United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-84-012
February 1984
Air
Optimum Sampling
Site Exposure
Criteria For Lead
-------
EPA-450/4-84-012
February 1984
Optimum Sampling Site
Exposure Criteria for Lead
by
D.J. Pelton and R.C. Koch
GEOMET Technologies, Inc.
Rockville, Maryland 20850
Contract Number 68-02-3584
Project Officer
David Lutz
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
DISCLAIMER
This report has been reviewed by the Office of Air Quality Planning
and Standards, U.S. Environmental Protection Agency, and approved for
publication as received from GEOMET Technologies, Inc. Mention of trade
names or commercial products does not constitute endorsement or
recommendation for use.
-------
CONTENTS
Figures iv
Tables v
1. Introduction 1
2. Monitoring Objectives for Lead 3
General 3
Requirements for monitoring lead 3
3. Characteristics of Lead Air Pollution 9
Airborne forms 9
Sources of emissions 9
Meteorological influences 10
Topographical influences 14
Observed patterns 20
Spatial scales of representativeness 29
4. Site Selection Methodology 33
Overview of methodology 33
Analyze existing monitoring data 35
Determine adequacy of mapping analysis and/or
select a modeling procedure 36
Air quality modeling 37
Selecting representative sites without monitoring
or modeling data 38
Network design 47
Specific site selection 48
Documentation 49
5. References 51
Appendixes
A. A Summary of Recent Findings on the Characteristics
of Lead Emissions 55
B. Plant Locations and Production Trends for Lead
Production and Refining 59
111
-------
F IGURES
Number Page
1 A typcial wind rose with wind speed information 13
2 A day-night wind rose showing, in this case, the diurnal
effect of the sea breeze 13
3 Characteristics of lake coast air flow 16
4 Hourly positions of lake breeze front of August 13, 1967 ... 17
5 Flow zones around a building 19
6 Flow characteristics among multiple buildings 19
7 Idealized urban heat island air flow 20
8 Observed daily mean concentrations of lead downwind of a
busy highway 23
9 Average 24-hour concentration of lead at various elevations
and setback distances 25
10 Illustration of various spatial scales of representativeness . 31
11 Procedure for selecting lead monitoring sites 34
12 Steps for locating micro and middle scale monitoring sites
in urban areas 39
13 Steps for locating a neighborhood scale monitoring site
in an urban area 40
14 Steps for locating a regional scale monitoring site 41
15 Steps for locating monitoring sites near isolated major
sources 42
16 Concentration as a function of wind speed, computed using
HIWAY2 model 44
17 Concentration as a function of wind direction, computed using
HIWAY2 model 45
18 Concentration as a function of stability class, computed using
HIUAY2 model 46
-------
TABLES
Number Page
1 Principal Uses for Lead Monitoring Data 4
2 Estimated 1981 Atmospheric Lead Emissions for the United
States 7
3 Uses of Lead--U.S. Data, 1981 11
4 Projected U.S. Use of Leaded and Unleaded Gasoline 11
5 Sources of Atmospheric Lead 21
-------
SECTION 1
INTRODUCTION
The primary purpose of this document is to guide Federal, state, and
local agencies in selecting sites for monitoring lead in the atmosphere.
This guideline provides more details on site-selection procedures than do the
Part 58 Regulations. Should any conflicts occur between the guideline and
the regulations, however, the regulations take precedence. In addition,
this guideline should not be used as a basis for rejecting data from existing
monitors sited prior to the publication of this guideline.
For monitoring networks to provide a representative sampling of air
quality in an area of concern, the number and locations of monitors must be
selected with care. This document emphasizes the concept of spatial repre-
sentativeness in selecting optimum monitoring sites to meet monitoring
objectives. A number of guidelines are given that can be used to identify
the types of representative sites that characterize exposure to lead in any
area of concern. Using these rules and knowing the objectives of a specific
monitoring group, the user of this document can select the number and loca-
tions of sites that best meet monitoring needs. Specific steps are recom-
mended for selecting monitoring sites with respect to each representative
type of site.
The contents of this document include the following subjects:
• Monitoring objectives and Federal requirements
t Characteristics of lead air pollution including airborne
forms, sources, distribution patterns, meteorological influ-
ences, and topographical influences
• Methods of site selection
t Site selection criteria.
This document extends and updates an earlier document prepared by PEDCo
Environmental, Inc. (PEDCo 1981a), which was published only in draft form.
-------
SECTION 2
MONITORING OBJECTIVES FOR LEAD
GENERAL
Monitoring objectives are established to fulfill authoritative requests
or provide information relevant to certain interests. The data obtained from
a planned ambient air monitoring network are examined to determine how well
the objectives are being met, and to revise the monitoring plan when necessary.
Air quality monitoring data are collected for the ultimate objective of
ensuring the protection of public health, but more immediate application of
the data may be intended for one or more of the following uses:
• Evaluation of ambient air quality
• Enforcement of source-specific regulations
• Evaluation/development of control plans
• Air quality maintenance planning
• Development and testing of models
• Research.
Further refinement and rationale of the objectives and data uses can
easily be established by the user. Table 1 lists a variety of uses for lead
monitoring data that are applicable to the six categories listed above.
More extensive discussions of these uses can be found in other guideline
documents such as those by Koch and Rector (1983); Ball and Anderson (1977);
Ludwig and Kealoha (1975); Ludwig, Kealoha, and Shelar (1977); Ludwig and
Shelar (1978), etc.
Once the objectives and data uses are determined, the monitoring network
and siting criteria are designed to accommodate the intended use.
REQUIREMENTS FOR MONITORING LEAD
National Ambient Air Quality Standard
The National Ambient Air Quality Standard (NAAQS) for lead, published
in the Federal Register (43 FR 46245, October 5, 1978), is 1.5 yg/m3, maximum
arithmetic mean averaged over a calendar quarter. The Federal reference
method for measuring atmospheric lead concentrations (given in Appendix G to
40 CFR 50) is by atomic absorption spectrophotometry analysis of particulate
matter collected on high-volume air sampler filters. Sampling every sixth
day will satisfy the monitoring requirements for an acceptable data base if at
-------
TABLE 1. PRINCIPAL USES FOR LEAD MONITORING DATA
Evaluate Ambient Air Quality
- Judge Attainment of NAAQS
- Establish Progress in Achieving/Maintaining NAAQS
- Establish Long-Term Trends
- Air Quality Indices
- Population Exposures Documentation
- Respond to Unique Citizen Complaints
- Develop/Revise Standards
Enforce Source-Specific Regulations
- Categorical Sources (New Source Review (NSR), Supplementary Control
Systems (SCS), Prevention of Significant Deterioration (PSD))
- Individual Sources
- Enforcement Actions
3. Evaluate/Develop Control Plans
- State Implementation Plan (SIP) Provisions
- Evaluate/Develop/Revise Local Control Strategies
4. Air Quality Maintenance Planning
- Establish Baseline Conditions
- Project Future Air Quality
5. Develop and Test Models
- Input for Receptor Models
- Validation and Refinement
- Assess Representativeness of Monitoring Networks
6. Research
- Effects on Humans, Plants, Animals, and Environment
- Characterize Source, Transport, Transformation, and Fate for
Anthropogenic and Natural Emissions
- Develop/Test New Instrumentation
-------
least 75 percent of the scheduled samples meet quality assurance guidelines.
A rigorous quality assurance program requires that all sources of sample
contamination be minimized, including surfaces of collection containers and
devices, hands and clothing of personnel, chemical reagents, laboratory
atmosphere, and labware and tools.
The monitors must be operated on a minimum sampling frequency of one
24-hour sample every 6 days, but the analysis of the 24-hour samples may be
performed for either individual samples or composites of the samples collected
over a month or quarter of a year.
Planning and Maintaining Control of Ambient Lead
Determination that an area is meeting the ambient air standard for
lead will depend heavily upon the selection of sites for monitoring lead.
As a minimum, SIPs are required to provide two lead monitoring sites (per
Appendix D of 40 CFR 58) in each urbanized area that has a 1970 population
greater than 500,000, or where lead air quality levels (measured since
January 1, 1974) exceed or have exceeded the lead standard. One of the two
monitoring sites must be located near a roadway in the area of expected
maximum concentration, and one site must be representative of a neighborhood
scale (see definition at end of Section 3). In addition, Subpart E (Control
Strategy—Lead) of Section 51.80 requires each SIP to demonstrate that the
NAAQS for lead will be attained and maintained in the following areas:
1. Areas in the vicinity of the following point sources of lead:
Primary lead smelters
Secondary lead smelters
Primary copper smelters
Lead gasoline additive plants
Lead-acid storage battery manufacturing plants that
produce 2,000 or more batteries per day
Any other stationary source that actually emits
25 or more tons per year of lead or lead compounds
measured as elemental lead.
2. Any other area that has lead air concentrations in excess of
the national standard concentration for lead, measured since
January 1, 1974. States may be allowed to limit the time
period to the last 3 years or since January 1, 1978, when
adequately justified (U.S. EPA 1983).
For lead SIPs, EPA has defined point sources differently than for other
pollutants. Point sources for lead are any stationary source causing emis-
sions in excess of 4.54 metric tons (5 tons) per year of lead or lead
compounds measured as elemental lead. This definition is important to
remember in planning monitoring sites.
-------
Many areas have no significant stationary sources of lead emissions.
Table 2 shows that most lead emissions come from gasoline consumption by
motor vehicles. Thus, lead concentrations near areas of heavy traffic and
on the downwind edge of dense urban developments are of major concern in
planning control measures. Ambient monitoring data are needed to confirm
that the SIP controls of motor vehicle emissions are adequate and working.
-------
TABLE 2. ESTIMATED 1981 ATMOSPHERIC LEAD EMISSIONS FOR THE
UNITED STATES
Annual U.S. Percentage
emissions of U.S. total
Source category (metric tons/yr) emissions
Gasoline combustion*
Waste oil combustion
Solid waste disposal
Coal combustion
Oil combustion
Gray iron production
Iron and steel production
Secondary lead smelting
Primary copper smelting
Ore crushing and grinding
Primary lead smelting
Other metallurgical
processes
Lead alkyl manufacture
Type metal
Portland cement
production
Miscellaneous
31,815
754
290
863
205
268
484
573
27
296
837
49
223
77
65
218
85.9
2.0
0.8
2.3
0.6
0.7
1.3
1.5
0.1
0.8
2.3
0.1
0.6
0.2
0.2
0.5
Total 37,032# 100
* Organolead vapors emitted to the atmosphere during the manu-
facture, transport, and handling of leaded gasoline are not
included in this inventory. In the October 1983 review draft
of Air Quality Criteria for Lead, it is estimated that these
emissions contribute less than 10 percent of the total lead
present in the atmosphere.
