External Review Draft No. 2
praff February 1981
Do Not Quote or Cite
Air Quality Criteria
for Particulate Matter
and Sulfur Oxides
r
Volume I
NOTICE
This document is a preliminary draft. It has not been formally released by EPA and should not at this stage be
construed to represent Agency policy. It is being circulated for comment on its technical accuracy and
policy implications.
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
-------
NOTE TO READER
The Environmental Protection Agency is revising the existing criteria
documents for particulate matter and sulfur oxides (PM/SO ) under Sections 108
and 109 of the Clean Air Act, 42 U.S.C. §§ 7408, 7409. The first external
review draft of a revised combined PM/SO criteria document was made available
/\
for public comment in April 1980.
The Environmental Criteria and Assessment Office (ECAO) filled more than
4,000 public requests for copies of the first external review draft. Because
all those who received copies of the first draft from ECAO are being sent copies
of the second external review draft, there is no need to resubmit a request.
To facilitate public review, the second external review draft has been
released in five volumes on a staggered schedule as the volumes were completed.
With circulation of this volume (Volume I), the release of all five volumes
of this second external review draft is completed.
The first external review draft was announced in the Federal Register of
April 11, 1980 (45 FR 24913). ECAO received and reviewed 89 comments from the
public, many of which were quite extensive. The Clean Air Scientific Advisory
Committee (CASAC) of the Science Advisory Board also provided advice and
comments on the first external review draft at a public meeting of August 20-22,
1980 (45 FR 51644, August 4, 1980).
As with the first external review draft, the second external review draft
will be submitted to CASAC for its advice and comments. ECAO is also soliciting
written comments from the public on this second external review draft and
requests that an original and three copies of all comments be submitted to:
Project Officer for PM/SOX, Environmental Criteria and Assessment Office, MD-52,
U.S. Environmental Protection Agency, Research Triangle Park, N.C. 27711. To
facilitate ECAO's consideration of comments on this lengthy and complex docu-
ment, commenters with extensive comments should index the major points which
they intend ECAO to address, by providing a list of the major points and a
cross-reference to the pages in the document. Comments should be submitted
during the public comment period from March 6 to May 5, 1981, as announced in
the Federal Register. There will be no extensions of this comment period, for
the reasons set forth in the Federal Register, which primarily concerns the
statutory deadline for completing appropriate revisions to the criteria for
PM/SOX.
XRD1B/A 3-3-81
-------
External Review Draft No. 2
Draft February 1981
Do Not Quote or Cite
Air Quality Criteria
for Particulate Matter
and Sulfur Oxides
Volume I
NOTICE
This document is a preliminary draft. It has not been formally released by EPA and should not at this stage be
construed to represent Agency policy. It is being circulated for comment on its technical accuracy and
policy implications.
Environmental Criteria and Assessment Office
Office of Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
-------
PREFACE
This document is a revision of External Review Draft No. 1, Air
Quality Criteria for Particulate Matter and Sulfur Oxides, released in
April 1980. Comments received during a public comment period from April
15, 1980 through July 31, 1980, and recommendations made by the Clean Air
Scientific Advisory Committee in August have been addressed here.
Volume I contains Chapter 1 which is the Executive Summary. A Table
of Contents for Volumes I, II, III, IV, and V follows.
n
-------
CONTENTS
VOLUMES I, II, III, IV, AND V
Page
Volume I.
Chapter 1. Executive Summary 1-1
Volume II.
Chapter 2. Physical and Chemical Properties of Sulfur
Oxides and Particulate Matter 2-1
Chapter 3. Techniques for the Collection and Analysis of
Sulfur Oxides, Particulate Matter, and Acidic
Precipitation 3-1
Chapter 4. Sources and Emissions 4-1
Chapter 5. Environmental Concentrations and Exposure 5-1
Volume III.
Chapter 6. Atmospheric Transport, Transformation and
Deposition 6-1
Chapter 7. Acidic Deposition 7-1
Chapter 8. Effects on Vegetation 8-1
Volume IV.
Chapters. Effects on Visibility and Climate 9-1
Chapter 10. Effects on Materials 10-1
Volume V.
Chapter 11. Respiratory Deposition and Biological Fate
of Inhaled Aerosols and S02 11-1
Chapter 12. Toxicological Studies 12-1
Chapter 13. Controlled Human Studies 13-1
Chapter 14. Epidemiological Studies of the Effects of
Atmospheric Concentrations of Sulfur Dioxide
and Participate Matter on Human Health 14-1
111
-------
TABLE OF CONTENTS
LIST OF FIGURES.
LIST OF TABLES..
1. EXECUTIVE SUMMARY 1-1
1.1 INTRODUCTION 1-1
1.1.1 Legal Requirements 1-1
1.1.2 Organization of the Document 1-2
1.2 PHYSICAL AND CHEMICAL PROPERTIES OF SULFUR OXIDES AND
PARTICULATE MATTER 1-4
1.2.1 Sulfur Oxides 1-4
1.2.2 Particulate Matter 1-4
1.3 TECHNIQUES FOR COLLECTION AND ANALYSIS 1-7
1.3.1 Sulfur Oxides 1-7
1.3.2 Parti cul ate Matter 1-8
1.4 SOURCES AND EMISSIONS 1-11
1.4.1 Sulfur Oxides 1-12
1.4.2 Particulate Matter 1-13
1.5 CONCENTRATIONS AND EXPOSURE 1-13
1.5.1 Sulfur Oxides 1-13
1.5.2 Particulate Matter 1-13
1.6 ATMOSPHERIC TRANSPORT, TRANSFORMATION, AND DEPOSITION 1-19
1.7 ACIDIC DEPOSITION 1-21
1.8 EFFECTS ON VEGETATION. 1-28
1.9 EFFECTS ON VISIBILITY AND CLIMATE 1-36
1.10 EFFECTS ON MATERIALS 1-43
1.11 RESPIRATORY TRACT DEPOSITION AND FATE OF SULFUR OXIDES AND
PARTICULATE MATTER 1-51
1.11.1 Exposures 1-51
1.11.2 Deposition and Clearance 1-51
1.12 TOXICOLOGICAL STUDIES 1-55
1.12.1 Metabolism of Sulfur Oxides and Particulate Matter... 1-55
1.12.2 Effects of S05 1-55
1.12.3 Effects of Paniculate Matter 1-58
1.12.4 Effects of Complex Mixtures 1-64
1.13 CONTROLLED HUMAN STUDIES 1-64
1.13.1 Effects of Sulfur Dioxide 1-68
1.13.2 Effects of SO, in Combination with Particulate
Matter r 1-71
1.13.3 Effects of Sulfur Dioxide-Ozone Exposure 1-72
1.13.4 Effects of Sulfate Aerosols 1-72
1.14 EPIDEMIOLOGICAL STUDIES. 1-73
1.14.2 Health Effects Associated with Acute Exposures to
Sulfur Oxides and Particulate Matter 1-73
iv
XRD1B/A 3-3-81
-------
LIST OF FIGURES
Figure Page
1-1 Size distributions of atmospheric particulate volume under a
variety of conditions/source locations 1-5
1-2 Idealized representation of typical fine and coarse particle mass
and chemical composition distribution in an urban aerosol 1-7
1-3 Sulfur dioxide second maximum 24-hour average by county, 1974-1976. 1-14
1-4 One example of rapid increases in ambient sulfur dioxide concen-
tration from near zero to 1.30 ppm (3410 ug/m ) during a period of
approximately two hours is shown 1-15
1-5 Total suspended particulate maximum annual average by county,
1974-1976 1-16
1-6 Seasonal variations in urban, suburban, and rural areas are shown
for four size ranges of particles 1-17
1-7 Complex processes affecting transport and transformation of air-
borne particulate matter and sulfur oxides 1-20
1-8 pH of rain sample as measured in the laboratory used in combination
with the reported amount of precipitation 1-23
1-9 Idealized conceptual framework illustrating the "law of tolerance,"
which postulates a limited tolerance range for various environmental
factors within which species can survive 1-25
1-10 Exposure thresholds for minimum, maximum and average sensitivity of
33 plant species to visible foliar injury by SQ« 1-32
1-11 Percentage of plant species visibly injured as a function of peak,
1-hr, and 3-hr S02 concentrations 1-33
1-12 Map shows median yearly visual range (miles) and isopleths for
suburban/nonurban areas, 1974-76 1-37
1-13 Inverse proportionality between visual range and the scattering
coefficient, b t, was measured at the point of observation 1-39
1-14 Simultaneous monitoring of b . and fine-particle mass in
St. Louis in April 1973 showlS a high correlation coefficient of
0.96, indicating that b . depends primarily on the fine-particle
concentration 1-40
1-15 The spatial distribution of 5-year average extinction coefficients
shows the substantial increases of third-quarter extinction
coefficients in the Carolinas, Ohio River Valley, and Tennessee-
Kentucky area 1-41
1-16 Seasonal turbidity patterns for 1961-66 and 1972-75 are shown for
selected regions in the Eastern United States 1-42
1-17 Steel corrosion behavior is shown as a function of average S02
concentration at 65% relative humidity 1-46
1-18 Steel corrosion behavior is shown as a function of average relative
humidity at three average concentration levels of sulfur dioxide... 1-47
1-19 Annual mean relative humidity (RH) in various U.S. areas 1-48
1-20 Division of the thoracic fraction into the pulmonary and tracheo-
bronchial fractions for two sampling conventions (ACGIH and BMRC)
as a function of aerodynamic diameter except below 0.5 pm where
particle deposition is plotted vs. physical diameter, from
International Standard Organization ad hoc group to TC-146, 1980... 1-53
XRD1B/A 3-3-81
-------
LIST OF TABLES
Table Page
1-1 (a) National estimates of participate emissions (10 metric tons
per year) 6 1-12
1-1 (b) National estimates of sulfur oxide emissions (10 metric tons
per year) 1-12
1-2 Sulfur dioxide concentrations causing visible injury to various
sensitivity groupings of vegetation (ppm SO-) 1-30
1-3 Selected physical damage functions related to S02 exposure 1-45
1-4 Results of regression for soiling of building materials as a
function of TSP dose 1-50
1-5 Effects of acute exposures to sulfur dioxide on pulmonary
f uncti on 1-56
1-6 Responses to acute sulfuric acid exposure 1-59
1-7 Responses to chronic sulfuric acid exposure 1-60
1-8 Responses to various particulate matter mixtures 1-61
1-9 Responses to acute exposure combinations of SO- and some
particulate matter 1-65
1-10 Responses to acute exposure combinations of sulfuric acid and
ozone 1-66
1-11 Pathological responses following chronic exposure to S0? alone
and in combination with particulate matter 1-67
1-12 Controlled human exposure studies - major studies cited in
Chapter 13 1-69
1-13 Summary of quantitative conclusions from epidemiological studies
relating health effects of acute exposure to S0? and particulate
matter to ambient air levels 1--74
1-14 Summary of quantitative conclusions from epidemiological studies
relating health effects of chronic exposure to S0? and particu-
late matter to ambient air levels 1-77
VI
XRD1B/A 3-3-81
-------
1. EXECUTIVE SUMMARY
1.1 INTRODUCTION
1.1.1 Legal Requirements
The purpose of this document is to present air quality criteria for particulate matter ar
-------
and averaged over appropriate time periods, and the biological responses to those levels and
interactions with other variable factors. This assessment must take into consideration the
temporal and spatial distribution of PM and SO , and is complicated by such factors as
J\
breathing patterns, individual activity levels, effects on sensitive populations, and the
complex and diverse chemical composition of PM.
The welfare effects which must be identified in the criteria document include effects on
soils, water, crops, vegetation, man-made materials, animals, wildlife, weather, visibility,
climate, damage to and deterioration of property, and hazards to transportation, as well as
effects on economic values and personal comfort and well-being [Clean Air Act Section 302(h),
42 U.S.C. §7602(h)]. Under Section 109(b) of the Clean Air Act, the Administrator must
consider such information in this document and set national secondary ambient air quality
standards which are based on the criteria and are requisite to protect the public welfare from
any known or anticipated adverse effects associated with the presence of such pollutants.
1.1.2 Organization of the Document
The present document consists of 14 chapters, currently organized into five separate
volumes as follows:
Volume I (containing Chapter 1);
Volume II (containing Chapters 2, 3, 4, and 5);
Volume III (containing Chapters 6, 7, and 8);
Volume IV (containing Chapters 9 and 10); and
Volume V (containing Chapters 11, 12, 13, and 14).
Volume I, the present volume, contains the general introduction and the executive summary
and conclusions for the entire document. Chapters 2 through 5, contained in volume II,
provide background information on: physical and chemical properties of PM and SO ; approaches
for the collection and measurement of such air pollutants; their sources of emissions; their
ambient air concentrations; and factors affecting exposures of the general population to them.
The third volume, containing Chapters 6 through 8, provides information on: atmospheric
transport, transformation and fate of PM and SO ; their contribution to and involvement in
acidic deposition processes and effects; and their effects on vegetation. Volume IV contains
Chapters 9 and 10, which describe effects on visibility and damage to materials presently
attributable to either PM or SO .
The fifth volume, containing Chapters 11 through 14, focuses on information concerning
the health effects of PM and SO . Chapter 11 discusses respiratory tract uptake and deposition
X
of sulfur dioxide (SO-), sulfur-related particulate matter species (e.g., sulfates and sulfuric
acid), and other types of PM, as well as factors affecting their deposition and biological
fate. Chapters 12 and 13, on the other hand, respectively discuss information derived from
controlled toxicological studies in animals and controlled human exposure studies, whereas
Chapter 14 discusses community health epidemiological studies.
The extensive literature on PM and SO is critically reviewed and evaluated in the present
document with emphasis on discussion of studies selected on the basis of their validity and
"XRD1B/D 1-2 3-3-81
-------
relevance for the assessment of human health and welfare effects. Consideration of air quality
information and measurement techniques in early chapters of the document is limited to eluci-
dating aspects of environmental pathways by which PM and SO move from natural and manmade
sources of emissions to receptor sites (i.e., biological organisms or materials affected by
them) most germane to understanding relationships between atmospheric levels of PM and SO and
health and welfare effects discussed later in the document. As indicated by the discussion of
air quality information, airborne particles of a wide variety of sizes, shapes, and chemical
composition are found in the ambient air of the United States in highly variable quantities
and combinations in different geographic locations and at various times at the same site.
Analysis of the effects of airborne particles is further complicated by complex transformations
of various particulate matter species or their precursor substances during atmospheric trans-
port from sources of emissions that may be hundreds or thousands of miles away from human popu-
lations, other biological organisms, or materials ultimately affected by the pollutants. Sul-
fur dioxide (SO-), a gaseous air pollutant exerting notable health and welfare effects in its
own right, is also the main precursor substance emitted from anthropogenic sources which con-
tributes to the secondary formation of sulfuric acid'and sulfate salts representing major
constituents of PM present in urban aerosols to which large segments of the U.S. population
are exposed. SO^ and sulfur-related PM species and their associated health and welfare effects
are accordingly discussed in considerable detail in the present document. Other specific PM
species of concern, however, are not discussed in as much detail here; but, rather the reader
is referred to other EPA air quality criteria or health assessment documents where the effects
of such substances are thoroughly reviewed. See, for example, Air Quality Criteria for Lead
(EPA, 1977), Air Quality Criteria for Oxides of Nitrogen (EPA, 1981), and so on.
In evaluating available information on the health effects of PM and S0x in man, the main
focus is on the effects of inhalation of these substances as the most direct and important
route of exposure, although it is recognized that biological effects may also be exerted by
some PM species via other routes of exposure, such as ingestion or contact with skin. Impor-
tant issues considered include: (1) patterns of respiratory tract uptake, deposition, and
biological fate of S02, sulfur-related PM, and other PM substances in relation to size and
other physical and chemical properties; (2) mechanisms of action by which such substances may
exert biological effects of potential concern from a health viewpoint; (3) qualitative charac-
terization of such biological effects and quantitative characterization of dose/response or
exposure/effect relationships for such effects in relation to air concentrations of S02 and PM
(defined either in general or, whenever possible, in terms of varying size or chemical composi-
tion); and (4) identification of population subgroups at special risk for the induction of
effects by PM and SO .
In relation to the evaluation of welfare effects of PM and SO , consideration is not only
^\
accorded to direct, acute effects of such substances on visibility, manmade materials, and
plant and animal species. Rather, as appropriate, there are assessed more-indirect, long-term
*XRD1B/D 1-3 3-3-81
-------
effects that might be reasonably anticipated to occur as a consequence of repeated or contin-
uous chronic exposures to low levels of such pollutants, their interactions with other agents
such as meteorological variables, and their secondary deposition on and movement through aqua-
tic and terrestial ecosystems.
In the summary materials that follow, citations are made only to selected, crucial
references that provide evidence in support of key points addressed. Complete bibliographic
information for each citation is provided in the reference list for the later chapter where
the particular point is discussed in more detail.
1.2 PHYSICAL AND CHEMICAL PROPERTIES OF SULFUR OXIDES AND PARTICULATE MATTER
1.2.1 Sulfur Oxides
Of four known sulfur oxides (sulfur monoxide, sulfur dioxide, sulfur trioxide, and disul-
fur monoxide), only sulfur dioxide (SO-) occurs at significant concentrations in the atmosphere.
SO. is a colorless gas with a pungent odor emitted from combustion of sulfur-containing fossil
fuels, such as coal and oil. Its physical and chemical properties are summarized in Chapter 2.
Oxidation of SO. to form sulfate particles is one mechanism by which this gas is removed
from the atmosphere; SO. is also removed directly by dry deposition on surfaces. Oxidation of
SO. to produce sulfate particles takes place by a variety of mechanisms, such as photochemical
reactions and catalysis by constituents of ambient aerosols. SO. interactions with airborne
particles have been studied for a variety of solids such as ferric oxide, lead oxide, aluminum
oxide, salt, and charcoal. One research group concluded that the main reaction between SO.
and particulate matter is adsorption, with most catalytic reactions occurring at high tempera-
tures near the combustion source.
SO. is readily soluble in water, forming a dilute solution of sulfurous acid including
the following species: S02 • H20; HSO^ (bisulfite); and SO^ (sulfite).
1.2.2 Particulate Matter
Airborne particles exist in diverse sizes and compositions that can fluctuate widely under
the changing influences of source contributions and meteorological conditions. In broad terms,
however, airborne particle mass tends to cluster in two principal size groups: coarse parti-
cles mostly larger than 2-3 micrometers (urn) in diameter, and fine particles mostly smaller
than 2-3 urn in diameter. The dividing line between the coarse and the fine sizes is frequent-
ly given as 2.5 um but the separation is neither sharp nor fixed; it can depend on the contri-
buting sources, on meteorology, and on the age of the aerosol. The curves in Figure 1-1
exemplify the influences of these parameters.
Fine particles can occur in two mass modes. Those in the nuclei mode—from 0.005 to 0.05
um diameter—form near sources by condensation of vapors produced by high temperature processes
such as fossil-fuel combustion. Particles in the accumulation mode—from 0.051 to( about
2 urn—form principally by coagulation or growth through vapor condensation of the short-lived
nuclei mode particles. Typically, 80% or more of atmospheric sulfate particle mass occurs in
*XRD1B/D 1-4 3-3-81
-------
io3
*£ IO2
I
\
I
5
io
10°
io-2
10-*
10s
PHYSICAL ROLES- SCATTERING OF
ATMOSPHERIC SOLAR RADIATION
ELECTRICITY CLOUD MICBOPHYSICS
PR IM ARY SOU RCES: FLY ASH
COMBUSTION SEA SPRAY
PROCESSES POLLEN AND SPORES
SECONDARY SOURCES:
GAS-TO-PARTICLE PHOTOCHEMICAL
CONVERSION REACTIONS S"1^
— FOREST FIRE \ ;
CLOUD DROP EVAPORATION /
AND PARTICLE COAGULATION I
SERIOUS "SMOG"
EPISODE /*
~ "CLEAN-
COMBUSTION
SOIL
EROSION
HEAVV INDUSTRY
MECHANICAL
FRACTURING
INCREASING WIND
DECREASING HEIGHT
INCREASING DISTANCE
FROM SOURCE
\
INCREASING
INSOLATION
Iff3
10'1
10°
10'
IO2 IO3
PARTICLE DIAMETER ID), fan
SINKS:
COAGULATION
PRECIPITATION SCAVENGING GRAVITATIONAL
DRY DEPOSITION SETTLING
AND EVAPORATION
COMMON NAMES:
AITKEN NUCLEI »R6E
CONDENSATION NUCLEI PARTICLES
LARGE IONS ' FINE COARSE PARTICLES
PARTICLES
PHOTOCHEMICAL
AEROSOLS SEA SALT
NUCLEI
OUST
GIANT SAND
PARTICLES
SMOKE
SOIL
PARTICLES
Figure 1-1. Size distribution of atmospheric paniculate volume under a variety of condi-
tions/source locations. Atmospheric particle volume is roughly proportional to mass.
