EPA-600/9-76-007a
June 1976
DENVER AIR POLLUTION STUDY - 1973
Proceedings of a Symposium
Volume I
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, Environmental
Protection Agency, have been grouped into seven series. These seven broad
categories were established to facilitate further development and application
of environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related
fields. The five series are:
1. ENVIRONMENTAL HEALTH EFFECTS RESEARCH
2. ENVIRONMENTAL PROTECTION TECHNOLOGY
3. ECOLOGICAL RESEARCH
4. ENVIRONMENTAL MONITORING
5. SOCIOECONOMIC ENVIRONMENTAL STUDIES
6. SCIENTIFIC AND TECHNICAL ASSESSMENT REPORTS
9. MISCELLANEOUS
This report has been assigned to the MISCELLANEOUS series.
This document is available to the public for sale through the National
Technical Information Service, Springfield, Virginia 22161.
-------
EPA-600/9-76-007a
June 1976
DENVER AIR POLLUTION STUDY - 1973
Proceedings of a Symposium
Volume I
Edited
by
Philip A. Russell
Denver Research Institute
University of Denver
Denver, Colorado 80210
Grant Number R-980590
Project Officer
Lester L. Spiller
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF RESEARCH AND DEVELOPMENT
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences
Research Laboratory, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents neces-
sarily reflect the views and policies of the U.S. Environmental Protec-
tion Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
-------
TABLE OF CONTENTS
INTRODUCTION 1
AIRFLOW STUDY RELATED TO EPA FIELD MONITORING PROGRAM DENVER
METROPOLITAN AREA NOVEMBER, 1973
L. W. Crow 3
THE ANALYSIS OF AMBIENT DENVER AIR FOR ORGANIC VAPORS INCLUDING
CARCINOGENIC POM COMPOUNDS
P. W. Jones 31
LIDAR OBSERVATIONS OF ATMOSPHERIC PARTICULATES NEAR DENVER,
COLORADO
V. E. Derr, G. T. McNice, N. L. Abshire, R. E. Cupp, R. F. Calfee
and M. J. Ackley 51
NUCLEATION CHARACTERISTICS OF DENVER AEROSOLS
C. C. Van Valin and R. F. Pueschel 87
ATMOSPHERIC AEROSOL DYNAMICS: THE DENVER BROWN CLOUD
P. B. Middleton and J. R. Brock 101
X-RAY FLUORESCENCE ANALYSIS OF DENVER AEROSOLS
T. G. Dzubay, M. Garneau, 0. Durham, R. Patterson, T. Ellestad
and J. Durham 141
AN ANALYSIS OF PARTICULATES FROM THE DENVER URBAN PLUME USING
SCANNING ELECTRON MICROSCOPY AND ENERGY DISPERSIVE X-RAY
SPECTROMETRY
P. A. Russell and C. 0. Ruud 165
ill
-------
-1-
INTRODUCTION
The Denver Urban Plume, often referred to as the "brown cloud," is
one of the area's most aesthetically unpleasing features. Although the
brown cloud was observed in the early 1950's, its occurrence and severity
have increased with the city's rapid population increase and urban
development. It is usually associated with the Northeast section of
Denver, where there is a concentration of industries, railways, freeways
and a major power plant. During the late fall and winter months, when
severe temperature inversions occur during periods of low wind speeds,
the visible and olfactory characteristics of the cloud are easily noticed
by residents throughout the metropolitan area, particularly along the
South Platte River basin.
For a number of years, the composition of Denver's Urban Plume was
unknown, and it became obvious that its control was impossible without
a detailed study to determine its particulate and gaseous composition,
and the influence of meteorological conditions. In November 1973, a
coordinated effort was initiated by the U.S. Environmental Protection
Agency to investigate Denver's brown cloud. Participants in the study
included:
1. Atmospheric Aerosol Research Section, Atmospheric Chemistry and
Physics Division, Environmental Sciences Research Laboratory,
U.S. Environmental Protection Agency, Research Triangle Park,
N.C. (AARS-EPA)
2. University of Denver, Denver Research Institute (DRI)
3. Loren Crow, Consulting Meteorologist
4. Meteorology Research, Inc.
5. Meteorological and Air Pollution Control Commission, Colorado
State Board of Health (CSBH)
6. Region VIII, U.S. Environmental Protection Agency
7. National Oceanic and Atmospheric Administration (NOAA)
8. National Center for Atmospheric Research (NCAR)
-------
-2-
9. General Motors Research Laboratories
10. University of Washington
11. University of Texas
12. Thermo-Systems, Inc.
13. United States Army, Rocky Mountain Arsenal
14. Battelle Columbus Laboratories
15. IIT Research Institute
About fifty samplers of several different types were used to collect
over 500 aerosol samples, and several types of gaseous testing equipment
were used at nearly ten different locations. Meteorological data were
collected at all the field laboratories and at other sites in the greater
Denver area. Some of the extraordinary research conducted during the
investigation included air pollution measurements from an instrumented
aircraft; LIDAR observations conducted by NOAA; non-particulate organic
contaminants analyzed by Battelle; radiation measurements made by NCAR;
and non-volatile particulate analyses, using scanning electron microscopy/
energy dispersive X-ray spectrometry, conducted by the Structures
Laboratory of DRI.