# Inventory does not include emissions from exhausting of workroom
air, burning of lead-painted surfaces, welding of lead-painted
steel structures, or weathering of painted surfaces.
Source: U.S. Environmental Protection Agency, Environmental Criteria
and Assessment Office, Research Triangle Park, N.C.
-------
SECTION 3
CHARACTERISTICS OF LEAD AIR POLLUTION
AIRBORNE FORMS
The lead compound that is emitted is dictated by the type of source
(e.g., alkyl lead compounds from petroleum refineries, lead salts from auto-
motive exhaust, elemental lead from smelters). However, automotive exhaust
accounts for 80 to 90 percent of lead emissions (see Table 2). Most research
on the forms of lead in the atmosphere has been directed toward characterizing
the fate of lead emitted from automobiles.
The chemical and physical form of lead emissions have important impli-
cations with regard to the sampling method used to measure atmospheric lead
concentrations. Not more than 10 percent of the airborne lead is associated
with particles exceeding about 2 vm diameter (Little and Wiffen 1978). Cars
driven at normal speeds emit aerosols mainly in submicron sizes. For a vehicle
operating on leaded fuel at idle, 20 mph, or 30 mph, the mode of the particle
size distribution occurs between 0.03 and 0.05 vm. At 50 mph, the particles
are slightly smaller. Although secondary aerosol formation causes a signifi-
cant change in the size distribution, the size distribution from auto emis-
sions is in close agreement with the size distribution of particles collected
near a freeway (Miller et al. 1976). Additional findings regarding lead
emissions to the atmosphere are summarized in Appendix A.
SOURCES OF EMISSIONS
Development of a monitoring strategy requires recognition of the emis-
sion sources. Procedures for preparing source inventories are described in
various other EPA guideline documents, e.g., Deyelopment of an Example
Control Strategy for Lead (EPA-450/2-79-002, OAQPS No. 1.2.123) and Supple-
mentary Guidelines for Lead Implementation Plans (EPA 450/2-78-038). An
updated revision to the latter document, Supplementary Guidelines for Lead
Implementation Plans—Updated Projections for Motor Vehicle Lead Emissions'
(EPA-450/2-83-002), should also be noted.Emission inventories will be
useful to identify the lead sources that must be included in the surveillance
plan.
The primary sources of lead to the atmosphere are automotive emissions
and waste oil incineration (see Table 2). An estimated 86 percent of lead
emissions to the atmosphere were due to automotive emissions from combustion
of leaded gasoline, based on 1981 data. The relative amount of lead emitted
to the atmosphere due to automotive emissions will decrease as unleaded
gasoline becomes the predominant fuel for automobiles. Coal combustion and
primary lead smelting contributed 2.3 percent each. Waste oil combustion
contributed 2.0 percent of the atmospheric emissions. All other categories
of stationary sources contribute 1.5 percent or less.
-------
Although highways are the main sources of lead emissions, major facili-
ties for lead production and refining and major industrial consumers of lead
can also be important local sources. Domestic ore, produced from eight mines
in Missouri, accounts for 87 percent of the domestic production. Mines in
Idaho and Colorado contributed 12 percent of the 1981 ore production. Of the
five primary lead smelters operating in 1981, the largest smelter is located
in Herculaneum, Missouri. Other primary lead smelters are located at Boss,
Missouri; East Helena, Montana; El Paso, Texas; and Glover, Missouri.
Consumption or uses of lead in the United States for 1981 are shown in
Table 3. Manufacture of storage batteries consumes the overwhelming share of
lead production. The use of lead in petroleum refining (nearly 10 percent of
the amount produced) is of concern to atmospheric concentrations of lead,
because most of the lead emitted to the atmosphere is from automotive emis-
sions. Lead emissions from automotive exhaust may be diminished by the
reduced use of leaded gasoline in conformance with the phasedown regulations
established by EPA in 40 CFR Part 80 and by the decrease in the number of
automobiles being driven that may use leaded gasoline. The use of unleaded
gasoline is projected to increase from 50 percent of the total gasoline sold
in 1980 to 81 percent of the gasoline sold in 1990 (see Table 4). Additional
data on lead production and consumption trends and the location of lead
producers and refineries in the United States are given in Appendix B.
METEOROLOGICAL INFLUENCES
The meteorological influences that need to be considered in selecting
monitoring sites can be described by a dispersion climatology that encompasses
those atmospheric parameters that affect the distribution of ambient concen-
tration. The parameters of primary concern are wind advection, horizontal
dispersion, and vertical mixing. With the exception of advection (i.e.,
surface winds), direct measures of these parameters are not routinely made in
most areas. The important fine structure needed to characterize significant
air pollution transport is generally not observed and must be inferred
indirectly (e.g., Hewson 1976, Holzworth 1974, and McCormick and Holzworth
1976). It is important to consider what regular data are available and what
additional parameters are needed.
Wind direction is the most obvious meteorological parameter influencing
the concentrations that will be observed. Wind speed influences the observed
concentrations by the rate of dilution of the emissions as well as rate of
transport while undergoing dispersion. Seasonal changes in wind patterns are
frequently observed at most U.S. cities. Changes in seasonal wind patterns,
apparent from climatological wind roses, are an important consideration in
selecting lead monitoring sites because the lead standard is based on a
quarterly average.
10
-------
TABLE 3. USES OF LEAD—U.S. DATA, 1981
Lead consumption — 1981
Metal Products
Ammunition
Bearing metals
Casting metals
Pipes extruded products
Sheet lead
Solder
Storage batteries
Other metal products
Pigments
Paints
Glass and ceramic products
Other pigments
Chemicals
Petroleum refining
Miscellaneous uses
TOTAL
Metric tons
49,514
6,922
18,582
8,829
19,355
29,705
770,152
50,648
16,316
44,339
19,510
111,367
21,862
1,167,101
Percent
of total
4.2
0.6
1.6
0.8
1.7
2.5
66.0
4.3
1.4
3.8
1.7
9.5
1.9
100.0
Source: Bureau of Mines Yearbook, 1981.
TABLE 4. PROJECTED U.S. USE OF LEADED AND UNLEADED GASOLINE
Percent of total gasoline sold
Year
1975*
1980*
1985t
1990t
Leaded
85
50
33
19
Unleaded
15
50
67
81
* Source: Chemical & Engineering News, No. 27, p. 12, 1980.
t Source: Federal Register, Vol. 47, No. 210, p. 49329.
11
-------
Monitoring sites may be selected by reviewing the frequency of wind
directions. River valley locations tend to have a high frequency of up-and-
down-valley air flow patterns due to channeling of the air along the valley,
especially during times when stable atmospheric conditions exist. Wind
patterns influenced by topographic features show a high frequency of wind
directions determined by the terrain-dominated circulation pattern. The high
frequency of the terrain-dominated wind directions will influence the long-
term average concentration.
The mixing height is frequently considered a seasonable variable, with
lower mixing heights occurring in the fall and winter when the atmosphere
is more stable than during the spring and summer. Low mixing heights
caused by atmospheric temperature inversions frequently occur in low-lying
areas and valleys. A sampling site placed in a locality with a high frequency
of low mixing heights may, over a period of months, result in a higher
pollutant concentration than would be observed from a site located on a hill
or level terrain.
Advection
For most monitoring objectives, advection is adequately defined by
the near-surface wind (speed and direction) measured at or adjusted to a
reference height of 10 m above ground. Routinely available observations
from the National Weather Service consist of short-term averages taken
hourly or every 3 hours. Although these are useful, vector averages based on
continuous recordings over 1-hour periods are more desirable where they are
available.
The frequency of air flow directions is an intuitively appealing siting
tool. One of the most useful summary depictions of wind flow shows the
frequency of occurrence of wind directions, with a breakdown of wind speed by
classes within each directional interval and is known as a wind rose (see
Figure 1). By convention, wind directions are denoted by the sector from
which wind is blowing. Wind roses may be constructed on an 8-sector basis, a
16-sector basis, or a 36-sector basis. Wind roses are commonly constructed
for annual, seasonal, or monthly distributions. Under some circumstances,
wind roses are devised to study winds under critical conditions. For example,
STAR* summaries offer a joint frequency distribution of winds and atmospheric
stability. These are available from the National Climatic Center^ and may be
compared for various time periods (e.g., see Figure 2). Additional categories
STability ARray, a broad-based algorithm for determining stability in the
Tower atmo?p~here using estimates based on winds and cloudiness. See Doty,
Wallace, and Holzworth (1976).
U.S. Department of Commerce, National Oceanic and Atmospheric Administra-
tion, Environmental Data Services, National Climatic Center, Federal
Building, Asheville, N.C. 22801.
12
-------
MILES PEH HOUR
1-5 6.15 16-30 >30
05 10 15 20 25 30
I 1 I I I 1 ' ' ' ' *
PERCENT FREQUENCY
Figure 1. A typical wind rose with wind speed information (Slade 1968
SEA
10 15 20 25
S2?
PERCENT FREQUENCY
Figure 2. A day-night wind rose showing, in this case, the diurnal effect
of the sea breeze (Slade 1968).
13
-------
of wind roses include winds under important pollutant index levels, distribu-
tion of persistent 24-hour winds, and distributions within key parts of the
day (i.e., morning versus afternoon).
Dispersion
Dispersion is the resultant effect of atmospheric turbulence to actively
dilute source material. Direct measurements of the three-dimensional wind
fluctuations that manifest turbulence are rarely made. Instead, various
methods of characterizing turbulence based on theoretical and empirical
relationships are employed. The most common system is based upon associations
among wind speed, solar insolation, and cloud cover. Many operational models
accept this type of data directly, and manual techniques have evolved to
treat these as well (see Turner 1970).
Mixing Height
Mixing height defines the vertical extent of mixing. Ground-based and
low-level inversions are the principal limiting factors. Mixing height is
determined from a thermodynamic analysis of vertical temperature soundings.
These soundings are routinely performed at 0000 GMT* and 1200 GMT each day at
a number of locations throughout the country.
Other Parameters
Additional parameters that may be useful are listed below:
• Precipitation—to relate to scavenging processes
• Air temperature—to be applied to plume rise estimates.
TOPOGRAPHICAL INFLUENCES
Uncomplicated (e.g., level, uniform terrain) settings for sampling are
rarely encountered. Distortions of the normal flow of air from the sources
to the monitor are caused by irregularities in the terrain and other physio-
graphic features. Two major factors in this regard are:
• Aerodynamic diversion—flow around and over obstacles.