Particle volume tends to be bimodally distributed, although the relative amounts and
peaks of both modes can vary significantly.
Source: W.G.N. Slinn. Dry Deposition and Resuspension of Aerosol Particles-a New Look at
Some Old Problems. In: Atmosphere-Surface Exchange of Paniculate and Gaseous Pollutants
(1974) Proceedings of a Symposium. Battelle Pacific Northwest Laboratories and U.S. Atomic
Energy Commission. Richland, Washington, September 4-6.1974. ERDA Symposium Seroes
38 Energy Research and Development Administration. Oak Ridge. TN. January 1976. pp. 1-40.
1-5
-------
the accumulation mode. Accumulation mode particles normally do not grow into the coarse
mode—those larger than about 2-3 urn. Coarse particles include re-entrained surface dust and
particles formed by anthropogenic processes such as grinding.
Primary particles are directly discharged from manmade or natural sources. Secondary
particles form by chemical and physical reactions in the atmosphere, and most of the reactants
involved are emitted to the air as gaseous pollutants. l
In the atmosphere, particle growth and chemical transformation occur through gas-particle
and particle-particle interactions. Gas-particle interactions include condensation of low-
vapor-pressure molecules, such as sulfuric acid (H2S04), which occurs principally on fine parti-
cles. The only particle-particle interaction important in atmospheric processes is coagulation
among fine particles.
As shown in Figure 1-2, major components of fine atmospheric particles include sulfates,
carbonaceous material, ammonium, lead, and nitrate. Coarse particles consist mainly of oxides
of silicon, aluminum, calcium, and iron, as well as calcium carbonate, tire particles, vege-
tation-related particles and sea salt. Note that some overlap into fine versus coarse size
distributions occurs for many chemical species found predominantly in one or the other
size mode.
o
I rf
I I I I 111
I I I I I I 11
FINE
COARSE
'SULFATES, ORGANICS,
AMMONIUM, NITRATES,
CARBON, LEAD, AND
SOME TRACE CONSTITUENTS
I I
I
CRUSTAL MATERIAL \
(SILICON COMPOUNDS. X
IRON, ALUMINUM), SEA V
SALT, PLANT PARTICLES \
I I I I
0.1
1.0
PARTICLE DIAMETER,
10.0
Figure 1-2. Idealized representation of typical fine and coarse particle mass and chemical
composition distribution in an urban aerosol. Although some overlap exists, note sub-
stantial differences in chemical composition of fine versus coarse modes. Chemical species
of each mode are listed in approximate order of relative mass contribution.
Source: After Husar et al. (1978).
*XRD1B/0
1-6
3-3-81
-------
The carbonaceous component of fine particles contain both elemental carbon (graphite and
soot) and nonvolatile organic carbon (hydrocarbons emitted in combustion exhaust and secondary
organics formed by photochemistry). In many urban and nonurban areas, these particles may be
the most abundant fine particle species after sulfates. Secondary organic particles form by
oxidation of primary organics, ozone, and nitrogen oxides. Atmospheric reactions of nitrogen
oxide gases yield nitric acid vapor (HNQ-3) which may accumulate as nitrate in fine and coarse
particles. Chemical pathways for forming nitrate particles and secondary organics are not
well established, and much doubt exists regarding the validity of historical nitrate data
bases.
Most atmospheric sulfates and nitrates are water soluble and hygroscopic (absorb
moisture). Hygroscopic growth of sulfate-containing particles has a profound effect on their
size, reactivity and other physical properties, which in turn influence their deposition in
the respiratory tract, toxicity, removal efficiency, and effects on weather and climate.
Deliquescence, or sudden uptake of water when relative humidity exceeds a certain level, is
exhibited by a number of organic and inorganic compounds, primarily salts.
1.3 TECHNIQUES FOR COLLECTION AND ANALYSIS
Since publication of the 1970 criteria documents, advances in technology have resulted in
a substantial number of new measurement techniques together with information on the quality of
data collected by older techniques, as discussed in Chapter 3 of the present document. This
summary focuses on those techniques principally used in health and welfare studies.
1.3.1 Sulfur Oxides
Three main measurement methods or variations thereof have been employed in generating
data cited for sulfur dioxide (S0«) levels in community health epidemiological studies: (1)
sulfation rate (lead dioxide) methods; (2) hydrogen peroxide measurements and (3) the West-
Gaeke (pararosanaline) method.
Sulfation rate methods involve reaction of airborne sulfur compounds with lead dioxide in
a paste spread over an atmospherically-exposed plate or cylinder. Rates of reaction of sulfur
2
compounds with surface paste compounds are expressed in S03/cm /day. However, the reactions
are not specific for S02, and atmospheric concentrations of S02 or other sulfur compounds can-
not be accurately extrapolated from the results, which are markedly affected by factors such
as temperature and humidity. Lead dioxide gauges were widely used in the United Kingdom prior
to 1960 and provided aerometric data reported for S02 in some pre-1960s British epidemiological
studies; sulfation rate methods were also used in certain American studies.
Use of the hydrogen peroxide method was gradually expanded in the United Kingdom during
the 1950s, often being coupled in tandem with apparatus for particulate matter (smoke) moni-
toring. The hydrogen peroxide method was adopted in the early 1960s as the standard S02
method used in the National Survey of Air Pollution throughout the United Kingdom and, as an
OECD-recommended method, elsewhere in Europe. The method can yield reasonably accurate esti-
mates of atmospheric S02 concentrations expressed in ug/m ; but results obtained with routine
*XRD1B/D 1-7 3-3-81
-------
ambient air monitoring can be affected by factors such as temperature, presence of atmospheric
ammonia and titration errors. Very little quality assurance information exists on sources and
magnitudes of errors encountered in use of the method in obtaining S0» data reported in
specific epidemiological studies, making it difficult to assess the accuracy and precision of
reported SO- values. Only in the case of the British National Survey has extensive quality
assurance information been reported (Warren Spring Laboratory, 1961; 1962; 1966; 1967; 1975;
1977; OECD, 1964; Ellison, 1968) for S02 measurements made in the United Kingdom and used in
various British epidemiological studies.
The West-Gaeke (pararosanaline) method has been more widely employed in the United States
for measurement of S02- The method involves absorption of S02 in potassium tetrachloromercu-
rate (TCM) solution, producing a chemical complex reacted with pararosanaline to form a red-
purple color measured colorimetrically. The method, suitable for sampling up to 24 hrs, is
specific for S02 if properly implemented to minimize interference by nitrogen or metal oxides,
but results can be affected by factors such as temperature variations and mishandling of rea-
gents. Only limited quality assurance information (Congressional Investigative Report, 1976)
has been reported for some American S02 measurements by the West-Gaeke methods.
In recent years, a number of automated methods have gained widespread use for air moni-
toring. Some of these have been used in studying the effects of S02 on vegetation. Con-
tinuous analyzers based on a variety of measurement principles have been designated by EPA as
equivalent methods for measurement of SO- in the atmosphere. Testing by EPA has verified
their performance and has demonstrated excellent comparability with the federal reference
method under typical monitoring conditions.
1.3.2 Particulate Matter
Sampling particulate matter suspended in ambient air presents a complex task because of
the spectrum of particle sizes and shapes. Particle separation by aerodynamic size provides a
simplification by accounting for variations in particle shape and particle settling velocity.
Samplers can be designed to collect specific size fractions or match specific deposition pat-
terns. Mass concentration measurements using gravimetric analysis provide direct measures of
atmospheric particulate matter levels. High-volume samplers, dichotomous samplers, cascade
impactors, and cyclone samplers are the most common examples of this type of measurement.
Mass concentrations have also been estimated using methods which do not employ direct
weighing. These methods utilize techniques which measure an integral property of particles
other than mass, such as optical reflectance. Examples of commonly used indirect methods
include the American Iron and Steel Institute (AISI) tape sampler version of the ASTM method,
the integrating nephelometer, and beta attenuation analysis.
Three main measurement approaches or variations thereof were used to obtain PM data
reported in community health studies. (1) the British Smokeshade light reflectance method or
variations used in the United Kingdom and elsewhere in Europe; (2) the American Society for
*XRD1B/D 1-8 3-3-81
-------
Testing and Materials (ASTM) filter soiling method based on light transmittance and used -'n
the United States; and (3) the high-volume sampling method most widely employed in the United
States.
As discussed in Chapter 3, the British Smoke (BS) method and various standard variations
of it typically have a DSQ cut-point of s 4.5 ym in field use (McFarland, 1979). Thus, regard-
less of whether or not larger coarse-mode particles were present in the atmosphere during the
sampling period, the BS method collected predominantly small particles. The D5Q of the instru-
ment may, however, shift slightly at higher wind speeds. The BS method neither directly mea-
sures the mass nor determines chemical composition of collected particles. Rather, it primari-
ly measures light absorption of particles as indicated by reflectance from a stain formed by
the particles collected on filter paper, which is somewhat inefficient for collecting very fine
particles (Lui, 1978). The reflectance of light from the stain depends both on density of the
stain or amount of PM collected in a standard period of time and optical properties of the
collected materials. Smoke particles composed of elemental carbon of the type found in incom-
plete fossil fuel combustion products typically make the greatest contribution to the darkness
of the stain, especially in urban areas. Thus, the amounj. of elemental carbon, but not organic
carbon, present in the stain tends to be most highly correlated with BS reflectance readings.
Other non-black, non-carbon particles also have optical properties such that they can affect
the reflectance readings (Pedace and Sansone, 1972).
Since highly variable relative proportions of atmospheric carbon and non-carbon PM can
exist from site to site or from one time to another at the same site, then the same absolute
BS reflectance reading can be associated with markedly different amounts (or mass) of parti-
cles collected or, even, carbon present. Site-specific calibrations of reflectance readings
against actual mass measurements obtained by collocated gravimetric monitoring devices are
therefore necessary in order to obtain approximate estimates of atmospheric concentrations of
PM based on the BS method. A single calibration curve relating mass or atmospheric concentra-
tion (in ug/m ) of particulate matter to BS reflectance readings obtained at a given site may
serve as a basis f9r crude estimates of PM (mainly small particle) levels at that site over
time, so long as the chemical composition and relative porportions of elemental carbon and
non-carbon PM do not markedly change. Of crucial importance for evaluation of BS data is the
fact that the actual mass or smoke concentration present at a particular site may differ
markedly from the corresponding mass or concentration (in ug/m ) associated with a given
reflectance reading on either of the two most widely used standard curves; great care
must be applied in interpreting exactly what any reported BS value in |jg/m means at all.
The ASTM or AISI light transmittance method is similar in approach to the British smoke
technique. The instrument has a 05Q cut-point of =5 urn and utilizes an air flow intake
apparatus similar to that used for the BS method, depositing collected material on a filter
paper tape periodically advanced to allow accumulation of another stain over a standard time
*XRD1B/D 1-9 3-3-81
-------
period. Opacity of the stain is determined by transmittance of light through the deposited
material and filter paper, with results expressed in terms of optical density or coefficient
of haze (CoHs) units per 1000 linear feet of air sampled (rather than mass units). CoHs read-
ings, however, are somewhat more markedly affected by non-carbon particles than BS measure-
ments. The ASTM method does not directly measure mass or determine chemical composition of
the PM collected. Any attempt to relate CoHs to ug/m would require site-specific calibra-
tion of CoHs readings against mass measurements determined by a collocated gravimetric device,
but the accuracy of such mass estimates could be subject to question.
The high-volume (hi-vol) sampler method, more widely used in the United States to measure
total suspended particulates (TSP), collects particles on a glass-fiber filter by drawing air
through the filter at a flow rate of approximately 1.5 m /min, thus sampling a higher volume
of air per unit of time than the above PM sampling methods. Recent evaluations show that the
hi-vol sampler collects a smaller particle size range than that stated in Air Quality Criteria
for Particulate Matter (U.S. Department of Health Education and Welfare, 1969). Under most
conditions the particle size fraction collected (D50) ranges from 0 to 25 - 30 urn. The
sampling effectiveness of the hi-vol inlet also is wind speed sensitive for larger (>10 urn)
particles. Wind speed could be estimated to produce no more than a 10 percent day-to-day
variability for the same ambient concentration for typical conditions. The hi-vol is one of
the most reproducible particle samplers currently in use, with a typical coefficient of
variation of 3-5 percent. A significant problem associated with the glass fiber filter used
on the hi-vol is the formation of artifact mass caused by the presence of acid gases in the
air. These artifacts can add 6-7 ug/m to a 24-hour sample.
One consequence of the broader size range of particles sampled by the hi-vol method
versus the BS or ASTM methods are severe limitations on intercomparisons or conversions of PM
measurements by those methods to equivalent TSP units or vice versa. As shown by several
studies, no consistent relationship typically exists, for example, between BS and TSP measure-
ments taken at various sites or even during various seasons at the same site (Commins and
Waller, 1967; Lee, 1972; Ball and Hume, 1977; Holland et al., 1979). The one exception
appears to be that, during severe London air pollution episodes when low wind speed conditions
resulted in settling out of larger coarse-mode particles and fine-mode particles markedly
increased to constitute most of the PM present, then TSP and BS levels (in excess of = 500
Mg/m ) tended to converge as would be expected when both methods are essentially sampling
predominantly fine-mode particles (Holland et al., 1979).
An extensive list of techniques is available to analyze particles collected on a suitable
substrate. Many of the techniques are more precise than the analyses for gravimetric mass
concentration. Methods are available to provide reliable analyses for sulfates, nitrates,
organic fractions, and elemental composition (e.g., sulfur, lead, silicon, etc.). Not all
analyses can be performed on all particle samples because of factors such as incompatible sub-
*XRD1B/D 1-10 3-3-81
-------
strates and inadequate sample size. Misinterpretation of analytical results occur when
samples have not been segregated by particle size and when artifact mass is formed on the
substrate rather than collected in a particle form.
Sampling technology is available to meet specific requirements such as providing: sharp
cutpoints, cutpoints which match particle deposition models, separate collection of fine and
coarse particles, automated sample-collection capability, collection of at least milligram
quantities of particles, minimal interaction of the substrate with the collected particles,
the ability to produce particle size distribution data, low purchase cost, and simple operat-
ing procedures. Not all of these sampling requirements may be needed for a measurement study.
Currently, no single sampler meets all requirements, but samplers are available which can meet
a majority of typical requirements.
A variety of devices including hi-vol samplers, size-selective samplers, nephelometers,
and precipitation collectors have been used to study the contributions of sulfur oxides and
particulate matter to such welfare effects as visibility reduction, soiling, and acidic
deposition.
1.4 SOURCES AND EMISSIONS
Both natural and manmade sources emit particulate matter and S02 into the atmosphere.
Natural emissions include dust, sea spray, volcanic emissions, biogenic emanations (such as
organic aerosols from plants), and emissions from wild fires. Manmade emissions originate
from stationary point sources, fugitive sources (such as roadway and industrial dust), and
transportation sources (vehicle exhausts). Reliable estimates for natural emissions of PM and
SO specific to the U.S. are not available. Proportional interpolations from global estimates
3\
indicate that in the U.S. natural sources emit 84 million metric tons of particles; estimates
of biogenic sulfur emissions from the northeast quadrant of the U.S. suggest a regional total
in the range of 0.1 million metric tons as SO-. Additional contributions from coastal and
oceanic sources may also be significant. Manmade sources emit 125 million to 383 million
metric tons of particulate matter and 27 million metric tons of sulfur oxides (mostly S02) per
year in the United States. These numbers should not be considered more than estimates because
of the assumptions and approximations inherent in emissions calculations.
The proximity of emissions to humans often is more important than relative intensity.
Emissions from combustion of home-heating fuels and transportation sources are minor on a
national level. However, they are emitted in densely populated areas and close to ground
level, thereby increasing the possibility of effects on human health and welfare. For such
reasons, certain manmade sources, particularly stationary point sources, have been given
special attention in this document. Historical trends in emissions of particulate matter
(excluding fugitive emissions) and sulfur oxide are shown in Table 1-1.
*XRD1B/D 1-11 3-3-81
-------
TABLE 1-1 (a) NATIONAL ESTIMATES OF PARTICIPATE EMISSIONS3
(10 metric tons per year)
SOURCE CATEGORY 1940 1950 1960 1970 1975 1978
Stationary fuel 8.7 8.1 6.7 7.2 5.1 3.8
combustion
Industrial processes 9.9 12.6 14.1 12.8 7.4 6.2
Solid waste disposal 0.5 0.7 0.9 1.1 0.5 0.5
Transportation 0.5 1.1 0.6 1.1 1.0 1.3
Miscellaneousb 5.2 3.7 3.3 1.0 0.6 0.7
TOTAL 24.8 26.2 25.6 23.2 14.6 12.5
Table 1-1 (b) NATIONAL ESTIMATES OF SULFUR OXIDE EMISSIONS
(10 metric tons per year)
SOURCE CATEGORY 1940 1950 1960 1970 1975 1978
Stationary fuel 15.1 16.6 15.7 22.7 20.9 22.1
combustion
Industrial processes 3.4 4.1 4.8 6.2 4.5 4.1
Solid waste disposal 0.0 0.1 0.0 0.1 0.0 0.0
Transportation 0.6 0.8 0.5 0.7 0.8 0.8
Miscellaneous15 0.4 0.4 0.4 0.1 0.0 0.0
TOTAL 19.5 22.0 21.4 29.8 26.2 27.0
Does not include industrial process fugitive particulate emissions, and non-
industrial fugitives from paved and unpaved roads, wind erosion, construction
activities, agricultural tilling, and mining activities.
Includes forest fires, agricultural burning, coal refuse burning, and structural
fires.
SOURCES: U.S. Environmental Protection Agency (1978b)
U.S. Environmental Protection Agency (1980a)
1.4.1 Sulfur Oxides
Most manmade sulfur oxide emissions come from stationary point sources, and more than 90
percent of these discharges are in the form of S0?. The balance consists of sulfates. Most
natural sulfur is emitted as reduced sulfur compounds, some portion of which probably becomes
oxidized in the atmosphere to S02 and sulfates.
*XRD1B/D 1-12 3-3-81
-------
1.4.2 Particulate Matter
Characteristics of particle emissions vary with the source and a host of other factors.
Primary particles from natural sources tend to be coarse. About 50 percent are smaller than
10 urn. Particles from non-industrial fugitive sources, such as unpaved roads and wind-eroded
farmland, are significant on a mass basis—an estimated 110 to 370 million metric tons a year.
However, only about 20 percent of this particulate matter is less than 1 urn in size. On the
other hand, most particles emitted by stationary and transportation sources are less than 2.5
urn in diameter. In addition, the variety of different toxic elements found in fine material
from stationary point sources tends to exceed that typically found in emissions from manmade
or natural fugitive sources.
Fugitive dust emissions exceed those from stationary point sources in most Air Quality
Control Regions with high total suspended particle loadings. However, the impact of this
pollution on populated areas may be lessened because: (1) a major portion of these emissions
consists of large particles which settle out in a short distance, and (2) most sources, such
as unpaved roads, exist in rural areas and their emissions spread over areas with low popula-
tion densities.
1.5 CONCENTRATIONS AND EXPOSURE
1.5.1 Sulfur Oxides
Sulfur oxide concentrations in the air have been markedly reduced during the past 15
years by restrictions on sulfur content in fuels, control devices on stationary and other
major sources, and tall stacks which relocate power-plant exhausts. Currently, only 1 percent
of the S02 monitoring sites show levels above 80 ug/m , as compared with 16 percent of the moni-
toring stations which reported annual means above this level in 1970. Despite this, some areas
still report very high short-term SO, concentrations (see Figure 1-3). Hourly values of 4000
3
to 6000 ug/m (1.5 to 2.3 ppm) are common near large smelters. Maximum hourly values above
1000 ug/m3 (0.4 ppm) exist in about 100 U.S. locations. Near isolated point sources, such peaks
may be reached very rapidly and be of only short duration (see, for example, Figure 1-4).