Prior to the research activities, DRI was active in coordinating the
preliminary efforts with EPA and the State of Colorado, in establishing
monitoring sites and the sampling networks, and furnishing logistic
support. During the investigation, DRI maintained the air pollution alert
system, continued the logistic support, and operated its own environmental
laboratory in the field.
In June 1974, preliminary results of many of the studies were pub-
lished as preprints for the Air Pollution Control Association meeting.
Only preliminary results were reported because of the early date required
for publication and restrictive page limitations.
In March 1975, DRI, through an EPA grant, conducted a three-day
symposium; comprehensive research reports from the Winter 1973 Denver
Urban Plume Study were presented. Volume I is the first of two volumes
that contain the proceedings of the symposium. DRI was responsible for
editing. Volume II will be published at a later date.
-------
-3-
AIRFLCW STUDY RELATED TO
EPA FIELD MONITORING PROGRAM
DENVER METROPOLITAN AREA
NOVEMBER, 1973
Loren W. Crow
Consulting Meteorologist
Denver, Colorado 80210
ABSTRACT
This paper describes the detailed hourly airflow, both at the
surface and at elevations above the surface, related to air pollution
episode days in Denver, Colorado, during November, 1973. Progressive
stages of the depth of the well-mixed layer near the ground are
documented. A near reversal of airflow on the more dense air pollution
days is a repeatable phenomenon. Rapid dispersal sometimes occurs when
strong winds assume control and move the polluted air rapidly away
from the city. There is a tendency for the "heat island" to develop a
"chimney effect" which carries effluents away from the metropolitan area
at a level above the ground to the north-northeast of Denver.
INTRODUCTION
This report covers a summary of airflow patterns related to
several air pollution episodes during November, 1973. This report will
emphasize data collected for November 14, 16, 17, and 21. The sponsor-
ing agency was the Atmospheric Aerosol Research Section, Chemistry and
Physics Laboratory, Environmental Protection Agency, National Environ-
mental Research Center - Research Triangle Park, North Carolina. The
author assisted in selecting appropriate days for concentrated
collection of air samples throughout the Denver metropolitan area. It
was also his responsibility to coordinate the collection of wind data
by use of pilot balloon (pibal) ascents at three locations on sampling
days and to collect a photographic record from advantegeous points
throughout the city.
-------
-4-
This airflow study adds to the body of data previously collected
during special study periods1' ''' in the Denver metropolitan area.
DATA SOURCES AND THEIR LOCATION
The geographic area of greatest concern for air pollution in the
Denver metropolitan area is primarily a segment of the Platte River
drainage extending from a few miles south-southwest to a few miles
north-northeast of the city center. The map shown as Fig. 1 identifies
most of the data collection points used in this study. The area shown
covers approximately 25 miles square. The Atomic Energy Commission's
facility at Rocky Flats, which is operated by Dow Chemical Company, is
located at the northwest corner of the map. The town of Henderson is
near the northeast corner and the Arapahoe County Airport is near the
southeast corner.
Two elevation contour lines, one at 5200 feet and the other at
5500 feet, indicate the downward slope of the terrain on both shoulders
of the Platte River drainage. Although not shown, a 6000-foot contour
line runs near the west edge of the map. There is a minor drainage
basin along Sand Creek, which runs from just west of Buckley Air National
Guard Base (ANGB) toward the northwest past Stapleton. The surrounding
terrain features help influence wind directions within that small drain-
age basin. Also, the combination of Clear Creek which drains from the
mountains near Golden toward the northeast, and Ralston Creek which
drains from the northwest toward Arvada, helps influence local windflow
patterns in that portion of the metropolitan area.
The locations where pibals were released are identified at the
cross points in the seven-point vertical scales at Arvada, Yellow Cab,
and the EPA Trailer site. The major highways and thoroughfares which
are shown carry a major portion of the automobile traffic in the metro-
politan area.
