendments and
Reauthorization
Act (SARA) Title
m
 Definition

 The assignment of a number to an entity; a method for determin-
 ing a number to be assigned to an entity; the act of determining,
 indicating, or expressing the quantity of an item.

 Evaluation of an uncertain outcome by estimation of maximal and
 minimal values.

 An estimate, with uncertainty  spanning a factor of 10, of the con-
 centration that could be inhaled for a lifetime with no adverse
 health effects.

 The probability  that a system will perform its required functions
 under conditions for a specified operating time.

 A measure of mutagenicity in  a short-term bioassay using bacte-
 ria.  Specifically, mutant strains of bacteria are exposed to pollut-
 ants, and only those bacteria that mutate back, or "revert," to
 their original genetic coding are able to  survive and produce
 colonies.

 The probability  of uncertain, undesirable consequences or out-
 comes; having a chance of injury or loss.

 A quantifiable estimate of a risk, based upon measurable and
 observable data  or statistics, without major assumptions or upper-
 limit estimates.

 An evaluation of the ranking of risks from a variety of causes in
 relationship to each other.

 The process of quantifying the level of risk associated with some
 situation or action.

 A systematic procedure  or mode of inquiry that may be employed
 as part of a risk assessment.

 The process of characterizing uncertainty (i.e., quantification of
probabilities) and consequence values for risk.

Title m of the Superfund Amendments and Reauthorization Act
of 1986, also known as  the "Emergency Planning and Com-
munity Right-to-Know Act of  1986," (EPCRA) which requires a
periodic (annual) inventory of toxic chemicals used, manufac-
tured, or processed in quantities above specified threshold
amounts at facilities in the U.S.  [See Toxic Release Inventory.]
                                         65

-------
 Term
 Definition
 Screening (stud-
 ies or hazards)


 Secondary prod-
 ucts
 Short-term test
 Smog
 (photochemical
 smog)

 Smog chamber
Source category



Stationary source


Target dose
Technology

Teratogenicity
 A preliminary process of hazard identification whereby a stan-
 dardized procedure is applied to classify products, processes,
 phenomena,  or persons with respect to their hazard potential.

 The products that are produced from the first, or primary, prod-
 ucts; specifically, photochemical reactions in polluted air produce
 primary products which themselves react further to produce the
 secondary products, many of which are the more stable products
 normally  associated with smog.

 Tests that take less time to complete than do other types of
 bioassays. Many short-term tests measure the biological interac-
 tions between the agent under test and deoxyribonucleic acid
 (DNA). Agents  that have effects hi short-term tests are generally
 considered more likely to be health hazards than those that have
 no effect.

 Air pollution containing ozone and other reactive compounds
 formed by the action of sunlight on nitrogen oxides and hydro-
 carbons (or other organic precursors).

 An experimental apparatus used to simulate the production of
 photochemical "smog"; often a large, Teflon-lined enclosure or
 bag, surrounded by lights that represent the sun's radiation or
 open to natural sunlight, with connections for inserting and
 withdrawing samples of pollutants.

 A grouping of individual sources for consideration together
 because of similarities in emissions, manufacturing processes, or
 other factors.

 A source  of pollutants hi a fixed position, the location of which
 can be specified.

 The amount of HAP that directly impinges on tissues or organs
 and induces a significant or toxic effect. Part of the consider-
 ations under  the Internal Dose component of the Environmental
 Health Paradigm.

 The tangible products of the application of scientific knowledge.

Production or induction of malformations or monstrosities,
especially of a developing embryo or fetus. One of the noncancer
health effects listed in Section 112(k) of the Clean Air Act.
                                         66

-------
 Term

 Threshold
 Toxicity


 Toxic Release
 Inventory (TRI)
Toxic substance


Uncertainty



Urban areas

Volatile organic
compounds
 Definition

 A discontinuous change of state of a parameter as its measure
 increases. One condition exists below the discontinuity, and a
 different one above it. In context of toxicity and this report,
 exposures above a threshold produce effects, whereas exposures
 below threshold do not produce effects.

 Inherent ability of a substance to adversely affect living organ-
 isms.

 The TRI is an  inventory  of releases to air, water, and soil,  or
 transfers to treatment facilities of 322 toxic chemicals. The  TRI
 was mandated by the Emergency Planning and Community  Right-
 to-Know Act (EPCRA) of  1986 (also known as Title HI of  the
 Superfund Amendment and Reauthorization Act). Manufacturing
 facilities that produce, import, or process 50,000 pounds or more
 per year, or facilities that use  10,000 pounds or more per year of
 the 322 chemicals specified in the EPCRA must report their
 emissions annually to EPA. [See SARA Title IE.]

 A substance of which exposure to humans or animals results in
 deleterious effects.

 A situation where there are a number of possible outcomes  and
 one does not know which of them has occurred or will occur;
 indeterminacy; unpredictability; indefiniteness.

 Areas in a city or town; areas that are city-like.

 Two major definitions are common: (1) Under the regulatory
 control program to limit production of ozone pollution, VOCs are
 organic chemicals, usually  hydrocarbons, that produce ozone at a
rate greater than ethane; (2) In a scientific sense, VOCs are
chemicals containing carbon that evaporate so readily that they
exist in the air  as vapors.
                                         67

-------