1.5.2 Particulate Matter
Following a downward trend from 1970 to 1974, TSP concentrations have not changed signi-
ficantly in recent years. Dusty, arid regions of the country still have high TSP values, as
do industralized cities in the east and far west. Ninetieth percentile values of 24-hr TSP
(values exceeded 10 percent of the time) above 85 ug/m3 are reported in every region of the
United States except Alaska (Figure 1-5). Annual mean TSP values range from 50 ug/m in the
New England region to 77 ug/m in the California-Nevada-Arizona region.
As discussed in Section 1.2, particulate matter is generally distributed in fine and
coarse mode size ranges of differing chemical compositions. A comparison of coarse, fine, and
hi-vol particle measurements for selected urban suburban, and rural sites is shown in Figure
1-6. Fine particles typically contribute about one-third of TSP mass in urban areas.
Sulfates often account for 40 percent of fine-particle levels which, in the eastern U.S., are
*XRD1B/D 1-13 3-2-81
-------
Figure 1-3. Sulfur dioxide second maximum 24-hour average by county, 1974-1976.
-------
1.50
1.40
1.30
1.20
£ 1.10
^ 1.00
2 .90
* .so
0 .70
0 .60
M „
8 -50
.40
.30
.20
.10
0
1
1 1 1 1 1 1 1 J 1 1 1 I 1 I 1 1 1 1 I I 1 1 1
— „
_ 1.30|ipmJ-33w>m JACKSON CO.. ALABAMA
|| SOURCE-ORIENTED SITE
-
-
-
-
~
-
-
~
-
-
\
I I I I I I \) I IV I I I I I I I I I I I I I
2 2 4 6 8 10 12 2 ' 6 8 10 12 2 4 6 8 10 12 2 4 6 8 10 12
NOON P.M. MID A.M. NOON P.M. MID A.M. NOON
NIGHT
JULY 29,1979
NIGHT
JULY 30,1979 JULY 31.
1979
Figure 1-4. One example of rapid increase in ambient sulfur dioxide concentration from near zero to
1.30 ppm (3410 ng/wr) during a period of approximately two hours is shown.
Source: Sulfur Dioxide One-Hour Values, National Aerometric Data Bank Standards Report for July
1979, Monitoring and Data Analysis Division, Office of Air Quality Planning and Standards.
US Environmental Protection Agency, Research Triangle Park, NC, (1981).
1-15
-------
en
•<— •—_BB^,-—_^B-S'_//? sSa?S=
-
[ |<
I J7.5 TO< 7B.5 (Jfl/ml
BrS.S TO 1003 p»/m3
H>
Figure 1-5. Total suspended particulate maximum annual average by county, 1974-1976.
-------
100 -
TSP (HI VOLI
•••--—•URBAN
— ^J^~~-^ SUBURBAN
- — CRURAL
2
-------
nearly the same in cities and rural areas. Sulfate and nitrate ions occur in high concentra-
tions during both summer and winter. Large areas of the United States experience 10 ug/m or
greater sulfate levels for one or two periods of a month or more every year. These areas are
so large that no background levels of fine particles are available for measurement east of the
Mississippi River. Southern California experiences high levels of sulfate and nitrate, parti-
cularly during photochemical (smog) incidents. Extremely high levels of organic aerosols also
occur in this area; from 2 to 4 p.m., during intensive periods of ozone formation, levels above
100 ug/m have been found. These organic aerosols consist largely of dicarboxylic acids and
other polyfunctional compounds. Concentrations of toxic organics and trace metals are highest
in cities. Levels of some fine-particle components have decreased because of control mea-
sures, such as reduction of lead in gasoline.
Coarse particles tend to settle close to sources. In most cases, these particles account
for 2/3 of TSP. During the summer, in dry regions such as Phoenix, Oklahoma City, El Paso, or
Denver they may contribute even higher proportions. The primary cause of high TSP appears to
be local dust, but in industrialized cities evidence exists for large contributions of soot,
fly ash, and industrial fugitive emissions.
Coarse particles are mainly composed of silica, calcium carbonate, clay minerals, and
soot. Chemical constituents in this fraction include silicon, aluminum, potassium, calcium,
and iron, together with other alkaline-earth and transition elements. Organic materials are
also found in coarse particles, including plant spores, pollens, and diverse biogenic detri-
tus. Much of this coarse material is road dust suspended by traffic action. Street levels of
resuspended dust can be very high. Traffic on unpaved roads generates huge amounts of dust
which deposit on vegetation and can be resuspended by wind action. Rain and snow can reduce
these emissions, but one study suggests that salting of roads is a major source of winter TSP.
Industrial fugitive emissions, particularly from unpaved access roads, construction activity,
rock crushing, and cement manufacturing can be major sources of coarse particles.
A number of calculational methods, generally categorized as source-apportionment or
source-receptor models, are being used to trace particle levels to their sources. The results
from chemical-element balance calculations or factor analysis are available for several
cities. Apportionments for these cities are presented in Chapter 5 as examples of results to
be expected in the future by application of these methods.
Although outdoor concentrations of pollutants can be measured at particular sites, the
highly mobile population can be exposed to either higher or lower values than community moni-
tors show. Some individuals in "clean" cities receive greater exposures than some individuals
in "polluted" cities. Indoor levels of S02, which tend to be lower than outdoor levels
because walls, floors, and furniture absorb the gas, are almost entirely due to >penetration
from outdoors. Presence or absence of air conditioning, air exchange rates, and activity
levels that resuspend dust all influence indoor particulate matter values. Also, outdoor fine
particles penetrate into buildings. Peak indoor TSP levels correlate to some degree with out-
door values after a time which depends on a building's air-exchange rate. Stationary ambient
*XRD1B/D 1-18 3-2-81
-------
air pollution monitors provide general statistics on composite population exposures; it is
extremely difficult to predict an individual's actual exposure to sulfur oxides and particulate
matter on the basis of community air-monitoring data alone.
1.6 ATMOSPHERIC TRANSPORT, TRANSFORMATION, AND DEPOSITION
The concentration of a pollutant at some fixed time and place beyond its source depends
on: (1) rate of emission and configuration of the source, (2) chemical and physical reactions
that transform one pollutant species to another, (3) transport and diffusion (dilution) as a
result of various meteorological variables, and (4) removal of the pollutant through inter-
action with land and water surfaces (dry deposition) and interaction with rain droplets or
cloud particles (wet deposition). Figure 1-7 schematically illustrates some of these pro-
cesses.
Processes governing transport and diffusion, chemical transformation, and wet and dry
removal of S02 and particulate matter are extremely complex and not completely understood.
The oxidation rate of SO- observed in urban and rural atmospheres is only partially accounted
for by gas-phase reactions. Liquid-phase catalytic reactions involving manganese and carbon
are possible contributing sources to observed rates, but further research is required to
quantify these processes under typical atmospheric conditions.
Dry deposition of S0_ is fairly well understood as a result of extensive measurements
over various vegetation canopies. Particle deposition has focused on physical aspects of the
process; that is, the aerodynamics, and little supportive measurement data exist on particles
with compositions typical of those in polluted atmospheres. It is apparent that coarse
particles are removed from the atmosphere much more rapidly than fine particles. Because of
this, the residence time of fine particles in the atmosphere appears to be in the order of 1
week and their transport distance can exceed 500 km.
Understanding of wet removal of S02 has progressed considerably in recent years, includ-
ing increased knowledge of solution-phase chemistry within rain droplets. Removal of parti-
cles, as well as gases, depends mainly on the physical character of precipitation events, which
in many instances may be the determining factor in accurate wet-removal prediction.
Characterization of the dynamics of the planetary boundary layer is essential to an ade-
quate understanding of pollutant transport and diffusion over all spatial scales. Though con-
siderable advances have been made in this area, ability to predict mean transport and dif-
fusion over long distances is less than adequate. This is due in part to the wide scatter,
spatially and temporally, of upper-air wind observations.
The long-range transport of the fine particle/SOp complex results in the superposition
and chemical interaction of emissions from many different types of sources. Present long-
range transport models are characterized by simple parameterization for chemical transfor-
mation and wet and dry removal, and by varying degrees of sophistication in treatment of
transport and diffusion. None of the models adequately treats the dynamics of the planetary
boundary. With further research and development, long-range transport models, though limited
*XRD1B/D 1-19 3-3-81
-------
FREETROPOSPHERIC
EXCHANGE
VERTICAL
DIFFUSION
AEROSOL
CONDENSATION
COAGULATION
CHEMICAL REACTIONS
ABSORPTION IN
CLOUD ELEMENTS
SEDIMENTATION
AS AEROSOL
1111 /
•
t"t"f'
////////'
DRY DEPOSITION ON
THE GROUND
O
KlmJJJ ANTHROPOGENIC ^
SOURCES SOURCES ^
fc <*•
-------
by a lack of data bases, should prove adequate for addressing issues associated with mo/eiriert
of pollutant emission over long distances.
1.7 ACIDIC DEPOSITION
The occurrence of acidic deposition, especially in the form of acidic precipitation, has
become a matter of concern in many regions of the United States, Canada, northern Europe,
Taiwan and Japan. Acidic precipitation in the Adirondack Mountains of New York State, in Maine,
in northern Florida, in eastern Canada, in southern Norway and in southwest Sweden has been
associated with acidification of waters in ponds, lakes and streams with a resultant disappear-
ance of animal and plant life. Acidic precipitation (rain and snow) is also believed to have
the potential to: (1) leach nutrient elements from sensitive soils, (2) cause direct and
indirect injury to forests, (3) damage monuments and buildings made of stone, and (4) corrode
metals.
Sulfur and nitrogen oxides have been most clearly implicated as pollutants contributing
to acidic deposition phenomena; and Chapter 7 of this document emphasizes the effects of wet
deposition of sulfur and nitrogen oxides and their products on aquatic and terrestrial
ecosystems. Dry deposition also plays an important rofe, but contributions by this process
have not been well quantified. Because sulfur and nitrogen oxides are so closely linked in
the formation of acidic precipitation, no attempt has been made to limit the present discus-
sion solely to a main topic of this document, sulfur oxides. A more thorough general review
of acidic deposition processes and associated environmental problems will be presented in a
future EPA document.
Sulfur and nitrogen oxides are considered to be the main precursors in the formation of
acidic precipitation. Emissions of such compounds involved in acidification are attributed
chiefly to the combustion of fossil fuels such as coal and oil. Emissions may occur at ground
level, as from automobile exhausts, or from stacks at times 1000 feet or more in height.
Emissions from natural sources are also involved; however, in highly industrialized areas,
emissions from manmade sources markedly exceed those from natural sources. In the eastern
United States the highest emissions of sulfur oxides derive from electric power generators
using coal. However, emissions of nitrogen oxides, mainly from automotive sources, tend to
predominate in the West. (Information regarding sources and emissions is discussed in Chapter
4 and is summarized in Sections 1.4 and 1.5 of this chapter.)
The fate of sulfur and nitrogen oxides, as well as other pollutants emitted into the
atmosphere, depends on their dispersion, transport, transformation and deposition. Sulfur and
nitrogen oxides or their transformation products may be deposited locally or transported long
distances from the emission sources (Altshuller and McBean, 1976; Pack, 1978; Cogbill and
Likens, 1974). Residence time in the atmosphere, therefore, can be brief if the emissions are
deposited locally or may extend to days or even weeks if long range transport occurs. The
chemical form in which emissions ultimately reach the receptor, i.e., the biological organism
or material affected, is determined by complex chemical transformations that take place between
*XRD1B/D 1-21 3-2-S1
-------
the emission sources and the receptor. Long range transport over distances of hundreds or
thousands of miles allows time for many chemical transformations to occur.
Sulfates and nitrates are among the products of the chemical transformations of sulfur
oxides (especially S0») and nitrogen oxides. Ozone and other photochemical oxidants are
believed to be involved in the chemical processes that form sulfates and nitrates. When sul-
fates and nitrates combine with atmospheric water, dissociated forms of sulfuric (HLSO.) and
nitric (HNO-) acids result; and when these acids are brought to earth in rain and snow, acidic
precipitation occurs. Because of long range transport, acidic precipitation in a particular
state or region can be the result of emissions from sources in states or regions many miles
away, rather than from local sources. To date, however, the complex nature of the chemical
transformation processes has not made it possible to demonstrate a direct cause and effect
relationship between emissions of sulfur and nitrogen oxides and the acidity of precipitation.
(Transport, transformation, and deposition of sulfur compounds are discussed in Chapter 6 of
this document; analogous information on nitrogen oxides is discussed in a separate document,
Air Quality Criteria for Oxides of Nitrogen, U.S. EPA, 1981).
Acidic precipitation has been arbitrarily defined as precipitation with a pH less than
5.6, because precipitation formed in a geochemically clean environment would have a pH of
approximately 5.6 due to the combining of carbon dioxide with air to form carbonic acid.
Currently the acidity of precipitation in the northeastern United States usually ranges from
pH 3.0 to 5.0; in other regions of the United States precipitation episodes with a pH as low
as 3.0 have also been reported in areas with average pH levels above 5.0 (see Figure 1-8).
The pH of precipitation can vary from event to event, from season to season and from geo-
graphical area to geographical area. Other substances in the atmosphere besides sulfur and
nitrogen oxides can cause the pH to shift by making it more acidic or more basic. For example,
dust and debris swept up in small amounts from the ground into the atmosphere may become com-
ponents of precipitation. In the West and Midwest soil particles tend to be more basic, but
in the eastern United States they tend to be acidic. Also, in coastal areas sea spray strongly
influences precipitation chemistry by contributing calcium, potassium, chlorine and sulfates.
In the final analysis, the pH of precipitation is a measure of the relative contributions of
all of these components (Whelpdale, 1978).
It is not presently clear as to when precipitation in the U.S. began to become markedly
acidic. Some scientists argue that it began with the industrial revolution and the burning of
large amounts of coal and others estimate that it began in the United States with the intro-
duction of tall stacks in power plants in the 1950's. However, other scientists disagree
completely and argue that rain has always been acidic. In other words, no definitive answer to
the question exists at the present time. Also, insufficient data presently exist to charac-
terize with confidence long-term temporal trends in changes in the pH of precipitation in the
United States, mainly due to the pH of rain not having been continuously monitored over
extended periods of time.
*XRD1B/D 1-22 3-2-81
-------
* NiUoMIAMiMpMflcDtimHIonProfln
O D*p«tiMfilof»
0 CmMlMAIniiMpMfleEiMlmMWMStnlMtCANSAP)
Figure 1-8. pH of rain sample as measured in the laboratory used in combination
with the reported amount of precipitation.
-------
Although acidic precipitation (wet deposition) is usually emphasized, it is not the only
process by which acids or acidifying substances are added to bodies of water or to the land.
Dry deposition also occurs. Dry deposition processes include gravitational sedimentation of
particles, impaction of aerosols and the sorption and absorption of gases by objects at the
earth's surface or by the soil or water. Dew, fog, and frost are also involved in the depo-
sition processes but do not strictly fall into the category of wet or dry deposition (Galloway
and Whelpdale, 1980; Schmel, 1980; Hicks and Wesley, 1980). Dry deposition processes are not
as well understood as wet deposition at the present time; however, all of the deposition
processes contribute to the gradual accumulation of acidic or acidifying substances in the
environment.
The most visible changes associated with acidic deposition, that is both wet and dry pro-
cesses, are those observed in the lakes and streams of the Adirondack Mountains in New York
State, in Maine, in northern Florida, in the Pre-cambrian Shield areas of Canada, in Scotland,
and in the Scandinavian countries. In these regions, the pH of the fresh water bodies has
decreased, causing changes in animal and plant populations.
The chemistry of fresh waters is determined primarily by the geological structure (soil
system and bedrock) of the lake or stream catchment basin, by the ground cover and by land use.
Near coastal areas (up to 100 miles inland) marine salts also may be important in determining
the chemical composition of the stream, river or lake. Sensitivity of a lake to acidification
depends on the acidity of both wet and dry deposition plus the same factors—the soil system
of the drainage basin, the canopy effects of the ground cover and the composition of the
waterbed bedrock. The capability, however, of a lake and its drainage basin to neutralize
incoming acidic substances is determined largely by the composition of the bedrocks (Wright
and Gjessing, 1976; Galloway and Cowling, 1978; Hendrey et al., 1980). Soft water lakes, those
most sensitive to additions of acidic substances, are usually found in areas with igneous bed-
rock which contributes few solids to the surface waters, whereas hard waters contain large con-
centrations of alkaline earths (chiefly bicarbonates of calcium and sometimes magnesium) derived
from limestones and calcareous sandstones in the drainage basin. Alkalinity is associated with
the increased capacity of lakes to neutralize or buffer the incoming acids. The extent to
which acidic precipitation contributes to the acidification process has yet to be determined.
The survival of natural living ecosystems in response to marked environmental changes or
perturbations depends upon the ability of constituent organisms of which they are composed to
cope with the perturbations and to continue reproduction of their species. Those species of
organisms most sensitive to particular environmental changes are first removed. However, the
capacity of an ecosystem to maintain internal stability is determined by the ability of all
individual organisms to adjust and survive, and other species or components may subsequently
be impacted in response to the loss of the most susceptible species.
The capacity of organisms to withstand injury from weather extremes, pesticides, acidic
deposition or polluted air follows the principle of limiting factors (Billings, 1978; Odum,
1971; Moran et al., 1980; Smith, 1980). According to this principle, for each physical factor
*XRD1B/D 1-24 3-2-81
-------
in the environment there exists for each organism a minimum and a maximum limit beyond which
no members of a particular species can survive. Either too much or too little of a factor such
as heat, light, water, or minerals (even though they are necessary for life) can jeopardize
the survival of an individual and in extreme cases a species. The range of tolerance (see
Figure 1-9) of an organism may be broad for one factor and narrow for another. The tolerance
limit for each species is determined by its genetic makeup and varies from species to species
for the same reason. The range of tolerance also varies depending on the age, stage of growth
or growth form of an organism. Limiting factors are, therefore, factors which, when scarce or
overabundant, limit the growth, reproduction and/or distribution of an organism.
ZONE OF
INTOLERANCE
ORGANISMS
ABSENT
LOWtH LIMITS
Of TOLEMMCE
ZONE OF
PHYSIOLOGICAL
STRESS
ORGANISMS
INFREQUENT
TOLERANCE RANGE
RANGE OF OPTIMUM
tun* units
or fOiEnuci
ZONE OF
PHYSIOLOGICAL
STRESS
ORGANISMS
INFREQUENT
GREATEST
ABUNDANCE
ZONE Of
INTOLERANCE
ORGANISVS
ABSENT
LOW4-
-GRAD1ENT-
-+-HIGH
Figure 1-9. Idealized conceptual framework illustrating the "law of tolerance/'
which postulates a limited tolerance range for various environmental factors
within which species can survive.
Source: Adapted from Smith (1980).
Continued or severe perturbation of an ecosystem can overcome its resistance or prevent
its recovery, with the result that the original ecosystem will be replaced by a new systen. In
the Adirondack Mountains of New York State, in eastern Canada, and parts of Scandinavia the
original aquatic ecosystems have been and are continuing to be replaced by ecosystems differ-
ent from the original due to acidification of the aquatic habitat. Forest ecosystems, however,
appear thus far to have been resistant to changes due to perturbation or stress froa acidifying
substances.
The impact of acidic precipitation on aquatic and terrestrial ecosystems is typically not
the result of a single or several individual precipitation events, but rather the result of
continued additions of acids or acidifying substances over time. Wet deposition of acidic
substances on freshwater lakes, streams, and natural land areas is only part of the problem
Acidic substances exist in gases, aerosols, and particulate matter transferred into the lakes,
*XRD1B/D
1-25
3-3-SI
-------
streams, and land areas by dry deposition as well. Therefore, all the observed biological
effects should not be attributed to acidic precipitation alone.
The disappearance of fish populations from freshwater lakes and streams is usually one of
the most readily observable signs of lake acidification. Death of fish in acidified waters has
been attributed to the modification of a number of physiological processes by a change in pH.
Two patterns related to pH change have been observed. The first involves a sudden short-term
drop in pH and the second, a gradual decrease in pH with time. Sudden short-term drops in pH
may result from a winter thaw or the melting of the snow pack in early spring and the release
of the acidic constituents of the snow into the water.