Since early 1965 Denver has been one of the cities where air
pollutants are being measured in the Continuous Air Monitoring Program
(CAMP). Denver's station is located at 2105 Broadway.
The types and frequency of observations and photographs collected
are indicated in Table I. Data for four days are being emphasized.
IMPORTANCE OF AIR STABILITY
When strong winds prevail in the Denver metropolitan area there is
near complete ventilation and a horizontal transport of air moving into
the city from one side and out and away from the city on the opposite
side. This type of ventilation occurred on November 14 and 21, 1973.
On November 14 ventilation occurred as air moved away from the city
-------
-5-
Fig. 1. Identification map.
-------
-6-
XXXXXX XXXXX
X XX
XXXXXX XXXXX
XX XXX
w
u
10 I- CM
XXXXXX X X X X X CN
04 •
H m
E-i r-
iJ tr.
Sw
PH ffl
fc* s
Q !>
W O
r)
a «
a o
O
ioun)
OIr-l
ni
!a)
cQ«^
(nid
ro
>,'O
jq in
a)Hrt;
Sfl:u
10
ft
"00)
CH
CO-H
b>
2Ofl:
w
ffi
(NT
OH
in
4-1
fi
tj>
C
-rl
g
O
CO
X X
X
-H
13
U 01
M k •
•H *H O
'rt TI 01
-iH
CO CN S
OO
-P-U
-
C0
iu >i t}
1)
4J
01 O
-P £ S
•H H j3 M
Id CE, 3 O
M -p
EH >i Id •
& M S
< CJ CU
^ o .^ •
W « W rf
-------
-7-
toward the east. On the 21st the city was ventilated by air moving
rapidly from east toward west. An illustration of such ventilation is
shown in the upper portion of Fig. 2.
W
o
9>
O
w-
Cli
a n
,_ Air _
1
N
Ventilation
N
s
0
u
R
C
E
West
East
Pig 2. Sketches of two alternate paths for
cleansing mechanisms.
The transport of air pollution, both from elevated sources,
ground sources, line sources, and area sources, is highly dependent on
the stability of the layer of air into which they are introduced.
Elevated plumes emanating from sources near the southern edge of the
metropolitan area can often be identified moving toward the north in
ribbon-like plume patterns. Each individual plume reaches its effective
stack height soon after being emitted from the source. As the plume
moves downslope it assumes an elevation somewhere between the initial
elevation above the ground and a continuing constant elevation with
respect to sea level.
A compound plume of pollution can be formed by several individual
elevated plumes. A similar compound plume can also be produced as a
layer of pollution is raised above the ground level as it passes over
-------
the heat island of concentrated buildings and/or highways near the
center of a city. Heat from the burning of 200,000 to 400,000 gallons
of gasoline during daily rush hour traffic periods helps to expand the
heat island. Much of the polluted air ascends to the upper part of the
heat island and then travels horizontally downwind at some layer above
the ground to the north-northeast. This combination of upper level
outflow is illustrated as a chimney effect in the lower part of Pig. 2.
Such a mechanism of outflow is fed by a net inflow toward the center of
the metropolitan area near the surface during early forenoon hours. An
earlier study found the net inflow rate to be approximately one-half
mph toward the center of the city.
During forenoon hours it is often possible to identify separate
layers of pollution. A polluted layer of air having a local source near
the ground can often be separated from an elevated pollution plume by
a non-polluted layer of clean air between.
The early morning temperature profiles from the 0500 Stapleton
Airport rawinsondes in the lowest 1500 feet are presented in Fig. 3.
The sloping dashed lines indicate the dry adiabatic lapse rate and the
depth of well mixed air which can be estimated for the times shown,
based on the subsequent surface temperature observations made during
the forenoon at Stapleton Airport. The upper portion of the 0500
temperature profile would have been abruptly changed at the time of
cross-city ventilation near noon on November 14 and November 21.
Using an approximate composite of the 12 low level a.m. soundings,
the corresponding resultant patterns of pollution layers can be esti-
mated. Three schematic diagrams which illustrate the influence of
stability factors are shown in Fig. 4. Under stable air conditions
each individual elevated plume moves off horizontally at its effective
stack height with no mixing to the ground until surface heating is able
to mix the layer up to and including the height range of that particular
effective stack height above the terrain over which the plume was moved.
This is illustrated in Part A of Fig. 4.
In Part B of Fig. 4, a comparative estimate is made of the tempera-
ture profile as it would occur within the heat island segment of air
near the north edge of the downtown portion of the city. The influence
of the heat island does not eliminate the air stability, but it can
produce an elevated compound plume for visible pollutant materials
moving farther downwind beyond the main portion of the heat island area.