Distortion of the flow field may be severe during moderate to
strong synoptic winds.
GMT means Greenwich Mean Time. The relation of GMT to U.S. standard times
can be determined by noting that 0000 GMT = 1900 EST = 1800 CST = 1700 MST
1600 PST.
14
-------
• Local circulations—mountain-valley winds, land-sea breezes,
and the like that may prevail when synoptic influences are
sufficiently weak. Under these conditions, flow patterns
within the scene may "wall off" subareas. Transport and
dispersion estimates at one place are unlikely to reflect air
motions elsewhere.
These factors will always influence the monitoring site selection.
However, because the standard for lead is based on average concentrations
measured over a calendar quarter, many of the terrain-induced influences may
be averaged over the sample period when the primary objective is to com-
pare air quality with the Federal standards. The influence of terrain and
physiographic features will be more critical when sampling for shorter
periods (especially less than 24 hours) and for source-oriented monitoring.
Some of the more salient considerations regarding terrain and physiographic
influences are described briefly in the following paragraphs.
Topographic Elements
Topographic elements become a factor when their influences extend into
the neighborhood scale (horizontal size order of kilometers). Because the
ratio of downstream aerodynamic effect to obstacle height is on the size
order of 10 to 1, obstacles on the order of 100 m will influence horizontal
sizes of the order of 1 km. The central problem that terrain introduces is
the added detail impressed upon the advection/dispersion field. That is,
a simple pattern that may be replicated consistently over level terrain
becomes distorted by three-dimensional perturbations in the presence of
substantial terrain relief. The principal types of flow distortion that
occur include separation flow on the downwind side of ridges when the flow is
perpendicular to the ridge, channeling of air flow by valleys, and local
circulations caused by differential heating of adjacent terrain slopes.
Coastal Settings
In coastal settings, during periods of light synoptic winds accompanied
by a sufficiently strong thermal contrast between water temperatures and land
temperatures, a land/sea breeze circulation (or land/lake breeze) will
control air motions in the vicinity of the shoreline.
Figure 3 displays the characteristic circulation patterns associated with
a lake (or sea) breeze (3a) and a land breeze (3b). This circulation system
is not static. As shown in Figure 4, the convergence zone migrates inland as
the land surface heats up. The intensity of the sea breeze may increase
through midafternoon, but dies out after sunset as the land surface rapidly
cools. At night, the land breeze sets up, but is generally less vigorous
because thermal contrasts are smaller.
15
-------
Lake
Land
Lake breeze
front
a. Lake breeze
Land breeze
front
Lake
Land
b. Land breeze
Figure 3. Characteristics of lake coast air flow.
16
-------
Hourly lake-breeze wind-shift positions
OPJ
Figure 4. Hourly positions of lake breeze front of August 13, 1967
(Lyons and Olsson 1972).
17
-------
The primary impact of this system is to recompose a coastal monitoring
scene into at least two siting domains: one area subject to the land/sea
breeze effects, another outside of this influence. The size and extent of
the land/sea breeze-affected subarea can be assessed in a number of ways. An
obvious factor of contrast is the horizontal distribution of wind directions
on appropriate days; however, few areas have sufficiently detailed meteorolog-
ical networks to define the horizontal extent of the area and the change in
size of the affected area with time. A more reasonable approach is to use
air temperature and relative humidity patterns to identify sites that are
affected. A distinctive signature will be observed in hygrothermograph
recordings that define the passage of the lake/sea breeze front.
Small-Scale Obstacles
Wind deflection around and over obstacles is a concern in selecting
specific sites in an urban area, because the effects occur on the microscale.
As shown in Figure 5, air does not simply slip past an isolated structure.
There are three distinguishable zones of air around a building:
1. Displacement zone—where streamlines are deflected upwind
and outward, remaining so for some distance
2. Wake zone—where streamlines gradually recover original
configuration
3. Cavity zone—return flow in the immediate vicinity of the
downwind side.
In terms of site selection, this effect is of obvious importance if an
intervening obstacle contains a strong enough source to generate a ground-
level impact that would be assigned to a source further upstream—particularly
if monitoring were to unwittingly take place in the cavity zone. This effect
is further complicated when many such obstacles are placed together, as shown
in Figure 6.
Urban Effects
In addition to the effects of individual buildings, a city induces
large-scale modifications to the local wind field. These modifications have
a bearing on site selection, due to the heat island circulation.
When a heat island circulation exists, there is a convergence zone over
the center of the city and a return flow into outlying areas, as illustrated
in Figure 7. This circulation pattern is most pronounced at night when
differential radiative cooling rates favor higher temperatures in the urban
center. The circulation pattern is generally weaker during the day when
urban/rural thermal contrasts are not as strong.
18
-------
lOHb
Figure 5. Flow zones around a building.
Figure 6. Flow characteristics among multiple buildings.
19
-------
Cold
/ / / / / 7/ X
Figure 7. Idealized urban heat island air flow (after Landsberg 1975)
Under sufficiently strong winds, the heat island circulation is over-
whelmed. Oke and Hannel (1970) have developed a simple relationship between
the threshold wind speed to prohibit the circulation and relative city size.
Oke and Hannel's empirical formulation is as follows:
UHm = 3.4 LogP-11.6
where P is the population number. Thus, a large urban area whose population
is counted in the millions can exhibit a heat island circulation even if
regional winds are quite strong. Although this relationship showed a high
correlation (94 percent variance explained) for the cities studied, it should
not be treated as an absolute measure. Each urban setting will have its own
idiosyncracies due to local terrain, presence of water bodies, or other
factors.
OBSERVED PATTERNS
The highest concentrations of lead air pollution have been observed in
the vicinity of major point sources, such as those listed in Table 5, and
near major highways and traffic interchanges. A map of the locations of
will identify areas of concern for monitoring.
inventory following the procedure described
pp. 5-19) is recommended if this has not
been done. Emissions of lead are also recorded in the Hazardous and Trace
Emissions System (HATREMS) data bank, and this source may also be of some
help.
point sources listed in Table 5
Development of a lead emissions
in detail by Smith et al. (1979
Lead emissions from automobiles are obviously significant. Guidance
for estimating automotive emissions is provided in EPA document number
20
-------
TABLE 5. SOURCES OF ATMOSPHERIC LEAD
Mining and milling—lead ore
Primary lead production
Primary copper production
Primary zinc production
Secondary lead production
Storage battery production
Gasoline additives
Solder
Cable covering
Type metal
Brass and bronze manufacturing
Waste oil combustion
Municipal incineration
Sewage and sludge incineration
Coal combustion
Distillate and residual oil combustion
Steel production
Gray iron foundaries
Cement production
Pigments
Silicomanganese electric furnaces
Ferromanganese electric furnaces—blast furnaces
21
-------
EPA-450/2-78-038a. Preparation of an emissions density map provides some of
the necessary information for determining the number and location of lead
monitors.
Many studies have been conducted to determine the pattern of lead air
pollution resulting from the dispersion and deposition of lead particles from
automotive exhaust. A recent study that provides the most complete data on
the distribution of automotive-generated lead particles near a highway is
referred to as the Philadelphia Roadway Study (Burton and Suggs 1982). The
study included horizontal and vertical arrays of samplers to collect particu-
late matter in fine (0 to 2.5 ym) and coarse (>2.5 to 15 urn) size fractions.
The average traffic density during the observation periods varied from
2119 to 3783 vehicles per hour. The average background concentration of
lead observed over a 2-month period was coarse particles 0.02 yg nr^
(25 percent) and fine 0.05 yg m~3 (75 percent), indicating lead was carried
primarily on the smaller particles. The horizontal array of samplers (2 m
above ground level) shows the distribution of lead in the downwind direction
from the roadway to be as follows:
• The highest concentration of lead occurs at the edge of the
roadway (0.15 yg m~3 (coarse), 0.53 yg m-3 (fine),
0.67 yg m~^ (total) above background).
• Concentrations decreased with distance from the roadway at a
rapid rate out to 75 m, then decreased at a much slower rate
out to 175 m.
• The downwind lead concentration stayed significantly above
background levels all the way to 175 m from the roadway.
• At 175 m downwind the total lead concentration was 0.1 yg m~3
above background; the lead content of fine particles accounted
for 0.09 yg irr3 of the total.
Sampling at 2, 7, and 15 m above ground level at downwind distances of
5 and 25 m showed that lead concentrations for fine and total particles were
significantly above background at all six sampling locations. Lead concen-
trations were significantly above background concentrations for all size
fractions to 175 m (the farthest distance at which sampling was done)
downwind of the roadway. The Philadelphia Roadway Study included sampling
only when the wind was ±45 degrees from a direction perpendicular to the
roadway; therefore, the rapid dropoff of higher lead concentrations close to
the roadway is more pronounced than if the samples had been collected over
the full range of wind directions that actually occurred.
Daines, Motto, and Chilko (1970) reported on the distribution of lead
in the vicinity of roadways with traffic density ranging from 19,800 to
58,000 vehicles per day. Samplers were placed 1.2 m above ground level and
22
-------
spaced from 3.0 to 152.4 m downwind of the roadway. The relationship
between lead concentration and distance from the highway was reported as a
function of traffic density (see Figure 8). Concentration of lead (above
background) was reduced by 50 percent between the 3- and 9-m sampling sites
with a traffic density of 58,000 vehicles per day. Beyond 45.7 m distance
from the roadway, the lead concentrations dropped off at a much slower rate.
The gradient of lead concentration close to the roadway was much less with
lower traffic density. Daines, Motto, and Chilko measured lead content in
various particle sizes and observed that the percentage of lead in larger
particles is above the background percentage only near the highway* The
percentage of lead present in smaller particles was above background when
sampled 533 m from the highway.
Daines, Motto, and Chilko concluded that a curvilinear decrease in lead
concentrations (as shown in Figure 8) can describe average concentrations
over long sampling periods but that the relationship may not describe short-
term conditions. For short-term periods, the distance to background levels
was found to be constantly changing due to meteorological parameters.
100 200 300 400 500
D-wancc lleell
Figure 8. Observed daily mean concentrations of lead downwind
of a busy highway (Daines, Motto, and Chilko 1970).
These investigators also note significant correlations of lead
concentrations with wind direction. When the sampler at 9.1 m setback was
downwind of the highway, the concentrations were 5.3 times higher than when
the sampler was upwind; when the sampler at 22.7 setback distance was down-
wind, the concentrations were 3.9 times higher than when the wind direction
placed the sampler upwind of the highway. A sampler at 37.9 m from the
highway had concentrations 5.0 times higher when it was downwind than when it
was upwind of the source. Their data indicate seasonal variations of lead
concentrations over the 2-year sampling period.