Long-term gradual increases in acidity, particularly below pH 5, interfere with reproduc-
tion and spawning, producing a decrease in population density and a shift in size and age of
the population to one consisting primarily of larger and older fish. Effects on yield often
are not recognizable until the population is close to extinction; this is particularly true
for late maturing species with long lives. Even relatively small increases (5 to 50 percent)
in mortality of fish eggs and fry can decrease yield and bring about extinction.
In some lakes, concentrations of aluminum may be as crucial or more important than pH
levels as factors causing a decline in fish populations in acidified lakes. Mobilization of
certain aluminum compounds in the water upsets the osmoregulatory function of blood in fish.
Aluminum toxicity to aquatic biota other than fish has not been assessed.
Although the disappearance of and/or reductions in fish populations are usually emphasized
as significant results of lake and stream acidification, also important are the effects on
other aquatic organisms ranging from waterfowl to bacteria. Organisms at all trophic (feeding)
levels in the food web appear to be affected. Species reduction in number and diversity may
occur, biomass (total number of living organisms in a given volume of water) may be altered
and processes such as primary production and decomposition impaired.
Significant changes that have occurred in aquatic ecosystems with increasing acidity
include the following:
1. Fish populations are reduced or eliminated.
2. Bacterial decomposition is reduced and fungi may dominate saprotrophic communi-
ties. Organic debris accumulates rapidly, tying up nutrients, and limiting
nutrient mineralization and cycling.
3. Species diversity and total numbers of species of aquatic plants and animals are
reduced. Acid-tolerant species dominate.
4. Phytoplankton productivity may be reduced due to changes in nutrient cycling and
nutrient limitations.
5. Biomass and total productivity of benthic macrophytes and algae may increase due
in part to increased lake transparency.
6. Numbers and biomass of herbivorous invertebrates decline. Tolerant invertebrate
species, e.g., air-breathing insects may become abundant primarily due to reduced
fish predation.
7. Changes in community structure occur at all trophic levels.
*XRD1B/D 1-26 3-2-81
-------
An indirect effect of acidification potentially of concern to human health is the possi-
ble contamination of edible fish and of water supplies. Studies in Canada and Sweden reveal
high mercury concentrations in fish from acidified regions. Lead has been found in plumbing
systems with acidified water, and persons drinking the water could be affected by the lead.
However, no examples have yet been documented of such human effects having actually occurred
in response to acidic precipitation processes.
Soils may become gradually acidified from an influx of hydrogen (H*) ions. Leaching of
the mobilizable forms of mineral nutrients may occur. The rate of leaching is determined by
the buffering capacity of the soil and the amount and composition of precipitation. Unless
the buffering capacity of the soil is strong and/or the salt content of precipitation is high,
leaching will in time result in acidification. Anion mobility is also an important factor in
the leaching of soil nutrients. Cations cannot leach without the associated anions also leach-
ing. The capacity of soils to adsorb and retain anions increases as the pH decreases, when
hydrated oxides of iron and aluminum are present.
Sulfur, like nitrogen, is essential for optimal plant growth. Plants usually obtain sulfur
in the form of sulfate from organic matter during microbial decomposition. Wet and dry deposi-
tion of atmospheric sulfur is also a major source. In soils where sulfur and nitrogen are
limiting nutrients, such deposition may increase growth of some plant species. The amounts of
sulfur entering the soil system from atmospheric sources is dependent on proximity to industrial
areas, the sea coast, and marshlands. The prevailing winds and the amount of precipitation in
a given region are also important (Halsteand and Rennie, 1977). Near fossil-fueled power plants
and industrial installations the amount of sulfur in precipitation may be as much as 150 pounds
per acre (168 kg/ha) or more (Jones, 1975).
At present there are no documented observations or measurements of changes in natural
terrestrial ecosystems or agricultural productivity directly attributable to acidic precipita-
tion. The information available regarding vegetational effects concerns the results of a
variety of controlled research studies, mainly using some form of "simulated" acidic rain,
frequently dilute sulfuric acid. The simulated "acid rains" have deposited hydrogen (H ),
sulfate (S04=) and nitrate (NOl) ions on vegetation and have caused necrotic lesions in a wide
variety of plants species under greenhouse and laboratory conditions. Such results must be
interpreted with caution, however, Because growth and morphology of leaves under such conditions
are not necessarily typical of field conditions.
Damage to monuments and buildings made of stone, corrosion of metals and deterioration of
paint may also result from acidic precipitation. Because sulfur compounds are a dominant com-
ponent of acidic precipitation and are deposited during dry deposition as well, the effects
resulting from the two processes cannot be clearly distinguished. Also, deposition of sulfur
compounds on stone surfaces may provide a medium for microbial growth that can result in
deterioration.
*XRD1B/D 1-27 3-3-81
-------
Several aspects of the acidic precipitation problem remain subject to debate because
existing data are ambiguous or inadequate. Important unresolved issues include: (1) the rate
at which rainfall is becoming more acidic and the rate at which the problem is becoming geo-
graphically more widespread; (2) the quantitative contributions of various acids to the over-
all acidity of rainfall; (3) the relative extent to which the acidity of rainfall in a region
depends on local emissions of nitrogen and sulfur oxides versus emissions transported from
distant sources; (4) the relative importance of changes in total mass emission rates compared
to changes in the nature of the emission patterns (ground level versus tall stacks) in contri-
buting to regional acidification of precipitation; (5) the relative contribution of wet and
dry deposition to the acidification of lakes and streams; (6) the geographic distribution of
natural sources of NO ,. SO and NH- and the significance and seasonality of their contributions;
XX >3
(7) the existence and significance of anthropogenic, non-combustion sources of SO , NO and
HC1; (8) the dry deposition rates for SO , N09, sulfate, nitrate and HC1 over various terrains
A. £m
and seasons of the year; (9) the existence and reliability of long-term pH measurements of lakes
and headwater streams; (10) the acceptability of current models for predicting long range trans-
port of SO and NO and for acid tolerance of lakes; (11) the feasibility of using liming or
other corrective procedures to prevent or reverse acid damage and the costs of such procedures;
(12) the effects of SO and NO and hydrogen ion deposition on ecosystem dynamics in both aqua-
A J\
tic and terrestrial ecosystems; (13) the effectiveness of fertilization resulting from sulfate
and nitrate deposition on soils; (14) the effects, if any, of acidic deposition on agricultural
crops, forests and other native plants; and (15) the effects of acidic deposition on soil
microbial processes and nutrient cycling. A more comprehensive evaluation of scientific
evidence bearing on these issues is being prepared as part of a forthcoming EPA document on
acidic deposition.
1.8 EFFECTS ON VEGETATION
The widespread occurrence of particulate matter, sulfur dioxide and other substances in
the atmosphere frequently results in exposure of terrestrial vegetation simultaneously to
these pollutants as well as other phytotoxic pollutants. More is known about the effects of
sulfur dioxide on vegetation than about the effects of particulate matter. Studies of the
effects of particulate matter have generally focused on the effects of heavy accumulations and
the reduction in photosynthesis resulting from these accumulations. The more subtle effects
of particulate matter on vegetation have not been extensively investigated and are, therefore,
not well understood. Plant response following exposure to sulfur oxides and particulate matter
in combination is even less well understood. Chapter 8 of this document discusses the effects
of sulfur oxides and particulate matter on vegetation.
Sulfur dioxide and particulate sulfate are the main forms of sulfur in the atmosphere,
and a plant may be exposed to these pollutants in several different ways. Dry deposition of
particulate matter and wet deposition of gases and particles bring sulfur compounds into
*XRD1B/D 1-28 3-3-81
-------
contact with plant surfaces and soil substrates. The effects of such exposure are more
difficult to assess than those associated with the entry of S02 through plant stomata.
Most sulfur dioxide enters leafy plants through the stomata (Chamberlain, 1980). After
entering the stomata and passing into cells within the leaf, sulfur dioxide is converted to
sulfite and bisulfite, which may then be oxidized to sulfate. Sulfate appears to be less toxic
than sulfite and bisulfite. As long as the absorption rate of S02 in plants does not exceed
the rate of conversion to sulfate, the only effects of exposure may be changes in opening or
closing of stomata or undetectable changes in the biochemical or physiological systems. Such
effects may abate if SOg concentrations are reduced. Both negative and positive influences on
crop productivity have been noted following low-dose exposures.
Symptoms of SOg-induced injury in higher plants may be quite variable since response is
governed by pollutant dose (concentration x duration of exposure), the kinetics of the exposure
(e.g., day vs. night, peak vs. long-term); the physiological status of the plant, the matura-
tion stage of plant growth, environmental influences on the pollutant/plant interaction, and
the environmental influences on the metabolic status of the plant itself. Although the product
of time and concentration may remain constant, the effect of exposure may vary for a given dose.
The relation of exposure to injury is generally more sensitive to changes in concentration than
to changes in duration of exposure. Plant response to dynamic physical factors such as light,
leaf surface moisture, relative humidity, and soil moisture may influence pollutant uptake
through internal physiological changes as well as stomatal opening and closing and hence play
a major role in determining sensitivities of species and cultivars or the time of sensitivity
of each on a seasonal basis. Dose-response relationships are significantly conditioned by
environmental conditions before, during, and following exposure to SQy.
Plants may respond to exposures of S0_ and related sulfur compounds in the following ways:
(1) no detectable response, (2) increased growth and yield resulting from fertilization, (3)
injury manifested as reductions in growth and yield without visible symptoms appearing on the
foliage or with only very mild symptoms that would be difficult to perceive as induced by air
pollution without comparing them with a control set of plants grown in pollution-free condi-
tions, (4) injury exhibited as chronic or acute symptoms on foliage with or without associated
reduction in growth and yield, and (5) death of plants and plant communities.
Under certain conditions, atmospheric S02 can have beneficial effects on agronomic vege-
tation (Noggle and Oones, 1979). The amount of sulfur accumulated from the atmosphere by leaf
tissues is influenced by the amount of sulfur in soil relative to the sulfur requirement of
the plant. After exposure to low doses of S02, plants grown in sulfur-deficient soils have
exhibited increased productivity.
Some species of plants are sensitive to low concentrations of S02, and some of these plants
may serve as bioindicators in the vicinity of major sources of SOg. Even these sensitive spe-
cies may be asymptomatic, however, depending on the environmental conditions before, during,
and after S0? exposure. Various species of lichens appear to be among the most sensitive
plants.
*XRD1B/D 1-29 3-3-81
-------
Because of space limitation, it is not possible to list all plants that are known to be
sensitive to various doses of SCL. It has also been demonstrated that plant response to air
pollutants varies at the genus, species, variety, and cultivar levels. Lists of sensitive
plants have been prepared on the basis of the expression of visible symptoms by any given plant.
Analyses of injury expressed in terms of growth or yield losses, however, have been limited
due to the relative lack of empirical data quantifying such losses in relation to SCL exposures.
Jacobson and Hill (1970) included a listing of plants sensitive to the major phytotoxic
air pollutants. Linzon (1972) has listed 36 tree species as being tolerant, intermediate, and
sensitive to S0_. Many of these sensitivity lists have not attempted to identify the dose
required to induce visible injury on indicator species. However, Jones et al. (1974) have
published such details based upon observations over a 20-year period of 120 species growing in
the vicinity of coal-fired power plants in the southeastern United States (Table 1-2).
TABLE 1-2. SULFUR DIOXIDE CONCENTRATIONS CAUSING VISIBLE INJURY TO
VARIOUS SENSITIVITY GROUPINGS OF VEGETATION3
(ppm S02)
Maximum Sensitivity grouping
concentration Sensitive
ppm SOg
Peak 1.0-1.5
1-hr 0.5-1.0
3-hr 0.3-0.6
Ragweeds
Legumes
Blackberry
Southern pines
Red and black oaks
White ash
Sumacs
Intermediate
ppm S02
1.5-2.0
1.0-2.0
0.6-0.8
Maples
Locust
Sweetgum
Cherry
Elms
Tuliptree
Many crop and
garden species
Resistant
ppm SCL
>2.0
>2.0
>0.8
White oaks
Potato
Upland cotton
Corn
Dogwood
Peach
Based on observations over a 20-year period of visible injury occurring on
over 120 species growing in the vicinities of coal-fired power plants in the
southeastern United States. Source: Jones et al., 1974.
*XRD1B/D
1-30
3-2-81
-------
Mclaughlin (1980) used symptom data as collected by Dreisinger and McGovern (1970) on 31
species of forest and agricultural plants following SCL exposure and plotted the average,
maximum, and minimum tolerances of individual species (Figure 1-10). The injury threshold for
most sensitive plants was 0.41, 0.37, 0.28, and 0.12 ppm S02 at averaging intervals of 1, 2,
4, and 8 hours, respectively.
As S02 exposure levels increase, plants develop more predictable and more obvious visible
symptoms. Foliar symptoms advance from chlorosis or other types of pigmentation changes to
necrotic areas and the extent of necrosis increases with exposure. Studies of the effects of
SOy on growth and yield have demonstrated a reduction in the dry weight of foliage, shoots,
roots, and seeds, as well as a reduction in the number of seed. At still higher doses,
reductions in growth and yield increase. Extensive mortality has been noted in forests continu-
ously exposed to S02 for many years.
The presence of acute or chronic foliar injury is not necessarily associated with growth
or yield effects. Furthermore, the degree of foliar injury, when present, may not always be a
reliable indicator of subsequent growth or yield effecfts.
Plant response to S02 may occur at many levels. However, two parameters, visible damage
to foliage and plant productivity, provide the most functional basis for evaluating response.
Both can be quantified as a "cost" to economic or ecological performance of many plant
species.
Dose-response relationships involving visible injury may be expressed in terms of the
level of injury (percent leaf area destroyed) produced for a single species, or as the upper
and lower limits of sensitivity of a group of species. The latter approach is presented here
because it provides data more applicable to responses of plant populations and because of the
difficulties in quantifying a dependable relationship between degree of visible injury and
growth responses (see Section 8.2.7). Data on generalized concentrations at which sensitive,
intermediate, and resistant species may be injured by S0» were presented earlier in Table 1-2.
In Figure 1-10, the data of Dreisinger and McGovern (1970) are graphed to show upper and lower
concentration limits of susceptibility of 31 species of herbs, trees, and shrubs to visible
foliar injury. Plotted as a function of S02 concentration and exposure time, these data demon-
strate a number of important points. First, the most sensitive plants at each concentration
were injured at S02 levels 6 to 7 times lower than the most resistant plants; secondly, the
dose or product of concentration x time (ppm-hr) required to cause injury was 30 to 60 percent
lower for 1-hr than 8-hr exposures. This emphasizes the importance of differences in exposure
duration when comparing specific degrees of injury associated with different exposure concen-
trations. Finally, exposure to S02 at 0.5 ppm for 3 hr represents a rather close estimate of
the injury threshold for about 50 percent of the species studied.
A second approach to defining dose-response relationships focuses on the numbers of indi-
viduals in a plant population which may be injured as function of exposure concentration. As
an example, a "worst-case" situation of S02 exposures and vegetation effects near a rural
coal-fired power plant in the southeastern United States provides some interesting data
*XRD1B/D 1-31 3-3-81
-------
AVERAGING INTERVAL (HI
Figure 1-10. Exposure thresholds for minimum, maximum, and average sensitivity of 33
plant species to visible foliar injury by S(>2.
Source: Dreisinger and McGovern (1970) as applied by Mclaughlin (1980).
1-32
-------
(Mclaughlin and Lee, 1974). During the period 1970-1973 (before partial sulfur scrubbing and
stack elevation improved surrounding air quality), surveys of vegetation in the vicinity of
this plant documented foliar injury of 84 plant species growing in the vicinity of continuous
S02 monitoring stations. Plotting of these data (Figure 1-11) as a function of exposure
100
o
in
EC
O.
CO
O
O
01
O
o
GC
Ul
Q.
SO2 CONCENTRATION, ppm
Figure 1-11. Percentage of plant species visibly injured as a function of peak, 1-hour, and
3-hour SC>2 concentrations.
Source: Mclaughlin and Lee (1974); Mclaughlin (1980).
concentration provides an index of probability of injury of species in a plant community as a
function of S02 concentration. Note, for example, 10 percent of the plant population (here 8
of 84 species) is injured at peak, 1-hr, and 3-hr concentrations of sl.OO, sO.50, and =0.30 ppa.
For agricultural crops, data on SO, effects on plant growth and yield, in most cases,
\ £.
provide the most relevant basis for studying dose-response relationships. As a whole-plant
measurement, plant productivity is an integrative parameter which considers the net effect of
multiple factors over time. Productivity data are presently available for a wide range of
species under a broad range of experimental conditions. Because results would not be expected
to be closely comparable across these sometimes divergent experimental techniques, data have
been tabulated separately for controlled field exposures (see Chapter 8, Table 8-3), labora-
tory studies with agronomic and horticultural crops (see Chapter 8, Table 8-4) and tree species
(see Chapter 8, Tables 8-5 and 8-6), and a variety of studies with native plants (see Chapter
8, Table 8-7).
*XRD1B/D
1-33
3-3-81
-------
Many presently available dose-response data sets have been derived from studies of con-
trolled exposures in the laboratory or in field chambers. In spite of the variety of species
studied and experimental protocols utilized, it is possible to derive potentially useful
generalizations from these data:
(1) The concentration threshold for visible injury is generally lower than the threshold
for effects on growth and yield, especially for acute exposure effects. Doses caus-
ing visible injury to 10 percent of a variety of southeastern plant species were
0.30 ppm for a 3-hr exposure and 0.50 for a 1-hr exposure.
(2) Visible injury data emphasize the greater relative biological effectiveness of short-
term higher concentrations than longer exposures with the same total dose.
(3) Plant responses to S0? may be positive, neutral, or negative over a rather wide range
of exposure dose. Positive responses were generally restricted to a very few species
or conditions when plants were known to have been grown in sulfur-deficient soils.
Negative responses constituted approximately 85 percent of all responses noted above
threshold levels for visible injury.
(4) Data derived from continuous or intermittent controlled chronic exposures (>4 weeks)
of six species or cultivars in field chambers provided a basis for estimating yield
responses from total logarithmically transformed exposure dose. Regression analysis
of these data provided a no-effects limit of approximately 6.0 ppm-hr. Yield losses
of 10 percent and 20 percent were similarly estimated at 10 and 27 ppm-hr, respec-
tively.
(5) An attempt to analyze data from 23 species or cultivars from laboratory and green-
house exposures generally indicated greater sensitivity than the six species or cul-
tivars tested in the field (above). A boundary line which delimited the maximum
observed response over the range of concentrations employed indicated that upper-
limit yield losses were approximately 10 percent and 20 percent for exposure doses
of 0.9 and 17 ppm-hr, respectively. Average responses determined by regression
analysis indicated that 10 and 20 percent yield losses would be produced by exposures
of 0.6 ppm-hr and 4.5 ppm-hr, respectively.
In interpreting the dose-response information, it should be noted that responses of plants
to SO^ in the field may occur as a consequence of one or more short-term episodes or as a result
of the cumulative dose experienced over an entire growing season. Regression analysis of data
from both controlled exposures in field chambers and from laboratory and greenhouse studies
showed positive and statistically significant correlations between degree of yield loss and
logarithmically transformed exposure dose in ppm-hr. The plants and conditions utilized in
field studies, which were heavily oriented toward crop plants, provided generally ,lower yield
losses for the same exposure dose than did laboratory and greenhouse studies.
A critical need in evaluating the likelihood of adverse effects occurring in association
with longer-term SO^ exposures is the identification of the fraction of the total SO- exposure
*XRD1B/D 1-34 3-2-81
-------
which may constitute a stress to plant growth and development. A review of dose-response data
indicates that this level may be approximately 0.05 ppm for some sensitive crop species or
lower (=0.02 ppm) for certain types of sensitive native vegetation (e.g., pines and lichens)
under field conditions. Data from studies involving SOp alone and in combination with other
pollutants may provide a more accurate basis for determining this level.
At present, data concerning the interactions of S02 with other pollutants indicate that,
on a regional scale, S02 occurs at least intermittently at concentrations high enough to pro-
duce significant interactions with other pollutants, principally 03- A major weakness in the
approach to pollutant interactions, however, is the lack of in-depth analysis of existing
regional air quality data sets for the three principal pollutants (S02, 03, and N02). These
data should determine how frequently and at what concentrations the pollutants occur together
both spatially and temporally within regions of major concern. The relative significance of
simultaneous versus sequential occurrence of these pollutants to effects on vegetation is also
not well documented and is critical in evaluating the likelihood and extent of potential pol-
lutant interactions under field conditions.