In Part C of Fig. 4, indication is given as to how the top of a
visible polluted layer emanating from ground sources increases as sur-
face temperatures increase during forenoon hours. When the air mixes
upward through sufficient depth to envelop the previously identifiable
plumes, separate layers can no longer be distinguished.
The nearly repeatable pattern of pollution over Denver on episode
days is illustrated in the cross-section diagram of Fig. 5. This
-------
-9-
- 780MILIBARS
I 790
- 800
- 810
-820
- 830
NOV. 14
05OO MST
- 800
NOV. 16
810 0500 MST
- «20
- 830
800
NOV. 17
810 05OO MST
820
830
- 820
- 790
NOV. 21
800 0500 MST
- 810
\
-5°C.
23°F
I
0°
32°
I
5°
41°
V?
\
\
10°
50°
r
15°
59°
o
z
o
IT
O
UJ
CD
LJ
Ul
////J
- 1500'
- 1000*
-500'
- 1000'
- 500'
h 1000'
- 500
///SS
- 1000'
- 500'
Fig. 3. Early morning temperature profiles for four
pollution episode days during November, 1973, as
measured by rawinsonde at Stapleton Airport.
-------
-10-
-5°C.
-5°C.
-5«C.
Z3°F.
0°
B
0°
INDIVUAL PLUME STABLE AIR
10°
W
-*
llk>
3jS COMBINED PLUMES
MOVING DOWNWIND FROM
CITY HEAT ISLAND
(COLDER AIR BELOW)
10°
TOP OF
•"".XEO.IJIY.ER
\
-1000'
-50O1
0°
32°
5°
41°
10°
50°
Fig. 4. Relationships of temperature profiles and polluted layers for:
A. elevated source moving hroizontally at effective stack height in
stable air; B. city heat island effect producing elevated combined
plume in stable air; C. comparative tops of polluted layer from ground
level sources with increasing surface heating during forenoon hours.
-------
-11-
ROCKY FLATS
PLANT AEC
8000 -i
-6000
Pig. 5. Typical WNW-ESE profile of pollution layer depths over metro-
politan Denver between 0800 and 1500 on days when temperature inversions
are not eliminated.
-------
-12-
cross-section has been prepared to show the vertical profile of terrain
and the pollution layer at various hours of the day as it is typically
observed. The profile, which is greatly distorted in the vertical
scale, runs from the mountainous area to the west-northwest of the Rocky
Flats Plant southeastward to Buckley ANGB. This profile crosses the
Platte River about three miles northeast of the downtown City of Denver
and passes through Stapleton Airport. Multiple photographs, taken during
the special study period of November, 1973, and during other periods
studied by this author, support the approximate elevations of the
polluted layer tops as shown in the cross-section for the hours from
0800 through 1500. At both 0800 and 1000 the heat island effect pushes
the upper boundary of the polluted layer somewhat higher over central
Denver. The polluted layer depths should be considered typical of days
in which the temperature inversion is NOT broken. In the summer months
the respective tops of the polluted layer may be slightly higher.
The ground level for most downtown buildings is approximately 5200
feet above mean sea level. The higher buildings extend slightly more
than 400 feet above ground level. Thus, they extend approximately 270
feet above the runway elevation for Stapleton Airport, which is 5331
feet. The elevation of the polluted layer tops is shown as above ground
level at Stapleton Airport since the surface temperatures used to help
derive the elevation tops are those measured hourly at the airport.
Temperature profiles from the 0500 rawinsonde at Stapleton have been
used to support the findings shown in Fig. 5.
Although not shown in this cross-section the sequence of air motion
changing from downslope to upslope between 0800 and 1500 must be con-
sidered in estimating the top of the polluted layer. The most frequent
direction of smoke layers as seen by the weather observer at Stapleton
Airport is to the southwest through north. There are relatively few
instances when the observer indicates that either smoke or haze is
observed in all directions from his point of observation.
The most frequent pattern of spreading for the pollution layer
from noon through 1500 is to the south and/or west of the downtown
area. This is aided by upslope motion above the higher terrain located
in that direction from the city. For instance, North Table Mountain,
South Table Mountain, and Green Mountain all extend above 6000 feet MSL.
At the tops of these relatively flat protrusions the layer of stable
air in early morning hours is very thin. Thus, a small amount of sur-
face heating is required to eliminate the stable air layer. The stable
layer is replaced by air which is well mixed vertically. In some
instances a chimney-like airflow pattern is produced. Air which is
heated and rises above the low foothills may move back toward the east
above the temperature inversion over the valley. This upward motion
of air above the higher terrain where the temperature inversion has been
eliminated induces a movement of air coming first southward and then
westward from the downtown area of Denver. Before it reaches the
mountainous terrain, vertical mixing often has destroyed the temperature
inversion, and the upper edge of the polluted layer becomes quite diffuse.