23
-------
The lowest seasonal average at the two setback distances (9.1 and
151.5 m) occurred during March, April, and May, while the highest seasonal
concentrations appear to occur in September, October, and November. The
seasonal influence may well be due to more stable conditions occurring more
frequently in the fall months. The Daines study also noted that the zone of
influence of the highway source is somewhat wider during the fall months
when atmospheric turbulence is at a minimum.
A recent field study, performed by PEDCo (1981b) to determine the
spatial variability of lead from roadways, indicates that higher average
concentrations were measured with a sampler inlet at 1.1 m above ground
level than at 6.3-m or 10.5-m elevations at all setback distances from the
highway. The experiment included setback distances of 2.8, 7.1, and 21.4 m
from the highway, with sampling at three elevations at each setback distance.
PEDCo also reports the lead concentrations at the upper elevation (10.5 m)
were lower at the tower nearest the highway (2.8 m) than at the same
elevation on the tower set 7.1 m back from the highway (see Figure 9).
Hunt (1983) reviewed the PEDCo data and noted that the confidence
interval about the mean for the sampler at 6.3 m elevation and 7.1 m setback
overlaps the confidence interval about the means of all other samplers in
the array, except for the sampers at a setback distance of 2.8 m and eleva-
tions of 1.1 m and 10.5 m. This finding was of special interest because the
sampler at 6.3 m elevation and 7.1 m setback distance was the only sampler
set up within the EPA criteria for the microscale roadway-type site. Hunt
noted that this indicates the EPA siting criteria are valid to obtain repre-
sentative samples in the vicinity of a roadway.
The setback distance from the roadway does not appear to be as critical
when comparing the means of all the 24-hour averages for the sampler placed
at 6.3 m elevation, e.g., the range and means of lead concentrations observed
are as follows:
Setback Range Mean
Distance (m) (yg/m3) (ug/m3)
2.8 0.18-2.13 0.96
7.1 0.40-2.35 1.07
21.4 0.29-2.22 0.97
Ondov, Zuller, and Gordon (1982) report results similar to those of
Daines, Motto, and Chilko, e.g., airborne lead concentrations drop off rapidly
with distance from the highway, especially within the 45 m bordering the
downwind side of the highway. Ondov, Zuller, and Gordon found that most of
the trace elements found in motor vehicle exhaust returned nearly to the
upwind concentration before reaching a site 45 m downwind. However, lead,
bromine, and chlorine that are emitted on fine particles in the exhaust
remained above background out to 90 m from the roadway.
24
-------
1.40
1.20
1.10
«->e i.oo
o>
3.
E
0.901
0.80
§ 0-70
o
<
bJ
0.60-
0.50-
0.40-
0.30-
0.20-
0.10-
TOWER
NO. 1
_L
_L
_L
_L
_L
_L
J_
6 8 10 12 14 16 18
MONITOR SETBACK DISTANCE, meters
TOWER
NO. 3
20 22
Source: PEDCo 1981b.
Figure 9. Average 24-hour concentration of lead at various elevations
and setback distances.
25
-------
The particle sizes of aerosols containing lead are primarily small
and remain airborne. Little and Wiffen (1978) observed that not more than
10 percent of the airborne lead samples collected near highways are asso-
ciated with particles exceeding about 2 pm diameter. In fact, within 1.5 m
of the roadway, Little and Wiffen found less than 10 percent of the airborne
lead is associated with particles larger than 5 ym. Little and Wiffen's
studies were conducted along two busy highways in England where the mix of
automobiles and exhaust characteristics may differ slightly from those in
the United States. They calculated the emission of lead from automobiles,
measured the deposition of lead within 100 m of the highway, and concluded
that only 9 percent of the lead emitted from automobiles at cruise speed
on a level roadway is deposited within 100 m of the roadway. Their results
demonstrate that although the concentrations fall off rapidly with distance,
most of the lead emissions from automobiles remain airborne for long-range
dispersal.
Evidence that significant deposition of lead to soil and vegetation
does occur due to fallout of lead aerosols has been investigated by Motto
et al. (1970). The lead content in soil increases with traffic volume and
decreases with distance from the roadway. Motto's observations indicate
the major effect of traffic occurs within 100 ft of the highway. Although
these observations made in northeastern New Jersey do not contradict those
of Little and Wiffen, they suggest that fallout must not be overlooked when
accounting for lead exposure due to pickup from the soil. This is especially
of concern in considering the exposure of young children.
Data collected along roadways in Texas by Bull in and Moe (1982) pro-
vide further information on the horizontal and vertical distribution of
lead particles near roadways. Sample sites were selected so the prevailing
wind could move perpendicular to the road section being studied; actual wind
data were not provided by the author. The data reported by Bull in and Moe
show the rapid decrease of atmospheric lead concentration with distance from
the highway that was reported by Daines, Motto, and Chilko. In general,
Bull in and Moe's results show the following:
0 Lead concentrations decrease rapidly with distance from the
highway to approximately 45 m from the road edge.
t Lead content is primarily in the fine particles.
t Vertical profiles of lead approximately 23 m from the road edge
show a decrease in concentration with height (on fine particles)
but quite low concentrations of lead on coarse particles
throughout the vertical profile.
26
-------
Feeney et al. (1975) determined the contribution of traffic-derived
aerosols to samplers adjacent to roadbeds. The study included sampling
activity near at-grade, raised, and depressed roadbed configurations.
Feeney's data show lead concentrations drop off rapidly with distance from
the roadbed for samplers spaced 27 m, 40 m, 100 m, and 160 m from the road-
way, at the at-grade and cut-section roadbed configurations. Near the fill
raised roadbed section, the concentrations remained remarkably constant out to
160 m from the road for a set of observations made in August. This is
consistent with concentrations close to the source being caused by fallout of
larger particles and concentrations further downwind being caused by vertical
dispersion of an elevated source. Feeney reported that dispersion calcula-
tions using an elevated source for this type of configuration showed reason-
able qualitative agreement with the observations.
Hutzicker, Friedlander, and Davidson (1975) have estimated that one-third
of the lead exhausted to the air in the Los Angeles basin is advected out of
the basin. That lead, carried out of the basin, is the principal source of
atmospheric lead for regions immediately downwind of the basin. Dramatic
evidence of the long-range transport of airborne lead is shown by the analy-
sis of lead in snow from the Greenland ice pack (National Academy of Sciences
1972). Lead in the ice pack increased gradually from the beginning of the
industrial revolution until 1950 when consumption of leaded gasoline in the
United States increased by doubling from 1940 to 1950, then more than doubl-
ing again by 1968.
Various researchers (Bullin and Moe; Burton and Suggs; Daines, Motto,
and Chilko; Little and Wiffen; Ondov, Zoller, and Gordon; PEDCo Environ-
mental, Inc.) have reported that lead concentration falls off rapidly with
increasing distance from the edge of the roadway over short-term (e.g.,
1 hour) periods. In general, the zone where the rapid decline occurs has
been reported at between 7 and 50 m downwind of the roadway. At distances
farther than about 50 m from the roadway, the concentration levels off and
declines at a much lower rate. Within the 50 m nearest the roadway, the
measured concentrations can vary considerably, due in large part to the
prevailing wind direction during the sample collection period.
Samples collected over a 3-month period may include wind directions
ranging from perpendicular to parallel for a roadway where a monitor is
located. The wind may blow from the road to the sampler or from the sampler
to the road. Therefore, the reported concentration for the quarter will
reflect a variety of sampling conditions.
As the wind direction becomes more aligned with the roadway, the zone
of influence due to traffic emissions becomes smaller, but the peak concen-
trations is higher and the concentration gradient is steeper (see Figure 17
27
-------
and later sections on selection of micro and middle scale monitoring sites).
Figure 17 shows how the concentrations may be expected to vary along a
roadway when the wind direction is at 90, 45, or 10 degrees in relation to
the roadway. The peak concentration from a wind blowing at a 10-degree
angle with the road would occur about 6 m from the median, but would be more
than twice as high as the concentration from a wind blowing at a 90-degree
angle with the road. As meteorological conditions vary over a long sampling
period, the observed concentrations will reflect the average of the condi-
tions that exist during the sampling period.
The high concentrations of lead near roadways are most prominent when
the sampling period is short and the wind speed remains constant and nearly
parallel to the roadway. However, the lead standard is based on a 3-month
average which, due to the normal variation of meteorological conditions
during the averaging period, will reduce the influence of the maximum concen-
tration conditions. The result of using a composite of samples collected
over a long period that includes a variety of sampling conditions is to
reduce the variation of high and low concentrations that is seen in short-
term samples taken at the same location. Another result is that the gradient
of long-term concentrations with distance from a highway is much lower than
the gradient of short-term concentrations. This is particularly evident in
the PEDCo observations (Figure 9). Therefore, the distance that measurements
are made from a nearby highway is not as critical for measurements of the
maximum quarterly mean as it is for measurements of the maximum 24-hour
concentration.
In summarizing, the following observations of airborne lead concentra-
tions are relevant to monitor siting:
• Emissions from motor vehicles using leaded gasoline are the
primary source of airborne lead.
• Airborne lead is principally associated with fine particles.
• Highest concentrations occur close to roadways with high
traffic density; the concentration gradient falls rapidly
with distance from the roadway during short-term periods
and much less rapidly over long-term periods.
• Fine particles are carried well beyond the immediate area of
roadways.
Studies of patterns of airborne lead concentrations show that a signifi-
cant portion of the lead from automobile exhaust is deposited near roadways,
but most is transported long distances in small particles. As a consequence,
the selection of lead monitoring sites must be influenced by proximity to
emission sources and meteorological parameters.
28
-------
SPATIAL SCALES OF REPRESENTATIVENESS
Monitoring sites required by SIPs describe a spatial scale of represen-
tativeness typically referred to as follows:
• Microscale—ambient air volumes ranging in horizontal extent
from a few meters to as much as 100 m. The microscale encom-
passes the immediate vicinity of the monitor. In the immediate
presence of lead sources, exposure may in reality be only
representative of the microscale. For this reason, the
microscale is the final judgmental factor in site selection
and requires a site visit to make this appraisal, because maps
rarely portray confounding influences in sufficient detail.
t Middle scale—ambient air volumes covering areas larger than
microscale but generally no more than 0.5 km in extent. In
settled areas, this may amount to several city blocks. This
is essentially the lower limit of resolution for most models.
• Neighborhood scale—ambient air volumes whose horizontal
extent is generally between 0.5 and 4 km. The neighborhood
scale is aptly named. It is useful in defining extended areas
of homogeneous land use.