A few studies have reported that combinations of particulate matter and SO^, or particu-
late matter and other pollutants, increase foliar uptake of S02, increase foliar injury of
vegetation by heavy metals, and reduce growth and yield. Because of the complex nature of
particulate pollutants, conventional methods for assessing pollutant injury to vegetation, such
as dose-response relationships, are poorly developed. Studies have generally reported vegeta-
tional responses relative to a given source and the physical size or chemical composition of
the particles. For the most part, studies have not focused on effects associated with specific
ambient concentrations. Coarse particles such as dust directly deposited on the leaf surfaces
result in reduced gas exchange, increased leaf surface temperature, reduced photosynthesis,
chlorosis, reduced growth, and leaf necrosis. Heavy metals deposited either on leaf surfaces
or on the soil and subsequently taken up by the plant can result in the accumulation of toxic
concentrations of the metals within the tissue.
Natural ecosystems are integral to the maintenance of the biosphere and disturbances of
stable ecosystems may have long-range effects which are difficult to predict. Within the
United States anthropogenic contributions to atmospheric sulfur exceed natural sources and in
the Northeast, approximately 60 percent of the anthropogenic emissions into the atmosphere are
deposited (wet and dry deposition) on terrestrial and aquatic ecosystems. The fate and
distribution of anthropogenic sulfur deposited in these systems is not well understood. Wet
deposition of sulfur compounds is discussed in Chapter 7.
Data relating ecosystem responses to specific doses of S02 and other pollutants are
difficult to obtain and interpret because of the generally longer periods of time over which
these responses occur and because of the many biotic and abiotic factors which modify them.
Vegetation within terrestrial ecosystems is sensitive to S02 toxicity, as evidenced by
changes in physiology, growth, development, survival, reproductive potential and commity
*XRD1B/D 1-35 3-3'81
-------
composition. Indirect effects may occur as a result of habitat modification through influences
on litter decomposition and nutrient cycling or through altered community structure. At the
community level chronic exposure to S02> particularly in combination with other pollutants such
as 0,, may cause shifts in community structure as evidenced by elimination of individuals or
populations sensitive to the pollutant. Differential effects on individual species within a
community can also occur through direct effects on sensitive species and through alteration of
relative competitive potential of species with which they compete within the plant community.
Particulate emissions have their greatest impact on terrestrial ecosystems near large
emission sources. Particulate matter in itself constitutes a problem only in those few areas
where deposition rates are very high. Ecological modification may occur if the particles con-
tain toxic elements, even though deposition rates are moderate. Solubility of particle con-
stitutents is critical, since water-insoluble elements are not mobile within the ecosystem.
Most of the material deposited by wet and dry deposition on foliar surfaces in vegetated areas
is transferred to the soil where accumulation in the litter layer occurs.
1.9 EFFECTS ON VISIBILITY AND CLIMATE
Atmospheric visibility is often used by airport weather observers and others to connote
visual range. Visual range is generally defined as the farthest distance at which one can see
a large, black object against the horizon sky. In the everyday sense, however, visibility
relates to the perceived characteristics of viewed surroundings including contrast and color
of objects and sky and atmospheric clarity. Pollution-derived effects on visibility can be
classified as: (1) coherent plumes or haze layers visible because of contrast with background,
(2) widespread, relatively homogeneous haze that reduces contrast of viewed targets and reduces
visual range. The kind and degree of effects are determined by the distribution and character-
istics of atmospheric particulate matter and nitrogen dioxide, which scatter and absorb light.
Reductions in visibility can adversely affect transportation safety, property values, and
aesthetics. When visibility (visual range) drops below 3 miles, FAA regulations restrict
flight in controlled air to those aircraft equipped with IFR instrumentation. Assessment of
the social, psychological, and economic value of visibility is difficult. Preliminary studies
of the economic value of visibility conducted in both urban and non-urban settings show pro-
mise but are currently too limited and premature to permit any large-scale evaluation.
Current U.S. visibility as indicated by regional airport visual range data is depicted in
Figure 1-12. Such human observations are subject to some limitations, but the data indicate
regional trends. The best visibility occurs in the mountainous Southwest where annual median
visibility exceeds 70 miles (110 km). East of the Mississippi and south of the Great Lakes
annual median visibilities are less than 15 miles (24 km) and are significantly lower in the
summer time, particularly during periodic episodes of regional haze.
On a regional scale, visibility reduction is generally dominated by light scattering and
by fine particles, particularly those in the 0.1 to 2 urn size range. In urban areas, absorp-
tion of light by fine carbanaceous particles and, to a lesser extent, N02 can become important.
*XRD1C/B 1-36 3-2-81
-------
u>
P: BASED ON PHOTOGRAPHIC
PHOTOMETRY DATA
N: BASED ON NEPHELOMETRY DATA
•: BASED ON UNCERTAIN EXTRAPOLATION
OF VISIBILITY FREQUENCY DISTRIBUTION
Figure 1-12. Map shows median yearly visual range (miles) and isopleths for suburban/nonurban
areas, 1874-1976,
Source: Trljonls and Shapland (1978).
-------
Total extinction is the sum of scattering and absorption by pollutants and Rayleigh or "blue
sky" scatter by air molecules. Visual range is inversely related to total extinction and can
be estimated, if extinction is known, by the Koschmieder relationship (see Figure 1-13).
Because extinction is wavelength and sun angle dependent, particle-derived haze may appear
bluish, white, grey, or brown under varying conditions.
Extinction due to scattering is closely proportional to the fine particle mass concentra-
tion (Figure 1-14) with typical extinction/mass ratios (for <70 percent humidity) in the range
of 0.003 to 0.005 km /jjg/m . Measurements in general areas suggest that the extinction due
to fine particle scattering will increase by a factor of two to three as relative humidity is
increased from 70 to 90 percent. This is due to absorption of atmospheric water vapor by
aerosol constituents such as sulfates. The major constituents of fine particles from natural
and anthropogenic sources contribute in varying degrees to visibility impairment. Theoretical
and empirical results suggest that two constituents, sulfates and elemental carbon, gener-
ally tend to be most significant. Sulfate, with associated ammonium and water, often
dominates the fine mass and visibility impairment, while elemental carbon can be a major
visibility-reducing specie in urban areas. Significant variations can occur at different times
and sites. Our knowledge of the roles of several possibly important species such as organics
is hindered by the lack of sufficient good data.
Studies of trends in Eastern airport visibility indicate that, while wintertime visibili-
ties improved in some Northeastern locations, overall Eastern visibility declined (Figure 1-15).
Summer, often the season of best visibility in the early fifties, is currently the worst season.
From 1948 to 1974, summertime haze (extinction) increased by more than 100 percent in the
central Eastern States, by 50 to 70 percent for the Midwest and Eastern sunbelt States, and by
10 to 20 percent for the New England area. Although the results of airport surveys should be
viewed with caution, the results are consistent from site to site. Similarities exist in the
long-term record of the spatial and seasonal trends in airport visibility and trends in sul-
fates, point source sulfur oxides emissions, and coal use. These similarities suggest, but do
not prove, that historical visibility trends in the East were caused, at least in part, by
regional sulfur oxides emissions and resultant sulfate aerosols.
The currently available visibility monitoring methods measure different aspects of visi-
bility impairment. Generally, contrast type measurements (such as photography, telephotometry,
and human eye observations) relate well to the perception of visual air quality, while extinc-
tion or scattering measurements (such as transmissometry and nephelometry) relate to the cause
of visibility degradation. Each of the above measurement methods can be used to approximate
visual range. No single method has been proven totally effective in measuring light absorption.
The methods used thus far include determining the difference between extinction and scattering,
several filtering methods, and a refractive index method.
The longer residence time and light attenuating properties of fine particles may also lead
to slow and subtle changes in the nature of the atmosphere and, possibly, in climate. For
XRD1C/B 1-38 3-3-81
-------
0.50
0.40
0.30
0.20
tu
u
8 0-10
z
E
0.05
0.03
I
I I I I i
10 20
VISUAL RANGE, km
50
100
Figure 1-13. Inverse proportionality between visual range and the
scattering coefficient, bscat, was measured at the point of obser-
vation. The straight line shows the Koschmieder formula for non-
absorbing (bext = bsca*) media. V = 3.9/bscat. The linear corre-
lation coefficient for V and 1/bscat is 0.89.
Source: Horvath and Noll (1969).
1-39
-------
60
w
II
Ul
40
20
- 0.2
I
! 0.1
4/17
4/18
4/19 4/20
TIME, days
4/21
4/22
Figure 1-14. Simultaneous monitoring of bscat and fine-particle mass
in St Louis in April 1973 showed a high correlation coefficient of
0.96, indicating that bscat depends primarily on the fine-particle
concentration.
Source: Macias and Husar (1976).
1-40
-------
194852
1960-64
1970-74
LZl CH
EXTINCTION
COEFFICIENT, km'1
VISIBILITY, miln
>0.36
<6.6
0.34.36
6.64
0.24-0.30
8-10
0.18-0.24
10-13.3
<0.18
>13.3
Figure 1-15. The spatial distribution of 5-year average extinction coefficients shows the substantial
increases of third-quarter extinction coefficients in the Carolina:. Ohio River Valley, and Tennessee-
Kentucky area. In the summers of 1948-52, a 1000-krn size multistate region centered around Atlanta.
GA, had visibility greater than 15 miles; visibility has declined to less than 8 miles by the 1970V The
spatial trend of winter (first quarter) visibility shows improvements in the Northeast megalopolis
region and some worsening in the Sunbelt region. Both spring and fall quarters exhibit moderate but
detectable increases over the entire eastern United States.
Source: Husar et al. (1979).
1-41
-------
example, a fraction of the solar radiation may be absorbed by aerosols, further reducing the
amount of radiation reaching the earth's surface and, at the same time, heating the aerosol
layer itself. On a hazy day, the direct solar radiation is reduced to about one-half of that
on a clear day, but most of the energy reappears as diffuse skylight. However, there is an
overall loss of up to about 10 to 20 percent of the radiation reaching the surface.
If there are no clouds between the observer and the sun, the intensity of direct solar
radiation for a given solar elevation depends on the variable amount of dust, haze, and water
vapor in the atmosphere. The extinction produced by these constituents is called "atmospheric
turbidity." During hazy episodes, turbidity coefficients of 0.6 to 1.0 are often reported,
resulting in a condition in which 75 to 90 percent of the solar radiation is removed from the
direct beam, 7.5 to 18 percent is lost to space and 7.5 to 18 percent is lost as atmospheric
heating. One of the consequences of such a hazy atmosphere is the disappearance of shadow
contrast. Long-term trends in atmospheric turbidity in the eastern U.S. are qualitatively
consistent with those for airport visibility (Figure 1-16).
TURBIDITY TREND
1961-66 -••-
1972-75
8
i
I
IT I I I'I'I'I'I'I'I
MEMPHIS. TN
.1.1.1.1.1.!. 1.1.1.1 I
OAK RIDGE, TN
1.1.1.1.1.1.i.r. i.1.1
I I'll I
GREENSBORO. NC
M J «l A SOI
MONTH
I TMM'rri'1'l'ITI'
BALTIMORE, MD
Figure 1-16. Seasonal turbidity patterns for 1961-66 and 1972-75 are shown for selected regions in the
Eastern United States.
Source: Flowers et al. (1969).
*XRD1C/B
1-42
3-2-81
-------
The attenuation of solar radiation from scattering and absorption by particles in the
atmosphere is probably an important factor in climatic change on all scales. On local scales
associated with urban and industrial areas, any significant attenuation of radiation by air
pollution can, in addition to other well-recognized factors, result in changes in local
weather. It is possible that local- and regional-scale changes in solar radiation caused by
human activity may ultimately influence the heat and water vapor content of the atmosphere on
very large scales, but solar radiation and aerosol levels measured at stations remote from
pollutant sources have not as yet displayed any trend that can be related to human activity.
Cloud- and precipitation-forming processes may be divided into two broad classes: macro-
physical and microphysical processes. Macrophysical processes involve the rise and descent of
air masses and the amount of water vapor available for condensation. Atmospheric aerosols,
primarily those that are strongly hygroscopic, influence the microphysics of cloud formation.
The incorporation of particles into rain and fog droplets can change the "quality" of precipi-
tation by changing its chemical composition. However/ the complex interactions of cloud- and
precipitation-forming processes obscure the specific role of manmade aerosols.
1.10 EFFECTS ON MATERIALS
The nature and extent of physical damage to materials by sulfur oxides and particulate
matter have been investigated by field and laboratory studies. Various approaches are used to
estimate economic damage; economic determinations may directly relate ambient pollutant levels
to economic damage estimates or may estimate economic damage based on physical damage functions.
The latter method of determination, termed here the physical damage function approach, has been
the past method of choice. Other studies, especially in the last decade, have employed the
first approach. Both approaches share a common element—an estimation of willingness to pay.
Physical damage function approaches have been most widely used and, therefore, have re-
ceived the most extensive treatment. The damage function, which is a mathematical expression
linking exposure to damage, is expressed in terms appropriate to the interaction of the pol-
lutant and material. For example, corrosion of metal would be expressed in units of thickness
lost, and deterioration of paint in units of reflectance or thickness lost. A major problem
in establishing reliable damage functions involves separating influences of the target pollu-
tant from that of meteorological parameters, (e.g., temperature, relative humidity, sunlight,
wind speed, wind direction) and other air pollutants. For the corrosion of metals, time of
wetness is the most important variable.
Economic valuations may require determinations of a critical damage level. This level
represents the point at which the service life or utility of the material ends or is severely
impaired. When this point is reached, replacement or repair costs are incurred. For exanple,
if a typical coat of paint is 60 urn thick, the critical damage level at which repainting is
necessary may be about 50 urn. Monetary value is determined through economic damage functions
which may be developed from physical damage functions. This approach includes exposure, re-
placement cost, protection cost, and other data, but it cannot account for damage to irre-
XRD1C/B 1-43 3-3-81
-------
placeable items, such as works of art, where the only measurable cost is that of preservation.
However, only a few of the functions developed to date are relatively reliable in determining
damage and none has been generally accepted for estimating costs.
The best documented and most significant damage from sulfur oxides and particulate matter
involves: acceleration of metal corrosion; erosion and soiling of paint; and soiling of build-
ings and other structures. Erosion of stone and other building materials due to sulfur oxides
is also well established, but the importance of sulfur oxides relative to other pollutants is
not clear. Although evidence of damage to fibers (cotton and nylon), paper, leather and
electrical components has been reported, reliable damage functions have not been developed.
There are some general conclusions which may be drawn from studies discussed in more
detail in Chapter 10. As noted above, it has been clearly established that increases in
sulfur oxide concentrations accelerate corrosion. Table 1-3 displays damage functions
developed for effects of Stk on zinc, steel, and house paint. These equations and the data
from which they are derived show that temperature and total suspended particulate matter are
relatively unimportant factors in metal corrosion, and that the most important factor is
surface wetness (Schwartz, 1972, Barton and Bartonova, 1969, Sydberger and Ericsson, 1976, and
Haynie and Upham, 1974). Corrosion will not take place when the metal surface is not wet.
This dominating factor is usually approximated by a "time-of-wetness" term, that is, the amount
of time the relative humidity exceeds some critical humidity; critical humidities have been
identified for various metals. There are, of course, several sources of moisture: rain, snow,
fog, condensation; but relative humidity is the usual proxy for contribution of moisture to
the surface from all sources. Metal corrosion initiated by surface wetness is accelerated by
sulfur dioxide. Increase in either sulfur dioxide concentration or relative humidity is accom-
panied by increase in rate of corrosion. The relative importance of the two factors in cor-
rosion acceleration is shown in Figures 1-17 and 1-18, which reflect analysis of field data
by Haynie and Upham (1976). As shown in the figures, a 100 percent increase in average sulfur
dioxide concentration has about the same effect as a 10 percent increase in relative humidity
above a critical humidity. In some areas of the country (see Figure 1-19) this humidity level
is usually exceeded; in others, rarely.
This fact has obvious and major implications for the probability of producing an accurate
aggregated national estimate of damage to metals related to S02 exposure. Average annual
relative humidity can vary 10 percent even within one region of the country; for instance,
included in the data base for Figures 1-17 and 1-18 are average relative humidities of 29 and 39
percent for Las Vegas and Phoenix, respectively. The range in RH for all 57 sites (covering
34 states and the District of Columbia) was 29 to 76 percent. Average sulfur dioxide concen-
trations measured at these sites ranged from 9 ug/m3 to 374 ug/m3 during the same period. As
noted by Haynie and Upham (1976), this wide variation is useful and desirable in regression
analysis for development of damage functions, but is quite the opposite in estimates of
aggregate damage. Not only do both relative humidity and S02 concentrations vary spatially,
*XRD1C/B 1-44 3-2-81
-------
TABLE 1-3. SELECTED PHYSICAL DAMAGE FUNCTIONS RELATED TO S02 EXPOSURE
Material
Reference
Dose-Response relationships
Zinc
Haynie and Upham, 1970 Y = 0.001028 (RH - 48.8) SO,
0.92
Galvanized steel Haynie et al., 1976 corr = (0.0187 S02 + e 41>85 " 23,240/RT)t^ Q_gi
Galvanized steel Haynie, 1980
corr = 2.32 t, + 0.0134v0'781SO,t
W £* W
Oil-base house paint Spence et al., 1975
Y = 14.3 + 0.0151 S02 + 0.388 RH
0.61
tn
Enameling steel Haynie and Upham, 1974 corr = 325 t* e<°-00275 S02 " 163.2/RH)
Weathering steel Haynie et al., 1976 corr = [5.64 VSO + e(55'44 " 31'150/RT)]
0.91
corr = corrosion, urn
Y = corrosion/erosion rate, pm/yr
S02 \ig/n .
R - gas constant (1.98 cal/gm mol/ *'
RH = percent average annual relative humidity
t = time-of-wetness in years
v = wind velocity in m/s
t = trme of exposure, years
-------
100 160 200
S02 CONCENTRATION,
250
Figure 1-17. Steel corrosion behavior is shown as a function of average SO2 concentra-
tion at 65% relative humidity.
Source: Adapted from Haynie and Upham (1974).
1-45
-------
I
GC
100
90
80
70
60
50
V)
g 40
oc
D
Ul
£
30
10
I 1 I I I I
SOj CONCENTRATION.
I I
1 I 1 I I I
10 20 30 40 50 60 70 80
AVERAGE RELATIVE HUMIDITY, %
90 100
Figure 1-18. Steel corrosion behavior is shown as a function of aver-
age relative humidity at three average concentration levels of sulfur
dioxide. ,
Source: Haynie and Upham (1974).
1-47
-------
Figure 1-19. Annual mean relative humidity (RH) in various U.S. areas.
Source: Office of Air Quality Planning and Standards. Protecting Visibility: An
EPA Report to Congress, EPA-450/5-79-008, U.S. Environmental Protection Agency,
Research Triangle Park, N.C. October 1979.
1-48
-------
they also vary seasonally. These spatial and temporal variations, further, are not the same
across the country (See Chapter 5). In some areas, then, the highest S02 concentrations
coincide with periods of highest relative humidity; in other areas, the reverse is true.
Setting all these uncertainties aside, even if there were a means to predict with perfect
precision that X level of SOy would result in Y urn corrosion to a metal surface, it would
still be difficult to arrive at an acceptable aggregate damage estimate. This is true because
one would have to know both the total thickness of the metal surface in question, the critical
thickness below which repair or replacement is necessary, and the total area of surface ex-
posed. This information is not available. Various surrogates have been used, typically
annual production modified by some service life factor. These surrogates do not account,
however, for such influences as indoor versus outdoor use, use of protective coatings, or
subjective judgments as to the point at which the object in use should or could be repaired or
replaced. This latter judgment is influenced both by willingness to pay and ability to pay
and is tied to the economic status of the individual, the corporation, the region, or the
nation, as appropriate. All of these difficulties disctissed are reasons why the trend in more
recent attempts to relate atmospheric pollutant levels to economic damage has been toward
development of direct relationships between pollutant concentrations and economic benefit or
disbenefit.
The kinds of limitations in available data necessary to estimate total metal corrosion
associated with ambient concentrations of sulfur dioxide also exist for estimates of total
erosion of paint or building materials. Factors such as humidity, nature and extent of expo-
sure, and critical damage points are quantifiable only with a great deal of uncertainty for
the national case. Costs assigned to repair or replacement are often necessarily arbitrary.