• Urban scale—ambient air volumes whose horizontal extent may
range between 4 and 50 km. This is frequently the most
desirable representative spatial scale, because it captures an
entire urban area. However, the diversity of sources that
prevail within such areas argue against homogeneity at this
scale.
• Regional scale—ambient air volumes whose horizontal extent
ranges from tens of kilometers to hundreds of kilometers.
Monitors that are unaffected by specific sources or by localized
groups of sources can be representative at this scale.
• National and global scales—seek to characterize air quality
from a national perspective (thousands of kilometers) or from
a global perspective (tens of thousands of kilometers).
The concept of representative spatial scale is used to define a charac-
teristic distance over which pollutant concentrations are uniform or nearly
so. As a corollary, we can define homogeneous areas in which measurements
performed in the relatively small air volume near a sampler (nominal horizontal
extent of 1 m) can represent conditions prevailing over some much larger
area.
29
-------
Representative spatial scales illustrated in Figure 10 have been pre-
viously identified (U.S. EPA 1977) and are compatible with spatial scales of
source areas. The scales of representativeness that will be of most concern
for lead monitoring are microscale for maximum concentrations and middle or
neighborhood scales for more general application.
To assess the scale of representativeness, the area must be analyzed
with respect to emissions, physical setting, and meteorological and climato-
logical characteristics.
30
-------
,' Micro-Scale
(<0.1 km)
Neighborhood Scales
(0.5 to * km)
URBAN COMPLEX
Regional Scales
(>50 Im)
Urban Scales
(4 to 50 km)
Figure 10. Illustration of various spatial scales of representativeness
(Ball and Anderson 1977).
31c
-------
SECTION 4
SITE SELECTION METHODOLOGY
The general procedure recommended for selecting sites to monitor airborne
lead concentrations is shown in Figure 11. Variations in the details performed
within each step are recommended for different topographical situations and
different configurations of sources of emissions.
OVERVIEW OF METHODOLOGY
The siting of monitors is part of a continuing planning cycle for moni-
toring. The three basic elements of the cycle are defining the objectives
of monitoring, reviewing monitoring data, and making judgments about the
adequacy of the air quality data. The iterative process provides flexibility
in the use of monitoring resources. The need is clearly recognized in EPA's
monitoring regulations and has resulted in the development of three types of
monitoring activities by state and local agencies, including National Air
Monitoring Stations (NAMS), State and Local Air Monitoring Stations (SLAMS),
and Special Purpose Monitoring (SPM). The locations of NAMS and SLAMS must
be coordinated with EPA regional offices because these must be designed to
meet EPA needs in addition to state and local needs. The siting methodology
is applicable to all three types of monitoring stations and will be useful to
industrial operating facilities as well as air pollution control agencies.
The NAMS monitoring sites for lead will include a roadside site and a
neighborhood site. The roadside site must be adjacent to and downwind of a
traffic volume that exceeds 30,000 vehicles per day. Additional SLAMS sites
may be established to determine that an area of special interest does not
exceed the ambient air quality standard, to establish that lead control
measures are effective in reducing exposure levels, or to measure background
levels. The following six-step procedure for selecting monitoring sites is
recommended:
1. Analyze existing monitoring data.
2. Determine adequacy of mapping analysis and/or select a
modeling procedure.
3. Perform air quality modeling if necessary.
4. Determine the number of monitoring sites required to describe
the area of interest.
5. Propose locations for those sites.
6. Document and update site exposure experience.
Specific suggestions for each of these steps is given in the following
paragraphs.
33
-------
Mom ton ng
data
Review lead
monitoring data
Are
data sufficient
for mapping
analysis?
Emissions and
topography
data
Is
analysis
adequate
Meteorological
data
Emissions and
topography
data
Determine network
requirements
(numbers and locations)
Step 4
Select monitoring
sites and
placements
Step 5
Figure 11. Procedure for selecting lead monitoring sites.
34
-------
Site planning may vary in scope of responsibility and may include any of
the fol lowing:
• Design multipurpose network
« Supplement existing network for specific purpose
0 Design single-source impact or compliance monitoring network
• Monitor a designated area or location.
The procedure for site selection can proceed once the objectives of the
monitoring effort have been established. The need for emissions maps and the
meteorological and topographical influences have been discussed in the
preceding section. In this section a process is presented for using the
information available for site selection to develop a monitoring network that
meets the operator's desired goals.
ANALYZE EXISTING MONITORING DATA
To select monitoring sites, the monitoring planner must form a conception
of the spatial distribution of lead concentrations over the area of concern.
If an adequate data base of ambient lead measurement is not available to meet
this need, the distribution must be estimated by mathematical simulation
modeling or by a reasonable, physically based qualitative analysis. The best
method of estimating the distribution of air quality levels will depend on
the amount, type, and quality of available information. The information of
interest includes the following categories:
• Ambient lead measurements
• Locations and amounts of lead emissions
t Air pollution climatology and meteorology data
• Maps of topographical features.
The amount of lead monitoring data available to help design a monitoring
network is likely to be incomplete. However, whatever data are available
will be valuable. The SAROAD data base, available from EPA regional offices,
is a convenient source of much of the available data. State and local air
pollution control offices are also important sources of additional data and
information about other data that may have been collected by nongovernment
parties or in special studies. The available ambient lead observations
should be critically reviewed to eliminate any data that are suspect because
of poor quality control.
-------
Mapping Analysis
Mapping the data will identify areas of special concern, e.g., locate
areas of high concentrations. The analyst must decide if there are sufficient
valid data within the area of concern to warrant mapping the data. If data
are available from fewer than six sites, single station analysis is likely to
be more practical than mapping. Mapping analysis will simply involve drawing
isopleths of lead concentration. The lead concentration data must represent
data from a time period common to all points and data collected by reasonably
similar sampling and analytical techniques. The number and value of contours
to be drawn will depend on the range of values observed and the nature of
their spatial distribution. Computer graphics packages are available to per-
form the contouring analysis if manual analysis is not practical. Generally,
about six contours will provide a useful display. However, as few as 1 or as
many as 10 may be appropriate, depending on the magnitude of the range rela-
tive to the mean of the values observed. The maps will be used to identify
representative spatial scales and preliminary siting selections.
Single-Station Analysis
Single station analysis of quarterly means will include analysis for
trends over time, peak concentrations, number of exceedances of the standard,
mean, and standard deviation. Single-station analyses may be performed to
identify the significant influencing factors that affect the lead air quality
levels observed. This identification process will help determine how wide an
area the station represents. Conclusions drawn from one station should be
compared with results from other stations in the area of interest. Trends
and frequency distributions help in analyzing single-station data. Case
study analyses of peak quarterly values will also be helpful.
Another useful single-station analysis is the pollution rose. The
pollution rose is constructed by computing the average measured concentration
for all values when the prevailing wind is in a given direction. The values
may be limited to days when the wind persistence index (ratio of vector to
scalar wind speed) exceeds a certain value.
DETERMINE ADEQUACY OF MAPPING ANALYSIS AND/OR SELECT A MODELING PROCEDURE
An important step in the process of selecting monitoring sites is to
identify the unique local influences that are affecting air quality. The
types of topographical features and the magnitudes and locations of lead
emissions have a major impact on where the worst air quality levels will
occur. In assessing the value of a"?ilable monitoring data and in selecting
an air quality simulation model, it is necessary to take these local influ-
ences into account.
While the mapping and station analysis data may be helpful in identifying
the spatial distribution of lead, they may be inadequate. Having analyzed
36
-------
the available data, the monitoring planner must consider whether modeling is
needed to supplement the available monitoring data. Consideration should be
given to gradients evident in the observations, locations of major sources,
terrain, and meteorology. In most cases, the available lead observations will
not be adequate for planning a new monitoring network.
The adequacy of the data analysis may be judged by whether the air quality
pattern can reasonably explain the inventory of sources and the influences of
terrain and meteorology. Two tests of the air quality pattern are suggested.
One test involves the time history of the pattern; the other test examines
the shape of the pattern of emission densities and topographical features.
If the patterns of annual means or maximum 24-hour concentrations for
several years show the same shape and same locations of peaks when superimposed
on each other, the pattern is consistent with time. This consistency is
evidence of a stable pattern, which is a reasonable guide for planning monitoring
sites. If the pattern is changing with time, the analysis may be adequate,
but the reasons for the changing pattern should make sense in terms of changes
in sources or in meteorological conditions. If there are no apparent reasons
for the changes, modeling results should be obtained and reviewed.
Emission densities that are chronologically consistent with the air
quality data should be plotted and used to generate contour patterns.
Topographical features may also be located on these patterns. When the
emission density contours are superimposed on the air quality patterns, there
should be a reasonable relationship. A reasonably consistent pattern would
be one in which the air quality pattern is offset from the emission pattern
in the direction of prevailing wind flow. If the influence of major peaks in
emission density are not evident in the air quality pattern, a modeling
analysis may be helpful in identifying the magnitude of the pattern deforma-
tion that can be expected.
AIR QUALITY MODELING
Computer models that mathematically simulate air quality levels can provide
help in selecting monitoring sites, especially where overlapping contributions
from multiple sources need to be considered. The following factors influence
how useful computer models will be:
t The air quality estimates are limited by the accuracy of the
assumed time, location, and rate of emissions.
• The air quality estimates are also limited by how representative
the meteorological data are of the area between each pair of
source and receptor locations.
• The cost of assembling and preparing data and of running computer
models for multiple sources can be expensive and must be carefully
and knowledgeably planned.
37
-------
Selecting a model to analyze lead must be done with due consideration
of uncertainties that will influence the results such as limitation of the
models to treat the processes that cause changes in the particle size
after it is emitted from the source, limitations of the models to treat wet
or dry deposition of the emitted particles, limitations of the quality of
emission factors, limitations due to a constantly changing mix of vehicles
using leaded and unleaded gasoline, and limitations of simulation models to
treat special terrain-induced changes to the dispersion pattern.
With appreciation of the limitations mentioned, modeling analysis will
be useful because lead monitoring sites are usually selected close to the
emission source. The dispersion models recommended in EPA's guidelines on
modeling and on lead SIPs (U.S. EPA 1978a, 1978b, and 1978c) provide the
capability required for the analysis. EPA regional offices can be of con-
siderable help in determining the value of modeling and in selecting an
appropriate model.
SELECTING REPRESENTATIVE SITES WITHOUT MONITORING OR MODELING DATA
There may be situations in which it is not possible to use monitoring
data or the results of a modeling analysis to define the pattern of air
quality levels in an area that is to be monitored. In this case, the moni-
toring network can be planned by identifying representative sites on the
basis of available information on sources of emissions, climatological data,
and topographical considerations. Observations from other locations and
previous modeling analyses of general classes of source influences may be
used to select monitoring sites for these situations. Requirements for
monitoring lead concentrations will cover various scales of representativeness,
including micro-middle scale, neighborhood scale, regional scale and monitor-
ing near major sources isolated from other significant sources. Steps for
locating monitors for each of these four types of representative sites
are suggested in Figures 12, 13, 14, and 15. The most frequently encountered
situation is the roadside monitor representing the micro and middle scales.
Selecting Micro and Middle Scale Monitoring Sites
The most common lead monitoring sites will be those near heavily
traveled roadways in urban areas where there are no major point sources.
Monitoring will be representative of areas classified as middle scale.
Figure 12 indicates appropriate steps for selecting monitoring sites for
this situation.
The first step is to obtain and analyze traffic and urban development
data that can be used to identify potential variations in otherwise homoge-
neous neighborhood-scale patterns of concentrations. Areas of high traffic
density, such as major highways, shopping centers, sports arenas,, amusement
parks, airports, and parking facilities, need to be identified. Other known
sources of lead emissions such as waste incinerators or sources that are not
considered major sources should be identified. Estimates of the impact of
38
-------
re there
any major
point
sources?
Analyze siting requirements for
isolated major source (Figure 15)
Assemble and analyze data on highway traffic,
major indirect sources of traffic concentration,
urban development, and wind direction frequencies
Determine number of peak
concentration monitors
Select sites on downwind side of major
roadway and indirect sources, and on
downwind side of urban area in maximum
impact zone for most prevalent wind direction
Select monitor air inlets that are not
shielded by structures or affected by
adjacent local sources
Figure 12. Steps for locating micro and middle scale monitoring sites
in urban areas.
39
-------
Are there
any major
point
sources?
Yes
Locate and characterize effective
height of major point sources
Determine distance of maximum
impact from each major source
Determine locations of overlapped
effects from multiple point sources
Assemble and analyze data on highway
traffic, major indirect sources, urban
development, and wind direction frequencies
Determine number of neighborhood
scale monitors
Divide area into neighborhood and
select neighborhoods to monitor
Select sites in each neighborhood
not influenced by major point sources
Select monitor air inlets that are not
shielded by structures or affected by
adjacent local sources
Figure 13. Steps for locating a neighborhood scale monitoring site
in an urban area.
40
-------
Locate and characterize:
Major urban areas
Major point sources
Wind direction frequencies
Major terrain features
1
Determine number
of sites required
Select site(s) using source
avoidance and wind direction
frequency considerations
I
Modify site selections based
on topography considerations
Figure 14. Steps for locating a regional scale monitoring site.
41
-------
Assemble and analyze emissions,
clinatological, and topographical data
Determine zones of maximum impact
based on climatology
Determine zones of maximum
impact based on
topography
Determine number of monitoring
sites for monitoring background,
maximum impact, and sensitive areas
Select sites in potential maximum
impact zones that are not shielded
by vegetation, terrain, or structure
Are
there any
sensitive
areas?
Select sites in sensitive areas
that are not shielded by veaetation,
terrain, or structure
Select background site
as suggested in Figure 14
Figure 15. Steps for locating monitoring sites
near isolated major sources.
42
-------
lead emissions from automobile exhaust on quarterly mean lead concentrations
can be prepared using the 1-hour concentration profiles for a variety of
conditions shown in Figures 16, 17, and 18 as a guide. These figures indi-
cate the areas of major impact along a highway as a function of meteorological
variables. Graphs similar to these will be helpful in determining which
sections of a roadway offer the greatest potential for high concentration of
lead merely by inspecting the data. For instance, if a major roadway is
oriented from northeast to southwest and the predominant wind speed and
direction is 4 m/s from the northwest, the zone of maximum concentration
would occur approximately 10 to 14 m from the median as shown in Figure 16.
The concentrations shown are for a traffic density of 3000 vehicles per hour
and an emission rate of 0.056 g per km per vehicle. The location of the
maximum is not affected by changes in the emissions or the traffic density.
The value of the maximum can be adjusted for different emission rates or
traffic densities by multiplying the values read from the graph by the ratio
of actual traffic or emission rate to the one shown in Figure 16. Figure 17
shows how the orientation of the highway to the prevailing wind direction
affects the magnitude and location of the peak. These curves, prepared
for guidance only, use calculations for 1-hour averages; longer-term averages
will flatten the slope from the peak somewhat.
On the basis of the concentrations predicted for all the traffic-
concentrated areas and the locations of the source areas relative to the
downwind edge of the city for the most prevalent wind direction, a decision
must be made on how many monitors will be used to measure the maximum lead
concentration. Unless a single source or source area is clearly more signif-
icant than any other, a number of sites should be selected as potential peak
concentration monitoring sites. These sites will be representative of micro
or possibly middle scale areas. The monitoring site should be located as
close to the source as possible without infringement or interference from
the source. The most suitable sites are within 5 to 15 m of the sources on
the downwind side of the prevailing wind direction. It is usually not
practical (nor acceptable, on the basis of Appendix D of 40 CFR 58) to
locate a site less than 5 m from a source. Generally, one site is sufficient
for each source area.
Neighborhood Scale Monitoring Sites
Steps for locating a neighborhood scale monitoring site are indicated in
Figure 13.
Neighborhood sites are needed to represent the areas that encompass or
surround the peak concentration sites. Due to variations in the type and
intensity of land uses throughout an urban area, a large metropolitan area
may be characterized by well over 1000 different neighborhoods. It is
possible to characterize neighborhoods in a qualitative fashion by preparing
a detailed emission inventory that identifies the spatial distribution of
lead emissions on a gridded basis using traffic and other relevant data. By
examining the locations and magnitudes of lead emissions by gridded area in
43
-------
•8 -S S
•r > o
"
w
iru
5 U &
IU
cu
ffi
(I
CD
o
UJ
o
O)
Q
LU
a.
t/o
Q
LU
C3
O)
-O
O
CM
cr-
0>
CL
O
O
TD
OJ
Ol
Q.
"O
c
+J
o
c:
o
(O
-(->
c
OJ
o
c:
o
o
CL)
s_
<-> C
t— o:
—< c
44
-------
aj
4-)
01
D
at
1
00
£
m wide highway
00 vehicles/h
I O
CNJ CO
<"" .y
4 H
0. 056 g/ vehicle/km
IE
a
o
w
j^
a
in
13
d
a
TJ
a
£
Q
CO
u
1— t
1
ri
PL.
0
D
"E
rjw
a
ex
LO
T3
a
ui
o
o:
u_
00
CL)
TD
O
C\J
zr.
o.
-o
d)
o
o
c
o
o
s_
TD
C
'S
-------
« !s
rt
* 0
-C o>
U> 5
60
S
> s
" e
o _o
^M ^*
2 1
ll
«
a
§
UJ
m
o
•d
o
LU
O
o:
o
CO
l«4
0
ni
CO
o-
00
o
z:
LjJ
UJ
d)
•a
o
CM
•a
O)
o
o
CO
CO
ro
-(->
CO
O
c
o
•i—
4-J
(_)
3
1C
co
re
fO
i-
ai
o
oo
CD
O
rg
zr H- o:
I— •- c
ro
_ E
en
46
-------
relation to the climatology of wind direction frequencies, one can rank
neighborhoods in terms of their expected levels of high concentrations.
Neighborhoods that encompass the middle or micro scale areas that are
expected to contain high concentrations are clearly high priority neighborhoods
for monitoring sites. One or two neighborhoods adjacent to the maximum
concentration neighborhoods are desirable secondary sites. A third category
of monitoring sites includes neighborhoods that are of special interest
because of large population density, because of rapid growth expectations, or
because of a highly sensitive population such as elementary school children.
Sites in the third category of interest may also meet the second category
of interest. There are no firm rules to determine how many sites to monitor.
Each monitoring jurisdiction must determine what its priorities are and how
far down the priority list of potential sites it is able and willing to go.
Regional Scale Monitoring Sites
Regional scale monitoring sites may be needed to measure background
levels of lead that are transported into the area being monitored. It is
important that regional scale monitoring sites not be affected by nearby
sources, which would significantly alter their scales of representativeness,
for large periods of time. It may be necessary to use two or more sites to
measure background concentrations when a single site cannot be found that is
never influenced by nearby sources. Figure 14 suggests four steps to follow
in selecting the site(s).
Monitoring Isolated Major Sources in Flat Terrain
Figure 15 suggests steps to be followed in selecting monitoring sites
near an isolated major source. A distinction must be made between sources
with the principal emissions from a tall stack and sources with the principal
emissions from ground level. For ground-level sources, the maximum concentra-
tions will occur immediately adjacent to the source in the most prevalent
downwind directions from the source. Wind observations will easily identify
the most suitable siting areas. Additional monitors may be used to help
define the extent of the area near the source that has high concentrations and
the neighborhood scale level of lead in the vicinity of the source. Two types
of information can be helpful in determining the extent of the high impact
area: (1) the relative concentration isopleths from the EPA (Turner 1970)
Workbook of Atmospheric Dispersion Estimates and (2) annual wind direction
frequency statistics published by the National Climatic Center.
NETWORK DESIGN
The information that will be gained from lead monitoring may include
determination of the maximum concentration, the background concentration, and
a determination of the area impacted by significant concentrations of lead.
The primary matter of concern is to determine if the air quality standard is
47
-------
exceeded, especially in areas where human exposure will occur. The discussion
on selection of monitoring sites has, for the most part, addressed the issue
of determining the pattern of atmospheric lead concentrations. Only in very
few situations will the emissions and dispersion patterns be simple enough to
require only one monitoring site to accurately reflect the maximum exposure
area. Such a situation would arise if the peak concentration occurred in the
same location most of the time. Complex patterns have two or more peaks that
may or may not lie within a single closed contour of impacted areas of
interest. Unless one peak is much higher than the others, two or more peak
areas will need to be monitored.
The number of monitors needed to define impacted areas will include a
minimum of two and may include six or more, depending on how large, how
complex, and how definitive the impacted area is. A single, well-sited
monitor, located well away from any nearby sources or source areas, may be
adequate for determining background concentrations. If it is impractical to
locate a monitor far away from nearby sources, it may be desirable to select
two nearby monitors, one or more of which is measuring background concen-
trations on any given day, depending on wind direction. Because lead concen-
trations are measured over 24-hour periods and because the wind direction is
frequently variable over a 24-hour period, this is a less desirable option
than a single, well-sited monitor.