Furthermore, existing damage functions for those materials are much less well documented than
those for metals.
The least reliable of the "significant" damage functions are those for soiling from par-
ticulate matter. Shown in Table 1-4 are the results of a regression analysis for soiling of
building materials, including paint, as a function of total suspended particulate exposure.
It is apparent from the results that reflectance, which formed the basis of the analysis, is
not the only property of particulate matter important to soiling. Also of importance is
particle size. The minimum deposition velocity for a particle is in a size range of 0.1 to
1.0 urn diameter (see Chapter 6). Those characteristics (stickiness, oiliness, tarriness)
which would increase adherence of deposited particles to a surface should also be considered.
There is at present no single technique which combines all relevant measurements: reflec-
tance, adherence, and particle size.
These limitations in physical damage estimates, as related to both sulfur oxides and
particulate matter, have presented major obstacles to accurate estimation of total material
damage and soiling by application of physical damage functions. Estimates of resultng econ-
omic damage based primarily on this approach have varied over a wide range, and have been
XRD1C/B 1-49 3-3-81
-------
TABLE 1-4. RESULTS OF REGRESSION FOR SOILING OF BUILDING MATERIALS AS A FUNCTION OF TSP DOSE
in
O
Material
Oil base paint
Tfnt base paint
Sheltered acrylic
emulsion paint
Acrylic emulsion
paint
Shingles
Concrete
Coated limestone
Uncoated limestone
Coated red brick
Uncoated red brick
Coated yellow brick
Uncoated yellow brick
Glass
400
400
400
720
48
160
80
80
80
80
80
80
45
89.43
86.13
91.54
90.79
43.50
41.75
44.57
46.99
12.95
14.88
45.05
43.21
0.2806
-0. 2768
-0.2618
-0.593
-0.4131
-0. 199
-0.0458
+0.0779
-0.0503
-0.0296
-0.0374
-0.1133
-0.1133
+0.0314
0.0641
0.0571
0.1156
0.0497
0.5771
0.1338
0.2464
0.1500
0.0223
0.0331
0.5337
0.2740
0. 008077
0.000069
0.000061
0.000123
0.000026
0.000258
0. 000080
0. 000164
0. 000089
0.000013
0. 000020
0.000317
0.000168
0. 000007
7.6510
6.8265
13.8143
8.3791
7.6992
7.5011
6.9046
4.2035
0.6255
0.9274
14.9533
7.6773
0.6851
0.745
0.738
0.880
0.902
0.769
0.143
0.347
0.266
0.459
0.477
0.342
0.503
0.340
.
Note: Equation used in this regression analysis was reflectance = B(TSP x months of exposure) + A.
N, Number of data sets (dependent upon the number of controlled variables in the factorial experiment).
Intercept of linear regression.
Slope of linear regression.
Estimated variance of intercept.
Estimated variance of slope.
Residual variance (error).
Correlation index (fraction of variability accounted for by regression).
A,
B,
S.
SD
Source: Abstracted from Beloin and Haynie, 1975.
-------
criticized both as underestimating total damage and overestimating it. For this reason, other
estimates of economic damage effects on materials and soiling related to ambient pollutant
concentrations have related ambient pollutant concentrations directly to economic benefit or
disbenefit. Such approaches are in the developmental stage and are limited by difficulties in
distinguishing the effects of one pollutant from another, and involve socioeconomic factors
that have yet to be dealt with satisfactorily. Though they show promise for future applica-
tion, to date these approaches have been found to be inadequate for decision-making guidance.
1.11 RESPIRATORY TRACT DEPOSITION AND FATE OF SULFUR OXIDES AND PARTICULATE MATTER
1.11.1 Exposures
Chapter 11 of the present document presents information on respiratory tract deposition
and fate of sulfur oxides and particulate matter useful in better understanding health effects
associated with such pollutants as determined by animal toxicological, controlled human
exposure and epidemiological studies. In animal laboratory or human clinical studies,
measurements of exposure levels can usually be made near the point of inhalation. Studies in
animals can often be used to determine the relationships between exposure levels and deposited
fractions or target-organ doses. The monitoring instruments used in laboratory and clinical
studies may, however, be very different from those used for environmental sampling in
epidemiological studies or for implementation of standards. These differences in exposure
characterizations have important quantitative effects upon the dose-response relationships
that may be derived from the different types of studies described in this report. They should
also be taken into account in developing exposure criteria and in specifying methods to be
used in complying with these criteria.
1.11.2 Deposition and Clearance
Removal of S02 by the upper airways during inspiration affects the penetration of S02 to
the tracheobronchial and possibly pulmonary regions of the lung. S02 removal by nasal
absorption, primarily under resting conditions, is nearly complete (95-99 percent) in both man
and laboratory animals (Frank et al., 1969; Speizer and Frank, 1966). S02 removal from the
respiratory tract during mouth breathing is significantly lower than during nose breathing,
although regional uptake has not been studied in man during mouth or oronasal breathing.
Since mouth breathing and higher airflow rates may be expected under exercise conditions, the
transition from rest to increased activity levels should significantly increase S02
penetration in man. These findings are in agreement with controlled human exposure studies
that have examined subjects under different breathing patterns (Lawther et al., 1975). Some
people are compulsorily or predominantly nose breathers; others favor breathing through their
mouths. However, most rely on both oral and nasal breathing under conditions of increased
respiratory workload.
Increasing the activity levels and respiratory rates of individuals can increase penetra-
tion of S02 into the trachea and bronchial airways. For example, animal studies show that
XRD1C/B 1-51 3-3-81
-------
absorption of S02 can be decreased to less than 50 percent by mouth breathing at relatively
high airflow rates. The presence of fine particles in inhaled air may result in gas-particle
interactions that also increase penetration of the reaction products into the lower
respiratory tract. Laboratory animals and people are similar in all of these aspects of SO^
deposition.
The majority of studies on the deposition of SO, in animals and people have been done at
3
concentrations greater than 2.62 mg/m (1 ppm). The high deposition of SO* in the upper
respiratory tract has not been confirmed at levels ordinarily found in ambient air [generally
less than 0.1 mg/m (0.038 ppm)]. It is anticipated, however, that similar deposition
patterns would be observed at these concentrations of SO^.
Of the total S0? inhaled, less than 15 percent is likely to be exhaled immediately. That
which is deposited is quickly absorbed into the secretions lining the respiratory passages.
Most of the sulfur is rapidly transferred into the systemic circulation from all regions of
the respiratory tract. Only small amounts have been observed to be exhaled at later times
(about 3 percent).
The deposition of inhaled particles in the respiratory tract is complex. Deposition in
different regions of the respiratory tract depends upon the physical properties and
aerodynamic diameters of the inhaled particles and upon breathing patterns. Detailed
information on regional deposition of inhaled particles for different modes of inhalation
cannot be summarized here and the reader is referred to Sections 11-1 and 11-2 in Chapter 11
of this document. These studies were made in individuals using a variety of air volumes and
flow patterns, so that the data relate to the resting state and states of light to moderate
exercise.
Particles inhaled through the nose have deposition patterns markedly different from
particles inhaled through the mouth. In nose breathing, most particles with greater than 4 urn
aerodynamic diameter are deposited in the respiratory tract (see Figure 11.3). With mouth
breathing, nearly complete deposition is observed only for particles greater than 10 urn
aerodynamic diameter (see Figure 11.4). Deposition of these size ranges is mainly in the
extrathoracic regions. However, 20 to 30 percent of particles between 5 and 10 urn aerodynamic
diameter inhaled with mouth breathing are deposited in the trachea and bronchial airways (see
Figures 11-7 and 11-8) and, with light activity levels, about 10 percent of particles as large
as 15 MID aerodynamic diameter are predicted to deposit in the tracheobronchial region. These
deposition patterns are depicted in Figure 1-20 in relation to varying physical or aerodynamic
particle diameters.
Inhaled particles with aerodynamic diameters less than about 4 urn have pulmonary
deposition fractions between 20 and 70 percent (see shaded area of Figure 1-20). About 20
percent pulmonary deposition is typical for these particle sizes when inhaled through the nose;
when particles between 2 and 4 urn aerodynamic diameter are inhaled through the mouth,
substantially greater (30-70 percent) pulmonary deposition results. For nose breathing, as
XRD1C/B 1-52 3-3-81
-------
en
ol_
0.1
"i—i—lUJ I III I I I I I II I I I I T
_^^^^^ * **»^. ACGIH CONV.
""~~~~~~^~~-^-^«^ *^ «•». — — «• BMRC CONV.
— BMRC CONV.
_._ STAHLHOFEN •»«!.
(19801
HHI PULMONARY VIA
^ MOUTH
"*>\ _,.-. PULMONARY VIA
^ NOSE
I -U-H-H
PHYSICAL DIAMETER
i | | A"""rl ( * '
0.2 0.3 0.5 0.7 1.0
I
23 5 7 10
-AERODYNAMIC DIAMETER,jum
30
Figure 1-20. Division of the thoracic fraction into the pulmonary and tracheobronchial fractions for
two sampling conventions (ACGIH and BMRC) as a function of aerodynamic diameter except below
0.5 Mm where particle deposition is plotted vs. physical diameter, from International Standard Organ-
ization ad hoc group to TC-146, 1980. Also shown are the band for experimental pulmonary depo-
sition data of Figure 11-9 and the tracheobronchial deposition data of one subject from Stahlhofen
et al. (1980).
-------
compared to mouth breathing, the peak of the pulmonary deposition curve shifts downward from
3.5 to 2.5 urn aerodynamic diameter. Depending upon the tidal volume and breathing frequency
used in various studies, pulmonary deposition of particles 5 urn in aerodynamic diameter can
vary from as little as 5 percent to as much as 50 percent. Also, with mouth breathing, about
5-13 percent of particles 8-9 urn in aerodynamic diameter are deposited in the pulmonary
region.
Regional deposition studies of particles less than 3 urn aerodynamic equivalent diameter
have been conducted using dogs and some rodents. In these species, the relative distribution
of these particles among the respiratory regions during nose breathing follows a pattern that
is similar to regional deposition in man during nose breathing. Thus, in this instance, the
use of rodents or dogs in toxicological research for extrapolation to humans entails
differences in regional deposition of insoluble particles less than 3 urn aerodynamic
equivalent diameter that can be reconciled from available data (see Section 11.2.1.5).
Although children are usually considered to be a subpopulation more susceptible to the
effects of environmental pollutants, deposition data for children are not currently available,
nor likely to be obtained soon. The little data that are available on other subpopulations,
such as asthmatics and chronic bronchitics, indicate that tracheobronchial deposition appears
to be enhanced at the expense of pulmonary deposition in most abnormal states.
Appropriate selective sampling procedures can and are being developed to provide more
meaningful data on inhalation hazard potential for particles as a function of their regional
deposition in the respiratory tract. Various sampler acceptance criteria reflective of
selective sampling procedures for various regions of the respiratory tract are shown in Figure
11-20. By taking into account the biological effects of the material and the population at
risk, air sampling procedures can be formulated which focus on the region(s) of the
respiratory tract pertinent to accurate health assessment (see Section 11.4 of Chapter 11).
Particles deposited in different regions of the respiratory tract are cleared by
different pathways and at different rates. Particles deposited in the anterior regions of the
nasal passages are cleared forward by nose blowing and sneezing. Beyond the middle turbinate
region, clearance to the pharyngeal regions occurs by mucociliary action, whereupon the
particles are swallowed. Likewise, most clearance of material deposited in the oral cavity is
by swallowing. All of these processes are relatively rapid and remove most of the deposited
material within minutes to hours.
Insoluble particles that deposit in the tracheobronchial region are cleared upward in the
respiratory tract by mucociliary action and then swallowed. This clearance is generally
complete within one or two days after particle deposition. In contrast, particles deposited
in the pulmonary region may be retained for several hundred days before they are cleared to
the conducting airways or to the pulmonary lymphatic system.
As the particles are cleared by mechanical processes from all regions of the respiratory
tract, chemical dissolution may remove soluble compounds which can then be absorbed directly
XRD1C/B 1-54 3-3-81
-------
into the systemic circulation. Since dissolution and absorption of substances from particles
deposited in the respiratory tract competes with mechanical clearance processes, the amount
absorbed depends upon the dissolution rate as compared to the mechanical clearance rate. The
absorbed fractions are markedly different for different regions of the respiratory tract
because of the large variations in clearance rates. Such variations may be calculated using
the equations seen on page 11-42 (see also Figures 11-12 and 11-13 of Chapter 11).
1.12 TOXICOLOGICAL STUDIES
Chapter 12 of this document discusses information derived from toxicological studies in
animals concerning the metabolism and effects of sulfur oxides and various forms of particu-
late matter. Such information is summarized next.
1.12.1 Metabolism of Sulfur Oxides and Particulate Matter
Although inhaled sulfur compounds are rapidly absorbed into the systemic circulation,
their main effect is observed in the respiratory tract. Prior to or during inhalation, S0?
* ^
may react with water to form sulfurous acid, or be oxidized to SO-. The latter reacts rapidly
with water to form sulfuric acid, which subsequently forms ammonium sulfate in the presence of
ammonia. The sulfurous acid readily dissociates to sulfite and bisulfite ions, which are in
rapid equilibrium. Bisulfite ions react with biological molecules by sulfonation, by auto-
oxidation and by addition to cytosine. Most of the inhaled S02 is presumed to be detoxified
by the sulfite oxidase pathway in the liver, which forms sulfate that can then be excreted in
the urine.
The metabolism of toxic substances that may be inhaled with atmospheric particles is
specific to the individual compounds. A discussion of the metabolism of all potentially
inhalable compounds in urban air is beyond the scope of this report. Detailed studies on the
deposition and clearance in laboratory animals of coal combustion products, automobile
exhausts, and silicates have been reported elsewhere in the scientific literature.
1.12.2 Effects of SO.,
A great number of studies have been conducted on the effects of exposing various species
of laboratory animals to differing concentrations of S02- Tables 1-5, 1-9, and 1-10 highlight
studies demonstrating various types of effects at different exposure concentrations. The
following discussion summarizes these findings and points out some of the inter- and
intraspecies variability observed.
Table 1-5 summarizes several animal studies on the effects of acute exposure to S02 on
pulmonary function. The most commonly observed response following 1 hour of exposure to S02
is bronchoconstriction. Only a few studies suggest acute exposure effects on pulmonary func-
tion at levels below 1 ppm, e.g., the increased airway resistance in guinea pigs reported by
Amdur et al. (1970, 1978a) at concentrations as low as 0.16 ppm. However, another study by
the same investigator (Amdur et al., 1978a) found no response at levels up to 0.8 ppm. It has
been hypothesized that SOy induces bronchoconstriction by interaction with a bronchial epi-
thelial receptor, which then initiates a reflex arc. This pathway is pharmacologically-
mediated by portions of the autonomic nervous system, particularly the vagus, and apparently
XRD1C/B 1-55 3-3-81
-------
TABLE 1-5. EFFECTS OF ACUTE EXPOSURES TO SULFUR DIOXIDE ON PULMONARY FUNCTION
Table
12-3
12-3
12-3
12-3
12-3
Concentration (ppm)
0.42 or 0.84 mg/m3 (0.16 or
0.32 ppm) S02
0.52, 1.04, or 2.1 mg/m9 (0.2,
0.4, or 0.6 ppm) S02
2.62, 5.24, 13.1, or 26.2 mg/m8
(1, 2, 5, or 10 ppm) S02
18 to 45 mg/m3 (7 to 17 ppm)
SO,
0, 44.5, 83.8, 162, 233, 322,
Duration
1 hr
1 hr
1 hr
1 hr
10 Bin
Species
Guinea pig
Guinea pig
Dog
Guinea pig
House
Results
Increase In resistance
No significant Increase In airway
resistance.
Increased bronchial reactivity to
aerosols of acetylcholine, a potent
bronchoconstrlctive agent
General decrease in tidal volume
and an increase In respiratory rate
Respiratory rate decreased proportion-
Reference
Amdur et al.,
Amdur et al . ,
Islam et al. ,
1970. 1978a
1978c
1972
Lee and Danner, 1966
Alarle et al.
, 1973d
12-3
519, or 781 mg/m3 (0, 17, 32,
62, 89, 123, 198, or 298 ppm)
S02
>50 mg/m3 (>19 ppm) S02
1 hr
Guinea pig
ally to the log of the concentration;
complete recovery within 30 min following
all exposures. The time for maximum
response was Inversely related to the log
of the concentration, being shortest at
highest concentrations
Increase in tidal volume and a decrease Lee and Danner, 1966
in respiratory rate
-------
involves the release of histamine (see Section 12.2.4). This hypothesis is supported
partially by investigations in man (see Section 1.13).
The respiratory response to histamine aerosols is similar, if not identical, to that of
SO*. If the SOg-initiated bronchoconstriction involves histamine release, as hypothesized
above, then the broad range of response of animals (and man) to different histamine concen-
trations could explain the similar broad concentration range of the S0?-bronchoconstriction
dose-response curve. As much as 200-fold differences in dose have been reported for
histamine-induced bronchoconstriction, for example. In addition, sensitivity to histamine may
decrease with age, depending upon the species involved. Further studies are needed to substan-
tiate this hypothesis and to determine if the variation in response to SO- represents sensi-
tive subpopulations.
Another alteration in breathing mechanics caused by S02 is a transient decrease in
respiratory rate. This may involve a chemoreceptor in the nasal passages, similar to the one
thought to be responsible for bronchoconstriction, which,is pharmacologically mediated by the
trigeminal nerve and which may involve the release of acetylcholine (see Section 12.2.4). The
S02~induced decrease in respiratory rate requires a higher concentration than, and differs
from, bronchoconstriction in several respects (see Table 1-5), among which are a concentration-
independent transience and a concentration-dependent period of desensitization.
The primary host defense mechanism of the respiratory tract is the clearance of foreign
objects from the lung, whether it be by mechanical (mucociliary transport) or biological
(phagocytosis or immunological function) means (See Section 12.2.5). The effects of S02 on
these mechanisms are variable and species-dependent (See Table 12-4). For example, rats
exposed to 0.1 ppm S02, 7 hours/day, 5 days/week, for 10 or 23 days exhibited accelerated
clearance of labeled particles. In the same study, however, 1.0 ppm exposures accelerated
clearance at 10 days, but depressed clearance rates after 25 days. In other studies, mucus
flow in the trachea of dogs was decreased following intermittent exposure for 1 year to 1 ppm
S02, whereas a single 30-minute exposure of 25 ppm did not affect clearance in donkeys. Also,
based on limited work with infectivity models, it appears that susceptibility to bacterial
infection is not affected by high concentrations of SO- (5 ppm up to 3 months). Antiviral
defenses were compromised by S02 in mice at 7-10 ppm for 7 days. Chronic exposure to S02,
however, can cause alterations of the pulmonary and systemic immune systems. In summary,
acute exposure to SO- can alter some aspects of host defenses, but concentrations in excess of
those currently found in the ambient air appear to be required. Unfortunately, few studies
have examined chronic effects on these parameters.
In regard to possible respiratory tract pathology, no remarkable alterations in lung
morphology were observed following chronic exposure of S0» in monkeys (0.14, 0.64, 1.28, and
5.12 ppm x 18 months) or dogs (5.1 ppm x 620 days) (see Section 12.2.3 and Table 12-2).
However, only conventional light microscopy was used, which is a method far less sensitive
than scanning or transmission electron microscopy for observing alterations in surface
XRD1C/B 1-57 3-3-81
-------
branes and cilia. Shorter exposures to much higher concentrations (10-400 ppm) generally did
cause morphological changes in mice, rats, and pigs (see Table 12-2).
The issue of mutagenesis is currently unresolved (see Section 12.5.2 and Table 12-15).
Although mutagenesis has been shown in two microorganisms in response to S02, no evidence
supports its occurrence in at least two higher systems, i.e., Drosophila and mouse oocytes.
With regard to the tumorigenic properties of S02, investigations of its jni vivo potential
oncogenicity are quite rare (see Section 12.5.3 and Table 12-16). Tumorigenesis after expo-
sure to S02, or to a combination of S02 and benzo(a)pyrene, has been examined in mice and rats,
respectively. In a single study involving S02, mice were exposed intermittently over an entire
lifetime. Increased incidence of primary lung carcinoma was reported for females, but not for
males. Because an adequate statistical analysis was not presented in the paper, a subsequent
statistical analysis was performed and revealed the increase in primary lung carcinoma in S0«-
exposed females to be significant (p = 0.011). Unfortunately, the exact duration of exposure
and concentration cannot be accurately determined from the publication.