In planning and revising air monitoring plans, it is important to bear
in mind that the need for monitoring data is dynamic and will change from
year to year. Once the nature of the air quality pattern for lead concen-
trations has been established or verified, fewer stations are needed to
evaluate general ambient conditions and trends. This is especially true for
areas where the ambient levels are well within acceptable limits and there is
no significant impact area. Reducing the amount of resources allocated to
fixed monitoring stations will allow resources to be reallocated to meet
other special-purpose monitoring needs.
Previous monitoring and modeling provide a first estimate of the lead
air quality patterns, but a large amount of uncertainty may still exist
regarding both the shape and the magnitude of the pattern. Therefore, some
monitoring resources should be allocated to verifying the assumptions made
regarding the pattern. Installation of temporary monitors can be a useful
approach to confirming the results or patterns observed from the previous
study. A monitoring site should be established in any area where there is a
good probability the lead concentration may exceed the standard and human
exposure will occur.
SPECIFIC SITE SELECTION
Once a general area for a monitoring site has been selected, it is
necessary to select a specific location for the sampling operation. The
intake for the monitor must be representative of the siting area, as close to
the breathing zone as possible, and not biased abnormally high or low by
influences that are representative only of the probe intake.
48
-------
Some requirements for monitoring sites are established by existing rules
and guidelines. The guidance for siting high-volume samplers that are used
for collecting airborne lead is given in Appendix E of 40 CFR 58. The
following guidelines for siting were promulgated in Section 2 of Appendix E
(40 CFR 58):
• 2-15 m above ground, as near to breathing height as possible,
but high enough not to be an obstruction and to avoid vandalism
0 At least 2 m away horizontally from supporting structures or
walls
• Should be 20 m from trees
• Should not be near furnace or incinerator flues
• No nearby obstructions to air flow due to buildings, structures,
or terrain, at least in directions of frequent wind.
DOCUMENTATION
With any worthwhile activity, documentation of the work is necessary to
substantiate the data that have been or will be produced. When a monitoring
site is established, a description of the site should be prepared. The site
description should include the following information:
• Exposure diagram
- Horizontal depiction showing location relative to nearby
streets, buildings, and other significant structures,
terrain features, or vegetation
- Vertical depiction showing location relative to supporting
structures, including buildings, walls, etc.
• Height of sampling intake above ground level
• Microinventory map showing locations of roads (with traffic
counts), open fields, storage piles, and any visible emissions
within 500 m of sampler
• List of all inventoried point and area sources within 1.5 km
of sampler and all major point sources within 8 km of sampler
• Types of meteorological and other air monitoring equipment
operated at the site
• Make, model, and serial number of the sampler.
49
-------
SECTION 5
REFERENCES
Ball, R.J., and G.E. Anderson. 1977. Optimum Site Exposure Criteria for
S02 Monitoring. EPA-450/3-77-013, U.S. Environmental Protection Agency,
Research Triangle Park, N.C.
Boyer, K.W., and H.A. Laitinen. 1975. "Automobile Exhaust Particulates--
Properties of Environmental Significance." Environ. Sci. Techno!. 9(5).
Bull in, J.A., and R.D. Moe. 1982. "Measurement and Analysis of Aerosols
Along Texas Roadways." Environ. Sci. Technol. 16(4):197-202.
Bureau of Mines. 1981. Bureau of Mines Minerals Yearbook 1981—Lead.
U.S. Department of Interior, Washington, D.C.
Burton, R.M., and J.C. Suggs. 1982. Project Report—Philadelphia Roadway
Study (Draft). Environmental Monitoring Systems Laboratory, Office of
Research and Development, U.S. Environmental Protection Agency.
Chemical and Engineering News (27)12, 1980.
Daines, R.H., R. Motto, and D. Chilko. 1970. "Atmospheric Lead: Its
Relationship to Traffic Volume and Proximity to Highways." Environ. Sci.
Technol. 4(4):318-23.
DeJonghe, W.R.A., and F.C. Adams. 1980. "Organic and Inorganic Lead Concen-
trations in Environmental Air in Antwerp, Belgium." Atmos. Environ. 14:
1177-80.
Doty, S.R., B.L. Wallace, and G.C. Holzworth. 1976. A Climatological
Analysis of Pasquill Stability Categories Based on STAR Summaries.
National Oceanic and Atmospheric Administration, Asheville, N.C.
Feeney, P.J., et al. 1975. "Effect of Roadbed Configuration on Traffic
Derived Aerosols." J. Air Pollut. Control Assoc. 25(11):1145-47.
Gerstle, R.W., and D. Albrinck. 1982. "Atmospheric Emissions of Metals
from Sewage Sludge Incineration." J. Air Pollut. Control Assoc. 32:1119-23.
Greenberg, R.R., W.H. Zoller, and G.E. Gordon. 1981. "Atmospheric Emissions
of Elements on Particles from the Parkway Sewage-Sludge Incinerator."
Environ. Sci. Technol. 15(1):64-70.
Habibi, K. 1973. "Characterization of Particulate Matter in Vehicle
Exhaust." Environ. Sci. Technol. 7(3):223-34.
51
-------
Hewson, E.W. 1976. "Meteorological Measurements." In: Air Pollution
(3d edition), Vol. I. A.C. Stern, ed. New York: Academic Press.
Holzworth, G.C. 1974. "Climatological Aspects of the Composition and
Pollution of the Atmosphere." Tech. Note No. 139. Secretariat of the
World Meteorological Organization, Geneva, Switzerland.
Huntzicker, J.S., S.K. Friedlander, and C.I. Davidson. 1975. "Material
Balance for Automobile-Emitted Lead in Los Angeles Basin." Environ.
Sci. Technol. 9(5):448-56.
Koch, R.C., and H.E. Rector. 1983. Network Design and Optimum Site
Exposure Criteria for Participate Matter.Draft Report, Office
of Air Quality Planning and Standards, U.S. Environmental Protection
Agency, Research Triangle Park, N.C.
Landsberg, H.H. 1975. "Atmospheric Changes in a Growing Community."
Tech. Note No. BN 823. University of Maryland, College Park, Md.
Lead Industries Association, Inc. Undated. Secondary Smelters and Refineries
in the United States. New York, N.Y.
Little, P., and R.D. Wiffen. 1978. "Emission and Deposition of Lead from
Motor Exhaust--!!. Airborne Concentration, Particle Size and Deposition
of Lead Near Motorways." Atmos. Environ. 12:1331-41.
Ludwig, F.L., and J.H.S. Kealoha. 1975. Selecting Sites for Carbon Monoxide
Monitoring. EPA-450/3-75-077, U.S. Environmental Protection Agency,
Research Triangle Park, N.C.
Ludwig, F.L., J.H.S. Kealoha, and E. Shelar. 1977. Selecting Sites for
Monitoring Total Suspended Participates. EPA-450/3-77-018, U.S. Environ-
mental Protection Agency, Research Triangle Park, N.C.
Ludwig, F.L., and E. Shelar. 1978. Site Selection for the Monitoring of
Photochemical Air Pollutants. EPA-450/3-78-013, U.S. Environmental
Protection Agency, Research Triangle Park, N.C.
Lyons, W.A., and L.E. Olsson. 1972. "Mesoscale Air Pollution Transport in
the Chicago Lake Breeze." J. Air Pollut. Control Assoc. 22:876-81.
McCormick, R.A., and G.C. Holzworth. 1976. "Air Pollution Climatology."
In: Air Pollution (3d edition), Vol. I. A.C. Stern, ed. New York:
Academic Press.
Miller, D.F., et al. 1976. "Combustion and Photochemical Aerosols Attribu-
table to Automobiles." J. Air Pollut. Control Assoc. 25(6):576-81.
52
-------
Motto, H.L., et al. 1970. "Lead in Soils and Plants: Its Relationship
to Traffic Volume and Proximity to Highways." Environ. Sci. Techno!.
4(3):231-37.
National Academy of Sciences. 1972. Lead: Airborne Lead In Perspective.
Washington, D.C.
Oke, T.R., and F.G. Hannel. 1970. "The Form of the Urban Heat Island in
Hamilton, Canada." In: Proceedings of the WHO/WMO Symposium on Urban
Climates and Building CMmato'Fogy. WHO Tech. Note No. 168. Secretariat
of the World Meteorological Organization, Geneva, Switzerland, 1970.
Ondov, J., W.H. Zoller, and G.E. Gordon. 1982. "Trace Element Emissions
on Aerosols from Motor Vehicles." Environ. Sci. Technol. 16(6)318-28.
PEDCo Environmental, Inc. 1981a. Optimum Site Exposure Criteria for Lead
Monitoring. Draft. Contract No. 68-02-3013, Task 2.U.S. Environmental
Protection Agency, Research Triangle Park, N.C.
PEDCo Environmental, Inc. 1981b. Field Study to Determine Spatial Varia-
bility of Lead from Roadways. EPA-450/4-83-002, U.S. Environmental
Protection Agency, Research Triangle Park, N.C.
Provenzano, G. 1978. "Motor Vehicle Lead Emissions in the United States:
An Analysis of Important Determinants, Geographic Patterns and Future
Trends." J. Air Pollut. Control Assoc. 28(12);1193-99.
Rohbock E., H.W. Georgii, and J. Muller. 1980. "Measurements of Gaseous
Lead Alkyls in Polluted Atmospheres." Atmos. Environ. 14:89-98.
Slade, D.H. (ed.). 1968. Meteorology and Atomic Energy. U.S. Atomic
Energy Commission, Oak Ridge, Tenn.
Smith, A.E., et al. 1979. Development of an Example Control Strategy^ for
Lead. EPA-450/2-79-002 (OAQPS No. 1.2-123), U.S. Environmental Protec-
tion Agency, Research Triangle Park, N.C.
Turner, D.B. 1970. Workbook of Atmospheric Dispersion Estimates. Report
No. AP-26, U.S. Environmental Protection Agency, Research Triangle Park,
N.C.
Turner, D.B. 1974. Dispersion Estimate Suggestion No. 4. National Environ-
mental Research Center, U.S. Environmental Protection Agency, Research
Triangle Park, N.C.
U.S. Environmental Protection Agency. 1977. Air Quality Criteria for Lead.
EPA 600/8-77-017, Office of Research and Development, Washington, D.C.
53
-------
U.S. Environmental Protection Agency. 1978a. Guideline on Air Quality Models.
EPA-450/2-78-027. Research Triangle Park, N.tT
U.S. Environmental Protection Agency. 1978b. Supplementary Guidelines for
Lead Implementation Plans. EPA-450/2-78-038 (OAQPS No 1.2-104).Research
Triangle Park, N.C.