Simultaneous exposure to S02 (4 ppm) and benzo(a)pyrene (10 mg/m ) for a lifetime was
subsequently reported for rats. The biological significance of these studies is complex and
difficult to interpret, particularly since statistical analyses were not given. However,
subsequent statistical analysis of the data reported for a combined exposure revealed the
increased incidence of lung tumors to be significant (p = 0.005), whereas exposures to S02 and
benzo(a)pyrene alone were not significant.
Numerous animal studies have investigated mortality induced by sulfur dioxide. However,
SOy causes high mortality only at high concentrations (>50 ppm) which are not relevant to
ambient air exposures (see sect. 12.2.3 and Table 12-1).
1.12.3 Effects of Particulate Matter
Characterizing exposures to particles in the atmosphere may be even more difficult than
characterizing exposures to S02- This is due, in large part, to the fact that the toxicity of
particulate matter depends greatly upon its chemical composition. In general, urban air is
quite heterogeneous in composition and may vary widely from one community to the next, or even
within a single community. It may contain both inert and chemically toxic fractions; the
potential impact of the latter is complicated by such considerations as dissolution, solu-
bility, and bioavailability (see Section 11.2.2). Although adequate physical and chemical
information can be obtained in studies with laboratory animals exposed to well-defined parti-
culate aerosols, the above types of data are not available for the heterogeneous mixture of
particles found in the environment. Therefore, the comparisons that can be made between toxi-
cological studies of defined, laboratory-produced aerosols and epidemiological studies off
people exposed to environmental aerosols is extremely limited.
A large number of studies have been conducted on the effects of exposure to sulfates and
certain other, well-defined particulate aerosols. Tables 1-6, 1-7, and 1-8 represent a sum-
mary of these studies demonstrating various types of effects at different exposure concentra-
XRD1C/B 1-58 3-3-81
-------
TABLE 1-6. RESPONSES TO ACUTE SULFURIC ACID EXPOSURE
t-1
$
Table
12-6
12-9
12-6
12-6
12-9
12-9
Concentrati on
100 Mg/m3
H2S04
500 ng/m3
H2S04
510 (jg/m3
H2S04
1000 'ng/m3
H2S04
190-1400 ug/m3
H2S04
1000 H9/m3
H2S04
Duration
1 hr
1 hr
1 hr
1 hr
1 hr
1 hr
Species
Guinea pig
Dog
Guinea pig
Guinea pig
Donkey
Dog
Results
Pulmonary resistance +47%
pulmonary compliance -27%.
Slight increases in tracheal
mucociliary transport velo-
cities immediately and 1 day
after exposure. One wk later
clearance was significantly
decreased.
Pulmonary resistance +60%
pulmonary compliance -33%.
Pulmonary resistance +78%
pulmonary compliance -40%.
Bronchial mucociliary , .
clearance was slowed.
Depression in tracheal
mucociliary transport
Reference
Amdur et al. , 1975,
1978b
Wolff et al., 1979a
Amdur et al . , 1978b
Amdur et al . , 1978b
Schlesinger et al., 1978
Wolff et al., 1979a
12-9 1400
H2S04
1 hr
Donkey
rate persisted at 1 wk
post-exposure.
No effect on tracheal
transport.
Schlesinger et al., 1978
-------
TABLE 1-7. RESPONSES TO CHRONIC SULFURIC ACID EXPOSURE
Table Concentration*
Duration
Species
Results
Reference
12-9
12-5
12-8
100
(0.3-0.6
H2S04
1 hr/day,
5 day wk,
several mo
Donkey
80 jjg/m3 (0.84 urn), 52 wks
100 M9/m3 cont.
(2.78 Mm)
H2S04
Guinea pig
Within the first few wk
all four animals developed
erratic bronchial muco-
ciliary clearance rates,
either slower than or faster
than those before exposure.
Those animals never pre-
exposed before 100 (jg/m3
H2S04 had slowed clearance
during the second 3 mo of
exposure.
No significant blood
effect; no lung alter-
ations; no effect on
pulmonary function.
Schlesinger et al., 1978
Alarie et al., 1973a,
1975
12-5 380 M9/in3 (1.15 M«I); 78 wks
12-8 480 M9/m3 cont.
(0.54
H2S04
Monkey
No significant blood
effect; no lung changes
380 M9/m3 increased respi-
ratory rate; 480 M9/m3
altered distribution of
ventilation early in exposure
period but not later.
Alarie et al., 1973a
12-13
900 M9/m3
(2.78
H2S04
12 mo
cont.
Guinea pig No significant effects on
hematology, pulmonary
function, or morphology.
Alarie, 1975
*A11 particle sizes in MMD
-------
TABLE 1-8. RESPONSES TO VARIOUS PARTICIPATE MATTER MIXTURES
Table
12-6
12-6
12-6
12-6
12-10
12-13
Concentration
100 ug/m3
open hearth dust
500 pg/rn3
(NH4)2S04
750 ug/m3
Na3V04
1000 ug/m3
FeS04
1000 ug/m3
MnCl2
500-5000 ug/m3
Mn304
1400 ug/m3
H2S04 +
1500 (jg/m3
carbon or
Duration
1 hr
1 hr
1 hr
1 hr
1 hr
2 hr
3 hr/day
5 day/wk,
20 wks
Species
Guinea pig
Guinea pig
Guinea pig
Guinea pig
Guinea pig
Mouse
Mouse
Results
Pulmonary resistance +9%
pulmonary compliance 0%.
Pulmonary resistance +23%
pulmonary compliance -27%.
Pulmonary resistance +7%.
Pulmonary resistance 2%.
Pulmonary resistance +4%.
The aerosols increased the
mortality from the subsequent
standard airborne strepto-
coccal infection: with Mn304
occurrence at 1550 vg/m3 Mn.
Altered the immune system.
Morphological changes
observed; more severe with
carbon only exposure.
Reference
Amdur and Underhill,
1968, 1970
Amdur et al . , 1978a
Amdur and Underhill,
1968
Amdur and Underhill,
1968
Amdur and Underhill,
1968
Gardner et al. , 1977b
Adkins et al. , 1979,
1980c
Renters et al . , 1979
1500 ug/m3
carbon only
12-13 1100 ug/m3
H2S04, or
1500 ug/m3
carbon or
in combination
3 hr
Hamster Carbon caused no change in
ciliary beat frequency.
Ciliary beat frequency was
depressed after H2S04 ex-
posure. The combination
produced similar effects,
but recovery had occurred by
48 hr. post-exposure. Up to
48 hr. after exposure H2S04
+ carbon resulted in more
tissue destruction than either
pollutant alone.
Schiff et al., 1979
-------
tions, durations and chemical composition. The following discussion highlights some of the
variability in response that has been observed.
Oxides of sulfur that are of air pollution significance or have been used in inhalation
studies as aerosols include: sulfuric acid (H-SO.), ammonium bisulfate and sulfate, and
sulfate salts of zinc, iron, copper, manganese and others.
Alterations of pulmonary function, particularly the increase in pulmonary flow resis-
tance, give the measure of response to acute exposure to sulfur oxide aerosols (see Section
12.3.3.1). Reports are variable regarding the irritation potency of various sulfate salts.
However, according to many investigators, sulfuric acid appears to be most irritating. Table
1-6 shows that, for short-term exposures, the lowest concentration of sulfuric acid found to
o
increase airway resistance was 100 ug/m (one-hour exposure of guinea pigs). Experiments with
various exposure protocols showed particle size to be an important factor; around 2 urn or
smaller sizes were generally more effective in producing the observed effects. Perhaps due to
variation in animal species and strains, and different particle sizes used in the experiments,
the effects noted with sulfuric acid and sulfate salts are often contradictory. Guinea pigs
are the most sensitive and show severe bronchoconstriction in response to sulfuric acid aero-
sols. Based on short-term (one-hour exposure) effects, a ranking of irritation potency (in
terms of increased airway reisitance) has been made for various sulfur oxides as listed below:
Relative Irritant Potency of Sulfates in Guinea Pigs
Exposed for One Hour (Amdur et al., 1978a)
Sulfuric acid 100
Zinc ammonium sulfate 33
Ferric sulfate 26
Zinc sulfate 19
Ammonium sulfate 10
Ammonium bisulfate 3
Cupric sulfate 2
Ferrous sulfate 0.7
Sodium sulfate (at 0.1 urn) 0.7
Manganous sulfate -0.9
aData are for 0.3 urn (HMD) particles. Increases in airway resistance were
related to sulfuric acid (0.413. increase in resistance per ug of sulfate
as sulfuric acid) which was assigned a value of 100.
The irritation potency of sulfuric acid aerosols is complicated by its partial neutrali-
zation by ammonia present in the breath or in the air of animal exposure chambers. The
resulting (NH4)2S04 and NH4HS04 lessen the expected effects of H2S04, but the extent to which
this affects the results of available animal studies cannot be quantified.
As summarized in Tables 1-7 and 1-8, chronic effects of sulfuric acid and sulfate salts
are less certain than acute exposure effects (see Section 12.3.3.2). Exposure to a concentra-
3
tion of 0.38 mg/m sulfuric acid for 78 weeks produces no pulmonary function changes in
monkeys, but measurable changes occurred at 0.48 to 4.79 mg/m concentrations. Concentration,
*XRD1C/B 1-62 3-2-81
-------
particle size, and chemical composition all appear to be important for pulmonary function
changes, as suggested by information in Tables 1-7 and 1-8 (as well as Table 12-8).
Exposure to sulfuric acid aerosol causes an alteration of mucociliary clearance of viable
and nonviable particles from the lung (see Section 12.3.4.1). The effects observed are vari-
able; e.g., tracheal mucociliary transport in dogs was increased after exposure to H?SO. aero-
sol (0.5 mg/m for one hour), but was decreased in rats and hamsters after one to three hour
exposure to 1 mg/m ^SO^. Prolonged exposure of donkeys to similar concentration of H2$04
caused a persistent slowing of mucociliary transport (see Table 12-9).
Resistance to bacterial infection was not affected by this sort of exposure to H^SO, aero-
sol. However, various metal sulfates adversely affected this defense mechanism (see Table
12-10). The potency of these metal sulfates, based on a three-hour exposure causing an in-
creased susceptibility to bacterial infection, may be ranked as: CdSO. > CuSO. > ZnSO. >
A12(S04), > A1 NH/,(SO.)2. Sulfuric acid and the following sulfates at concentrations greater
than 2.5 mg/m were ineffective in this bacterial infection model: (NH4)2S04, NH.HSO.,
Na2SO., Fe2(SO.)_ and Fe(NH.)2(S04)2. It should be noted, however, that various non-sulfate
metallic aerosols, especially Ni- or Cd-containing compounds, have substantial inhibitory
effects on host defense mechanisms in general. A thorough discussion of this may be found in
Sections 12.3.4.2, 12.3.4.3, and 12.3.4.4 of Chapter 12 of this report.
Morphological changes in the lung have been studied mostly after chronic exposure to
sulfuric acid (see Section 12.3.2). Morphological changes were evident, such as in monkey
lung, after a long-term (78 wk) exposure to relatively high levels of sulfuric acid (2.43
3 3
mg/m , 3.6 urn, aerodynamic diameter to 4.79 mg/m , 0.73 urn aerodynamic diameter). The major
findings included thickening of the bronchial wall and bronchiolar epithelium, which may con-
tribute to the changes in lung function. In other studies involving guinea pigs (0.1 mg/rn
a
H2S04 for one year) and dogs (0.89 mg/m H2SO. for about two years), neither morphologic nor
physiologic changes were noted (see Table 12-5).
The lethal effects of sulfate aerosols are dependent on animal age (see Section 12.3.1).
For example, 18 mg/m was lethal to 1-2 month old guinea pigs as opposed to 50 mg/m for 18
month old animals. Particle size is also important; LC5Q for guinea pigs was 30 mg/m with an
aerosol size of 0.8 urn (aerodynamic equivalent diameter) as opposed to 109 mg/m with an
aerosol size of 0.4 urn (aerodynamic equivalent diameter). Bronchial spasm may be the najor
cause of animal death.
Suspended particles not related to sulfur oxides are also of considerable concern.
However, because of the wide variety of such substances, it is difficult to summarize perti-
nent toxicological results. Information on the inhalation toxicology of several individual
substances found in ambient air particulate matter is summarized in Table 1-8. Other relevant
information can also be derived from non-inhalation toxicological studies. For exanple,
numerous trace metals have been found as components of airborne particulate matter (see
Section 12.5.4). In addition to being generally toxic, certain compounds of some of these
XRD1C/B 1-63 3-3-81
-------
metals, including beryllium, cadmium, cobalt, chromium, iron, lead, nickel, titanium, and
zinc, have been identified as carcinogenic under specific, non-respiratory laboratory exposure
conditions.
Silicon is ubiquitous in the earth's crust and in coarse mode particles. Silicon dioxide
(SiOp), which is responsible for the disease silicosis, is found in 3 crystalline forms
(quartz, Cristobalite, and tridymite). As a generalization, the ranking of toxicity is tri-
dymite > Cristobalite > quartz. These uncombined forms of Si02 are generally called "free
silica." Si02 is also found combined with cations, in which case the term silicates is
applied. Very few animal toxicological studies of silicates exist. Several hypotheses of the
etiology of silicosis have been developed, but no single one has been proven definitively.
Although many animal toxicological studies of Si Op exist, comparisons are difficult
because of the species and strain of animal used, accidental infections, the size of Si02
particle used, and the crystalline form of Si02 used. Silicosis, similar to that observed in
man, has been produced in animals exposed to high concentrations of quartz and other SiO-
dusts via intratracheal instillation (30-50 mg) or chronic inhalation. Chronic exposures (2.5
yr) of dogs to earth containing 61 percent cristobalite produced fibrotic nodules in hilar
lymph nodes, but not the lungs.
1.12.4 Effects of Complex Mixtures
It is difficult to assess with accuracy the toxicity of complex sulfur-containing aero-
sols in u.rban atmosphere based simply upon the sulfuric acid or sulfate content. The chemical
composition of sulfate aerosols, particularly the metallic or cationic counterparts, are
important in determining their relative toxicity. Since atmospheric aerosols may contain
varying proportions of sulfuric acid and ammonium- and metal sulfates, it is not possible to
extrapolate from animal toxicological data obtained with single compounds to human environ-
mental situations.
Exposure to SCL together with liquid or solid aerosols, which may act as a carrier for
the gas, seems to enhance the toxic effects of S02 in some cases (see Section 12.4). Table
1-9 shows some examples of this, where the aerosols are solutions of a salt such as NaCl,
MnCl2, or FeSO^. Although the evidence is not clear, additive as well as synergistic effects
have been observed by some investigators using H2$04 and ozone (see Table 1-10), or S02 to
H2S04 may have increased the toxic potency. On the other hand, it may be seen from Table 1-11
that the addition of fly ash to mixtures of S02 and H2SO. had no significant effect.
1.13 CONTROLLED HUMAN STUDIES
Human experimental studies of the health effects of exposure to specific pollutants in
the ambient environment are of significance in scientific assessments of air pollution risks
since they can demonstrate relationships between pollutant exposure and short-term health
effects. Such studies utilizing man require strict controls so that their findings are rele-
vant to larger populations. To date no human study of the health effects of exposure to
*XRD1C/B 1-64 3-2-81
-------
TABLE 1-9. RESPONSES TO ACUTE EXPOSURE COMBINATIONS OF S02 AND SOME PARTICULATE MATTER
Table
Concentration
Duration
Species
Results
References
12-11
1.0 ppm S02 +
1 hr
Guinea pig
No increase in
pulmonary flow
McJilton et al,
1973
NaCl Aerosol
(<40% RM end
>80% RH)
resistance at low RH; at high
RH, potentiation was marked
and evident during both early
and late parts of exposure.
12-11 1.0 ppm S02 + 1 hr
Aerosols of salts
12-11 2.0 ppm S02 + 1 hr
NaCl Aerosol
en
01
Guinea pig Presence of soluble salt
increased pulmonary flow
resistance about three fold.
The potentiation was evident
early in the exposure.
Guinea pig S02 alone produced an
increase of 20% in pulmonary
flow resistance; with 10 mg/m3
NaCl the increase was 55% and
the potentiation occurred later
in exposure; with 4 mg/m3
potentiation was reduced.
Amdur and Underhill, 1968
Amdur, 1961
-------
TABLE 1-10. RESPONSES TO ACUTE EXPOSURE COMBINATIONS OF SULFURIC ACID AND OZONE
Table Concentration
Duration
Species
Results
Reference
12-14
§ 12-14
880 M9/BI3
H2S04 +
0.1 ppm 03
12-14 900
H2S04 +
0.1 ppm 03
1000 ug/m3
H2S04
0.4-0.5 ppm 03
3 hr 0-
2 hr H2S04
3 hr 03
2 hr H2S04
3 days
cont.
Hamster
Mouse
Rat
Gardner et al., 1977a
H2S04 depressed ciliary beat Grose, 1980
frequency. By 75 hr after
exposure, recovery had occurred
03 exposure had no effect.
Sequential 03 then H2S04
exposure decreased ciliary beat
frequency significantly but to a
lesser extent than that caused
by H2S04 alone.
In response to air-
borne infections a
significant increase in
mortality only when 03 was
given immediately before
exposure to H2S04, and the
response was additive.
Synergistic effects.
Glycoprotein synthesis was
stimulated in tracheal ring
explants; lung DMA, RNA and
protein content increased.
Last and Cross, 1978
-------
TABLE 1-11.
PATHOLOGICAL RESPONSES FOLLOWING CHRONIC EXPOSURE TO SO,
IN COMBINATION WITH PARTICULATE MATTER i
ALONE AND
Table
12-2
Concentration
10 pp» S02
Duration
72 hr cont.
Species
House
Results
Pathological changes In the
nasal mucosa appeared after
Reference
GHddens and Fair-child,
1972
12-2 0.14, 0.64, 1.28
PP» S02
12-13 0.11, 1, 5 ppm
SO, + 560
(jg/m3 fly ash
12-13 1.0 ppa S02 +
H2S04 +
fly ash
78 wk cont.
78 wk
cont.
18 no
Cynomolgus
monkey
Monkey
Honkey
12-13 0.11, 1, 5 ppm
S02 + 560
ug/H3 fly ash
12-13
6.1 ppm SO,
+ 890 M9/«i
H2S04
52 wk
cont.
21 hr/day
620 days
Dog
24 hours and Increased in
severity after 72 hours.
Nice free of upper respiratory
pathogens were significantly
less affected than the conven-
tionally raised animal. Morpho-
logical alterations were quali-
tatively Identical in both groups.
No remarkable morphologic
changes in the lung.
No effects on morphology.
No significant effects on
henatology or pulmonary
function tests during exposure.
At end of exposure to 0.99 ppm
S02 + 930 ug/m3 H2S04 (0.5 urn,
HMD) lungs had morphological
alterations In the bronchial «
mucosa. Exposure to 1.01 ppm
S02 + 880 ug/m3 H2S04 (0.54
Mm, HMD) + 410 pg/in3 fly ash
(4.1 MR, MHO) had similar
alterations; thus fly ash did
not enhance effect. Exposure
to 990 M9/m3 H2S04 Q.64 urn.
HMO) + 550 pg/n3 fly ash (5.34
Mm, HMD) had slight alterations.
Alarie et al., 1972, 1973c
Alarie et al., 1973b
Alarie et at.. 1975
Guinea pig No effects on morphology.
Alarie et al., 1973b
After 225 days, dogs
receiving H2S04 had a lower
diffusing capacity for CO than
those that did not receive
H2S04. After 60 days, no
morphological changes occurred.
H2S04 decreased net lung volume
and total weight.
Lewis et al., 1969, 1973
-------
sulfur oxides and particulate matter meet all the requirements of strict controls. Some basic
information, however, can be garnered from many studies that have been published.
1.13.1 Effects of Sulfur Dioxide
S0« has been found to have effects on several physiologic functions. Through subjective
reports, the reliability of which has been questioned, a level of 5 ppm has been established
for detecting S0?, with considerable variation below that level. Several sensory processes
are affected by generally agreed-upon levels of concentration of SO^. The odor threshold
averages 0.8 to 1 ppm, with 0.47 ppm set in one study performed under ideal conditions. The
sensitivity of the eye to light increases at 0.34 to 0.63 ppm, is maximal at 1.3 to 1.7 ppm,
o
and decreases to normal by 19.2 mg/m during dark adaptation. During light adaptation, the
figures increase and decrease similarly but at slightly higher levels of exposure. The alpha-
o
wave has been found to be attenuated by 0.9 to 3 mg/m SO- during 20 seconds of exposure.