U.S. Environmental Protection Agency. 1978c. Supplementary Guidelines for
Lead Implementation Pians--Revised Section 4.3 (Projecting Automotive LelTd
EmlssionsHEPA 450/2-78-038a, OAQPS No. 1.2-104a, U.S. Environmental
Protection Agency, Research Triangle Park, N.C.
U.S. Environmental Protection Agency. 1983. Updated Information on Approval
and Promulgation of Lead Implementation Plans"Draft. 9246.00/73,76.
Office of Air Quality Planning and Standards, Research Triangle Park, N.C.
54
-------
APPENDIX A
A SUMMARY OF RECENT FINDINGS
ON THE CHARACTERISTICS OF LEAD EMISSIONS
Automotive exhaust contributes up to 90 percent of the measured atmo-
spheric lead. As a result of the dominating effect of automotive emissions
on atmospheric lead, a considerable amount of research has been directed
toward characterizing the exhaust particles and their environmental fate.
Provenzano (1978) has discussed a number of factors that influence automotive
lead emissions. Among the variables listed by Provenzano are the following:
1. The rate of lead emissions is dependent upon the mode of
vehicle operation, "...at higher speeds larger percentages of
lead burned are emitted. Emission rates averaged 33 percent
at 60 miles per hour but ranged from 27 to 71 percent at
45 mph to from 49 to 91 percent at 70 mph."
2. Average lead emission rates varied from 28 to 45 percent for
cars tested under city-suburban driving conditions.
3. Some of the retained lead is re-entrained into the exhaust
stream at high speeds. Full throttle acceleration to high
speed can produce emissions of 900 to 2000 percent of the lead
burned.
4. There is an increase in the lead emission rate as vehicles
accumulate mileage. Exhaust lead is deposited and accumulated
in the exhaust system.
5. Less lead is added to gasoline refined for winter use than is
added to gasoline for summer use. Less lead is added to
gasoline sold in the northern states.
Finally, Provenzano states, using 1975 data, for the entire United
States over 60 percent of motor vehicle lead emissions occur in urban areas.
Habibi (1973) has reported on the character of vehicular exhaust particu-
lates, noting the following:
1. Reports from early literature (prior to 1958) indicate
lead particle sizes range from 0.01 micron to several
millimeters in diameter. Under city driving conditions,
50 to 75 percent of lead exhausted is associated with
particles 5 microns and smaller in diameter. The size of
lead particles from the 1957 data has been questioned,
however.
55
-------
2. The effects of mileage accumulation on the automobile was
characterized in two ways: (a) Lead salt emission increased
with increasing mileage on the vehicle, (b) Particle size
distribution shifted; a higher percentage of lead was emitted
on larger particles as mileage accumulated on the vehicle.
3. The results of data including tests on 26 cars indicated
55 percent of the exhausted lead is associated with particles
>5 vm equivalent diameter.
4. Particle size and exhaust particle composition are related.
Large particles have a composition similar to exhaust system
deposits and are 60 to 65 percent lead salts; the major lead
salt is PbBrCl. The submicron particles contain more soot
and carbonaceous material with the primary lead salt being
2PbBrCl • NH/jCl.
Boyer and Laitinen (1975) state the mass emission rate of filter particu-
lates is about a factor of 20 higher when lead gasoline is burned in an
automobile without a catalytic converter than when nonleaded gasoline is
burned under the same conditions, but the mass of ether extractable organics
stayed about the same, indicating the increase in mass is due to inorganic
substances. In the same paper, they reported that although the gasoline
consumption rate doubled for a vehicle operating at 60 mph, as compared with
30 mph, the filter particulate emission rate was approximately 6 times
greater. Boyer and Laitinen reported that the filter particulates loading
was much higher for tests with leaded gasoline than for nonleaded gasoline.
The experimental design included sampling the engine exhaust with a cyclone
sampler to obtain a gross size separation of particles. Sampling was also
done with a filter sampler. The emission rate for particles collected by
cyclone was much more variable than for particles collected by filter. They
suggested this may mean the formation of small particles is dependent on
nonvariable factors such as engine condition, but the formation of large
particles is more dependent on variable factors such as condition of the
exhaust system.
Lead, as organic compounds, was reported by Rohbock, Georgii, and Muller
(1980) to be less than 1 to 9 percent of the total atmospheric lead based on
a study in Frankfurt/Main area in Germany. They note that gaseous lead/
particulate lead ratios are low at a site near a highway where automobile
engines are hot, whereas higher gaseous lead/particulate lead ratios occurred
in areas where automobile engines were cold and evaporative emissions would be
more likely. Higher ratios of gaseous to particulate lead were measured in
urban and residential settings (e.g., 2 to 7 percent of total lead was
gaseous lead). Inner-city air samples were 4 to 15 percent gaseous lead; air
in a garage contained about 30 percent gaseous lead of the total lead.
DeJonghe and Adams (1980) measured organic lead as 2 to 24 percent of total
lead, depending to a large extent upon the siting of the sampler. Samples
from rural locations had less than 1 percent organic lead, whereas samples
56
-------
collected near a gasoline station were 24 percent organic lead. It may be
deduced that gaseous lead may in some situations be an important source of
exposure to some portions of the population. The reference method for
measuring lead concentrations does not account for the gaseous lead because
this portion of the atmospheric lead passes through the filters.
Gerstle and Albrinck (1982) have reported the content of metal found
in typical municipal sewage sludges. Lead content of sludge was reported to
range from 80 to 26,000 mg lead per kg dry sludge. Gerstle and Albrinck
report that a large incinerator K200 t/d) could emit from 20 to 30 pounds of
lead per day depending on the amount of lead in the sludge and the efficiency
of the emission control system. The form of the lead may be lead oxide, lead
chloride, or elemental lead. Lead chloride is classified as intermediate in
volatility. At the temperatures that are reached in an incinerator (980° C,
1800° F), lead is potentially volatile. Greenberg, Zoller, and Gordon (1981)
reported lead emissions of 2.0 ± 0.84 g/d from a sewage-sludge incinerator
with a capacity of 540 to 610 kg of dry sludge per hour (~14-16 t/d). The
emission control devices at that incinerator were 99 percent efficient; the
lead content of the dry sludge was reported as 430 ± 20 yg lead per gram of
dry sludge. Gerstle and Albrinck reported the lead content of dry sludge
from 16 cities to average 1940 mg/kg; the median lead content from those
16 cities was 600 mg/kg.
57
-------
APPENDIX B
PLANT LOCATIONS AND PRODUCTION TRENDS
FOR LEAD PRODUCTION AND REFINING
An air monitoring strategy can be developed more rationally if there is
some appreciation for the production, use, and consumption of lead. Therefore,
some basic data regarding the amount of lead produced and consumed in the
United States are provided. The data presented in Table B-l indicate a
decline of approximately 15 percent in the amount of lead produced and
consumed over the past 5 years. The Bunker Hill smelter-refinery at Kellogg,
Idaho, terminated operations in 1981. Locations of primary and secondary lead
smelters are listed in Tables B-2 and B-3. Figure B-l shows the locations of
lead mines, and primary and secondary smelters in the United States. Mines
and smelters listed in Tables B-2 and B-3 are keyed to the locations shown in
Figure B-l.
TABLE B-l. U.S. LEAD PRODUCTION AND CONSUMPTION IN METRIC TONS
1977 1978 1979 1980 1981
Production
Domestic ores 537,499 529,661 525,569 550,366 445,535
Primary lead refined 486,659 501,643 529,970 508,163 440,238
Foreign ores 62,041 63,530 45,641 39,427 55,085
Consumption
Primary and
secondary 1,435,473 1,432,744 1,358,335 1,070,303 1,167,101
Source: Minerals Yearbook 1981--Lead, Bureau of Mines, U.S. Department of
Interior.
59
-------
TABLE B-2. PRIMARY LEAD PRODUCTION AREAS IN THE UNITED STATES
State
No.
District
County
Mines
Missouri
Idaho
Colorado
Idaho
Montana
Texas
Nebraska
Missouri
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Smelters
1 S&R
2 S
3 S
4 R
5 S«R
6 S&R
7 S&R
Buick
Magmont
Vi burnum
Fletcher
Mini ken
Brushy Creek
Viburnum
Indian Creek
Vi burnum
West Fork
Lucky Friday
Bunker Hill
Star Unit
Leadville Unit
Bulldog Mountain
(S) and refineries (R)
Kellogg
East Helena
El Paso
Omaha
Glover
Boss
Herculaneum
Iron
Iron
Iron
Reynolds
Reynol ds
Reynol ds
Washington
Washington
Crawford
Reynol ds
Shoshone
Shoshone
Shoshone
Lake
Mineral
Shoshone
Lewis & Clark
El Paso
Douglas
Iron
Iron
Jefferson
Source: Lead Industries Association, Inc., New York, New York.
60
-------
TABLE B-3. SECONDARY SMELTERS AND REFINERIES IN THE UNITED STATES
State
Alabama
California
Florida
Georgia
Illinois
Indiana
Kansas
Louisiana
Minnesota
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
City
Leeds
Troy
Carson
City of Industry
City of Industry
Los Angeles
San Francisco
Jacksonvi 1 le
Tampa
Tampa
Atlanta
Atlanta
Cedar town
Columbus
Chicago
Granite City
McCook
Savanna
Beech Grove
East Chicago
East Chicago
Indianapolis
Muncie
Whiting
Olathe
Baton Rouge
Heflin
Eagan
St. Louis Park
St. Paul
Source: Lead Industries Association, Inc., New York, New York.
(continued)
61
-------
TABLE B-3. (continued)
State
Mississippi
Missouri
Nebraska
New Jersey
New York
North Carolina
Ohio
Oregon
Pennsylvania
Tennessee
Texas
Virgi nia
Washington
No.
31
32
33
34
35
. 36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
City
Florence
Forest City
Omaha
New Brunswick
Newark
Pedricktown
DeWitt
Walk ill
Charlotte
Cleveland
St. Helenes
Lancaster
Lyons Station
Nesquehoning
Philadelphia
Reading
College Grove
Memphis
Rossville
Dallas
Dallas
Dallas
Frisco
Houston
Richmond
Seattle
Source: Lead Industries Association, Inc., New York, New York.
62
-------
s
CD
OJ
o
o
o
in
Oi
•i—
$_
+->
1/1
-o
(O
cu
cu
o
s-
o
oo
ai
CO
13
OJ
i.
U
3
TD
O
S_
CL
-a
03
a>
ai
en
63
------- |