Studies of the effects of S02 on the respiratory system of the body have arrived at con-
flicting conclusions (see Table 1-12). Although two studies found respiratory effects after
exposure to as little as 0.75-1 ppm S02 (Bates and Hazucha, 1973; Amdur et al., 1953), others
could find no effect below 5 ppm for normal healthy subjects at rest. At the latter level,
pulmonary flow resistance increased 39 percent in one study. Respiratory effects have been
found to be proportional to the concentration of S0« to which study subjects are exposed.
Although the bronchoconstrictive effects of exposure to SO^ have been found to be fairly con-
sistent, subjects vary considerably in response to exposures, and there are some especially
sensitive subjects, which may represent as much as 10 percent of the population. Recent
studies have suggested that asthmatics may be particularly sensitive.
In asthmatic subjects, mid maximum expiratory flow rate (MMFR) was significantly reduced
3
after oral exposure to 1.31 mg/m (0.5 ppm) SO* for 3 hours (Jaeger et al., 1979). As noted
in Table 1-12, three of the subjects (2 asthmatics, 1 normal) incurred delayed effects
(wheezing attacks at night) that may have been related to the exposure. Recent studies by
Sheppard et al. (1980, 1981) and Koenig et al. (1980, 1981) have demonstrated pulmonary
function changes observed in asthmatic subjects both at rest and during exercise. In the
Sheppard et al. studies, at rest SRaiJ increased significantly at SO, concentrations of 1, 3, 5
3W £,
ppm for asthmatics and only at 5 ppm for normal and atopic (viral rhinitis) subjects. During
exercise specific airway resistance (SR ) significantly increased in the asthmatic group at
aW
0.5 and 0.25 ppm S02 and even at 0.1 ppm in the two most responsive subjects. At 0.5 ppm,
three of the subjects experienced wheezing and shortness of breath and at 1.0 ppm, all six
subjects experienced such symptoms. Sheppard et al. concluded that with resting subjects, the
exaggerated bronchoconstriction produced by SO- in persons with asthma and the lesser broncho-
constriction produced in normal persons both appear to be mediated by parasympathetic neural
pathways. Under S02 exposure concentrations more likely to be encountered in polluted cities
(0.1 0.5 ppm S02) they also demonstrated that exercise modifies the bronchoconstriction
*XRD1C/B 1-68 3-2-81
-------
TABLE 1-12. CONTROLLED HUMAN EXPOSURE STUDIES - MAJOR STUDIES CITED IN CHAPTER 13
Concentration Duration of Number of
(ppm) exposure (mins) subjects
S02 (5 - 30 ppm)
3
S02 (2.5, 5.0, 10.0 ppn)
SO. (5 ppn)
£
S02 (1.1 - 3.6 pp»)
v S02 (1 - 8 ppn)
-------
TABLE 1-12. (continued)
Concentration
(PP-)
S02 (0.1, 0.25, 0.5,
SO, (0.15 ppm)
and
03 (0.15 ppm)
Oral or
Duration of Number of nasal Rest (R) or
exposure (mlns) subjects exposure exercise (E)*
10 7, 6 0 E
asthmatics
120 6 E* -
Effects Reference
SR significantly increased Sheppard, et
in the asthmatic group at
0.5 and 0.25 ppm SO. and
even at 0.1 in the two most
responsive subjects. At 0.5
ppm, 3 subjects experienced
wheezing and shortness of
breath. At 1 ppm all six sub-
jects experiences these symptoms.
Significant enhanced Kagawa et al.
decrease in SG after
exposure to SOj - 0, in
al., 1981
, 1979
0.35 - 5.0 mg/m3 H,SO.
HMO 1 urn * *
H2SO. (0, 100,, 300,
Or 1,000 ug/nT
HHAD 0.5 urn
1.9)
SO, (1-60 ppm) plus
H,0, to form H,SO,
aerosol * *
CMD 1.8 and 4.6 uo
15
60
Variable
15 Mask (rest)
10
Nasal
24
(Rest)
Respiratory rates increased,
max. insp. and expiratory
flow rates and tidal
decreased volumes
No pulmonary function
effects
Broncial mucociliary
clearance t following
100 ug/m%but * following
1000 Mg/m • Mucociliary
clearance distal to trachea
more affected
Airway resistance
increased especially
with larger particles
Aadur et al., 1952
Lippmann et al., 1980
Toyama and Nakamura,
1964
-------
*
produced by S02 in subjects with mild asthma. However, extrapolation of these observed
quantitative exposure/effect relationships to what might be expected under ambient condi-
tions could be questionable because of the use of forced mouth breathing with nosedips
and rapid step-function changes in S02 exposure concentrations.
Because S(>2 is readily water soluble and the nasal and mouth passages differ in available
moist surface area, the route of exposure will affect the response of individuals. Subjects
report less throat and chest irritation when breathing through the nose, and pulmonary flow
resistance increases are less in subjects who are nose breathing. Regardless of the route of
exposure, 5 ppm S02 had limited effects on specific airway conductance (airway bronchoconstric-
tion), although higher levels had a dose-dependent effect; that is, higher concentrations de-
creased SG more than lesser concentrations. The average decrease was greater after oral
exposure than after nasal administration.
The level of activity of the subjects tested affects the results because the actual dose
delivered to lungs and airways is greater when subjects'breathe through their mouth, as during
exercise. Just having subjects breathe deeply through the mouth significantly affected speci-
fic airway resistance during exposure to 1 ppm SOp in one study, although another study found
no such effect. Respiratory effects of exposure either by nose or by mouth are greatest after
5 to 10 minutes of exposure. Recovery takes about 5 minutes in normal subjects, but much
longer (10 to 60 minutes) in sensitive subjects and those who are asthmatic. Studies of nasal
mucus flow rates and airway resistance following about 6 hours of exposure to 1 and 5 ppm SO-
per day for 3 days found some effects maximal after 1 to 6 hours.
An early study found mucus clearance increasingly reduced as length and concentration of
exposure to S02 increased. Long exposures to 5 ppm S02 increased mucociliary clearance in one
study; a decrease had been found in nasal clearance rates in another study. Available studies
have not found a significant interaction of smoking with S02>
1.13.2 Effects of SOp in Combination with Particulate Hatter
Particulate matter has been shown to be potentially important in enhancing the effects of
S02 exposure. Airway resistance increased more after combined exposure to S02 and sodium
chloride in several studies, although others have failed to reach the same conclusions.
MEF50% (maximal expiratory flow rate at 50% vital capacity) was found to be significantly
reduced after exposure to a combination of saline aerosol and 13.1 mg/m (5 ppm) S02 (Snell
and Luchsinger, 1969). Most recently (Koenig et al., 1980, 1981), studies have been reported
on pulmonary function changes observed in extrinsic asthmatics both at rest (1980) and during
exercise (1981) with exposure to 2.62 mg/m3 (1 ppm) S02 and 1 mg/m3 NaCl (See Section 13.4).
Significant decreases in Vmax50^ and Vmax75«g were observed under aerosol conditions both at
rest and during exercise for asthmatics but not for all normals. However, NaCl alone did not
produce such effects, suggesting that either the exercise or the combination of S0£ and NaCl
was important.
XRD1C/B 1-71 3-3-81
-------
1.13.3 Effects ofCombined Sulfur Dioxide-Ozone Exposure
Previous studies Involving sulfur dioxide and ozone together support the view that ozone
is more toxic at a given concentration. Whether sulfur dioxide or its reaction products ever
exacerbate the irritant response to ozone is still unresolved. For most studies involving S02
(0.37 ppm) and ozone (0.37 ppm) there are conflicting reports with respect to lung function
decrements and synergistic effects of the combined pollutants. Recently, however, Japanese
studies (Kagawa and Tsuru, 1979) involving a combination of SO, exposure at 0.15 ppm (393 ug/m )
3
and 0, at 290 ug/m (0.15 ppm) have observed a significantly enhanced decrease in
specific airway conductance (SG ) compared to the decrease in SG observed in these subjects
3W clW
when exposed to 0,, alone. They suggest that the effect observed during exercise was
synergistic and not just additive and refute the idea that sulfuric acid formation was
responsible for the marked effects of 0, in the presence of S02 (Section 13.4).
1.13.4 Effects of Sulfate Aerosols
Sulfuric acid and sulfates have been found to affect both sensory and pulmonary function
3
in study subjects. The odor threshold for sulfuric acid aerosol has been set at 750 ug/m in
one study and 3000 ug/m in another. Light sensitivity has been found to be consistently
3 3
increased by 25 percent at 700 to 960 ug/m concentration of sulfuric acid mist (300 ug/m ).
Optical chronaxie has also been found to be increased after exposure of subjects to 750 ug/m
sulfuric acid mist.
3
Respiratory effects from exposure to sulfuric acid mist (350 to 500 ug/m ) include
increased respiratory rate and decreased maximal inspiratory and expiratory flow rates and
tidal volume. Several studies of pulmonary function in normal subjects indicated that
pulmonary function was not affected when subjects were exposed to 100-1000 ug/m sulfuric acid
for 10 to 60 minutes, although in one study the bronchoconstrictor action of carbachol was
potentiated by the sulfuric acid and sulfate aerosols more or less in relation to their
acidity. In studies with asthmatic subjects, generally no changes in airway function have
been demonstrated after exposure to sulfuric acid and sulfate salts at concentrations less
3 3
than 100 ug/m . However, at higher concentrations (1000 ug/m ) reduction in specific airway
conductance and FEVj have been observed after H2S04 and NH^HS04 exposures. Mucociliary
clearance was affected by exposure to sulfuric acid, being significantly increased after
3 3
exposure to 100 um/m and significantly decreased after exposure to 1000 ug/m . Another study
found no pulmonary effect of exposure of normal and asthmatic subjects to sulfuric acid,
ammonium bisulfate, and ammonium sulfate.
In summary, the available evidence generally suggests that thresholds of detectable
effects are somewhat above maximum likely ambient exposure concentrations for sulfate
compounds. Also, in studies with asthmatic subjects, generally, no changes in airway function
have been demonstrated after exposure to sulfuric acid and sulfate salts at concentrations
3
less than 100 ug/m . However, reduction in specific airway conductance (SG=1J and FEV, n have
aW Q . J. • U
been observed after HgSO^ and NH^HSO^ exposures at higher concentrations (1000 ug/m).
XRD1C/B 1-72 3-3-81
-------
1.14 EPIDEMIOLOGICAL STUDIES
Chapter 14 of this document discusses epidemiologies! studies on the effects of sulfur
oxides and particulate matter. Some of the epidemiological studies reviewed appear to provide
meaningful quantitative information on health effects associated with ambient air exposures to
PM and S02- Others, however, do not meet as fully various objectives regarding study design
and analysis to allow for quantification of exposure/effect relationships, or ambiguity exists
regarding clear interpretation of their reported results. Only relatively few of the study
results can, therefore, be accepted with a relatively high degree of certainty or confidence,
whereas others may be seen as providing, at best, only suggestive evidence for reported asso-
ciations between air pollutant parameters and health effects. The main focus of the present
section will be on summarizing results and conclusions derived from selected key studies
having a relatively high degree of certainty associated with their findings.
In general, the epidemiological studies reviewed provide evidence for severe health
effects, such as mortality and respiratory disease, being associated with marked elevations of
atmospheric levels of sulfur dioxide and particulate matter for certain populations at special
risk. Those populations at special risk for such effects appear to include, mainly, the
elderly and adults with chronic pre-existing cardiac or respiratory diseases (e.g.,
bronchitics). Increased lower respiratory tract illnesses and more transient (likely
indicator) effects, e.g., decrements in pulmonary function, also appear to be associated for
children with lower chronic exposures to sulfur dioxide and PM. Also, some qualitative epi-
demiological evidence suggests that asthmatics may be a susceptible population at special risk
for experiencing pulmonary function decrements in response to elevations in SO,, and PM.
1.14.2 Health Effects Associated with Acute Exposures to Sulfur Oxides and Particulate Matter
As discussed in Chapter 14, it is widely accepted that increases in mortality occur when
either sulfur dioxide (S0«) or particulate matter (PM) levels increase beyond 24-hr levels of
•a £•
1000 ug/m . Such increased mortality, mainly in the elderly or chronically ill, may logically
also be most directly attributed to very high short-term peak levels in the pollutants, which
at times increased to several thousand ug/m during certain major pollution episodes.
Much more difficult to establish is to what extent significant increases in mortality and
morbidity are associated with exposures to S02 and/or particulate matter levels below 1000
ug/ra3. Concisely summarized in Table 1-13 are several key studies that appear to demonstrate
with a reasonably high degree of certainty, mortality and morbidity effects associated with
acute exposures (24 hrs) to these pollutants. The first two studies cited, by Martin and
Bradley (1960) and Martin (1964), deal with a relatively small body of data from London in the
late 1950s. No clear "threshold" levels were revealed by their analyses regarding S02 or BS
levels at which significantly increased mortality began to occur. Based on their findings,
there appears to be little question that mortality in the elderly and chronically ill was
elevated in association with exposure to ambient air containing simultaneous S02 and BS levels
XRD1C/B 1-73 3-3-81
-------
TABLE 1-13. SUMMARY OF QUANTITATIVE CONCLUSIONS FROM EPIDEMIOLOGICAL STUDIES RELATING HEALTH
EFFECTS OF ACUTE EXPOSURE TO S02 AND PARTICULATE MATTER TO AMBIENT AIR LEVELS
Type of study
Effects studied
24-hr average pollutantlevel
particulate matter
CoH BS TSP
(Mfl/m3)
SO,
Reference
Mortality
Likely increases in daily
total mortality above a
15-day moving average during
winter 1958-59 among persons
with existing respiratory
or cardiac disease in London.
>1000
>1000
Martin and
Bradley (1960)
Slight indication of likely
increases in daily total
mortality above a 15-day
moving average during winters
of 1958-59 and 1959-60 among
persons with existing respira-
tory or cardiac disease in
London.
750-1000
710-1000
Martin (1964)
Morbidity
Likely worsening of health
status among a group of
chronic bronchitis patients
250-500
500-600
Lawther et al.
(1958, 1970)
No apparent response or
worsening of health status
among a group of chronic
bronchitis patients
<250
<500
Lawther et al.
(1970, 1975)
-------
somewhere in the range of 500-1000 ug/m3. Greatest certainty applies for levels in excess of
700-750 ug/m • Much less certainty is attached to lower estimates possibly derived from a
reanalysis of the same data set by Ware et al. (1981), which applies to mortality data from
very brief periods during the two London winters. It seems more likely that levels well in
excess of 500 ug/m BS and S02 are typically necessary in order to induce mortality among
highly susceptible elderly or chronically ill individuals.
Only very limited data exist by which to attempt to delineate any specific physical and
chemical properties of PM associated with the observed increases in mortality. Based on
information summarized in Section 1.3, it would seem that marked increases in small particles
3
to levels above 500-1000 ug/m appear to be most clearly associated with increased mortality,
based on the BS aerometric measurements reported, although contributions from larger
coarse-mode particles cannot be completely ruled out. It is not possible to state with
certainty specific PM chemical species associated with the increases in mortality. We do know
that large amounts of pollutants (e.g., elemental carbon, tarry organic matter, etc.) from
incomplete combustion of coal were present in the air,"but no single component or combinations
of particulate pollutants can clearly be implicated. The relative contributions of S02 or
particulate matter cannot be clearly separated based on these study results, and neither can
possible interactive effects with increases in humidity (fog) be completely ruled out.
However, temperature changes do not appear to be important in explaining the mortality effects
observed in Martin's studies.
A study by Glasser and Greenberg (1971), not listed in the table, appears to suggest with
less confidence that slight mortality increases were associated with increases in SO- above
786-1048 ug/m3 and in CoHs levels of 5.0-7.0 in New York City. These latter levels likely
correspond to concentrations in excess of 570-720 ug/m BS equivalent units based on
calibration studies by Ingram alluded to in Section 14.2 of Chapter 14. Again specific parti-
culate chemical species cannot be clearly implicated nor the relative contributions of S02 and
particulate matter separated. It should be noted that, whatever the causal agents, only very
small increases in mortality may have been detected at the above pollutant levels in New York
City.
Similar analysis of Lawther's morbidity studies listed in Table 1-13 suggests that acute
exposure to elevated 24 hr PM levels in the range of 250-500 ug/m (BS) in association with
24-hr S02 levels of 500-600 ug/m3 were likely associated with the worsening of respiratory
disease symptoms in chronically ill London bronchitis patients. Again, little can be said,
however, in terms of specifying physical or chemical properties of PM associated with the
observed effects beyond the comments noted above in relation to Martin's studies on mortality.
In regard to chronic exposure effects of S02 and particulate matter, the best pertinent
epidemiological health studies are summarized in Table 1-14. No epidemiological studies are
presently well-accepted as demonstrating associations between mortality and chronic (annual
average) exposures to sulfur oxides or particulate matter. The Lambert and Reid (1970) study
*XRD1C/B 1-75 3-2-81
-------
does suggest, however, that respiratory disease symptoms (cough and phlegm) are associated
3
with long-term (annual -average) exposures of adults to PM levels in the range of 100-200 ug/m
3
(65) or above in association with SCL levels in the range of 150-200 ug/m or above. The
studies by Ferris et al. (1973, 1976) also suggest that lung function decrements may occur in
3
adults at TSP levels in excess of 180 |jg/m in the presence of relatively low estimated S02
level, whereas no effects were observed by the same investigators at TSP levels below 130
or by other investigators (Holland and Stone, 1965; Deane et al., 1965; Comstock et al.,
3
1973) at TSP levels in the range of 70-163 ug/m , based on surveys of chest illness and
symptoms prevalence. Other studies by Lunn et al. (1967, 1970) listed in Table 1-14 suggest
that significant respiratory effects (increased respiratory disease; decreased lung function)
occur in children in association with long-term (annual average) PM levels in the range of
230-301 ug/m (BS) in association with S0_ levels of 181-275 ug/m3- No effects were seen for
3
children, however, at PM levels in the range of 48-169 ug/m (BS) or at SO- levels of 94-253
Again, no specific particulate matter chemical species can clearly be implicated as
causal agents associated with the effects observed in those studies listed in Table 1-14. Nor
can potential contributions of relatively large inhalable coarse mode particles be ruled out
based on these study results; and, it should be noted that various occupational studies listed
in Appendix C of Chapter 14 at least qualitatively suggest that coarse-mode size particles of
many different types of chemical composition can be associated with significant pulmonary
decrements, respiratory tract pathology, and morphological damage.
*XRD1C/B 1-76 3-2-81
-------
Cross-sectional
(5 cities)
TABLE 1-14. SUMMARY OF QUANTITATIVE CONCLUSIONS FROM EPIDEMIOLOGICAL STUDIES
RELATING HEALTH EFFECTS OF CHRONIC EXPOSURE TO S09 AND PARTICULATE
MATTER TO AMBIENT AIR LEVELS *
Type of study
Cross-sectional
(4 areas)
Cross-sectional
(4 areas)
Cross-sectional
(4 areas)
Longitudinal
and cross-
sectional
Longitudinal
and cross-
sectional
2
Annual average pollutant levels (pg/m )
parti cu late matter
Effects studied CoH BS TSP S02
Greater prevalence of cough - 100-200+ - 150-200+
and phlegm in areas of elevated
BS and S0» pollution, observed
in survey of 10,000 British
postal workers.
Likely increased frequency - 230-301 - 181-275
of lower respiratory symp-
toms and decreased lung
function in children
No observed effect on res- - 48-169 - 94-253
piratory symptoms and lung
function in children
Apparent improvement in - - 180
lung function of adults
in association with decreased
PM pollution in Berlin, N.H.
Apparent lack of effects - - 80-131
and symptoms, and no
apparent decrease in lung
function in adults
Reference
Lambert and Reid
(1970)
Lunn et al .
(1967)
Lunn et al.
(1970)
Ferris et al.
(1973, 1976)
Ferris et al.
(1973, 1976)
No apparent evidence of
increased symptom pre-
valence or chest illness
among telephone plant
workers
70-163
Holland and
Stone (1965)
Deane et al.
(1965)
Comstock et al
(1973)
------- |