-------
-13-
The top of the polluted layer is best characterized as having small
undulations rather than being absolutely flat. An instantaneous photo-
graph often makes it appear as quite level but time-lapse photography
shows variations in the upper level.
The frequency with which the western edge of polluted layer reaches
Rocky Flats is less than one-third of the days when relatively dense
pollution can be noted in the downtown area of Denver. Likewise, there
are very few instances when the polluted layer envelops Buckley ANGB.
When a layer of pollution extends over the downtown area it corres-
pondingly decreases the rate at which surface heating can be produced by
solar radiation. Therefore, by the early afternoon hours it is generally
true that the surface temperatures are cooler under the polluted layer
than in the surrounding terrain where clearer air has permitted more
rapid surface heating. Thus, the profile pattern of Fig. 5 shows the
outer edges of pollution to be somewhat higher during afternoon hours.
The higher ground surfaces above mean sea level near the outer rim of
the basin, especially to the south and west of Denver, also help to
support an earlier elimination of the temperature inversion with the
same amount of solar radiation.
Although the peak density near the ground occurs most frequently
near the end of the morning rush hour period for sampling stations near
the center of the metropolitan area, the deepest layer of visible
pollution generally occurs between noon and 1500. The aerial photograph
presented as Fig. 6 was taken November 9, 1973, at approximately 1200
from an elevation of 7000 feet. The layer of pollution extends approxi-
mately 1100 feet (335m) above street level in the center of the city at
that time. An abbreviated record of a typical airflow reversal day,
November 9, is shown in Fig. 7. There was downslope drainage at 0300,
upslope reversal at 1100, and stable conditions at both 0500 and 1200.
DETAILED POLLUTION EPISODE PERIODS
The following episode periods have been examined to determine the
airflow patterns both near the surface and through several hundred feet
above the surface. The stability has been determined by temperature
profiles obtained from the Denver Airport rawinsonde with some supple-
mental aircraft soundings. Individual hourly maps of surface winds
show a high variability of airflow at different times for each day.
Pibal runs were made on an hourly basis at the EPA Trailer during
daylight hours beginning at 0900 on most episode days. At the other
two locations pibal runs were made each half hour beginning at 0900.
At all three locations theodolite readings were made at 150-foot
intervals along the vertical ascent path.
In addition to the fixed time photographs which were taken from
various places throughout the city, an additional record of the changes
in visible pollution layers was obtained using time-lapse camera
-------
-14-
Fig. 6. Aerial photograph from aircraft at 7000 feet MSL near noon
on November 9, 1973. Scene is toward the southwest.
-------
-15-
- 760 MB
780
-800
-820
C°-5
DENVER
RAWINSONDE
0500 MST
F°23
32
50
1 I
39
2500-,
2000-
1500-
1000-
500
- 760 MB
5331'
MSL
-780
hBOO
-820
C°-3
DENVER
RAWINSONDE
1200 MST
F°23
32
50
59
2500
2000-
N8
I50O-
(000-
>-^
6
500-
4
20
5331'
MSL
68
Fig. 7. Example of surface airflow patterns during: a. nighttime
downslope converging drainage flow at 0300, and b. daytime upslope
flow at 1100 on November 9, 1973. The temperature inversions was not
broken that day.
-------
-16-
techniques. One camera, using Super 8 film, took pictures from the 13th
floor of the Bureau of Reclamation Building located near the west edge
of the metropolitan area. The second camera, using 16 mm film, took
pictures from the roof of a hanger located at Buckley ANGB,
Episode Day November 14
The day of the 14th was forecast to be a relatively light air
pollution day. However, it was the first day that aircraft sampling
facilities became available and wind measurements were made to fit with
the aircraft measurements.
During the early forenoon hours the air pollution buildup was
typical for a weekday morning. However, the temperature inversion
which existed at 0500 was eliminated before noon and the layer of air
near the surface was mixed with strong synoptic airflow from the west-
northwest. Synoptic linkage occurs when the pressure airflow identifi-
able on synoptic weather maps is sufficiently strong to override all
local terrain control features.
One photographic record of the air mass changes that occurred
shortly before 1100 on November 14 was obtained from the observing site
on the eighth floor of the Sheraton Inn-Denver Airport. A set of before
and after photographs is shown in Fig. 8. The scene at 0900 toward the
northwest shows a relatively dense layer of pollution in the north
portion of Denver. The lower photograph in Fig. 8 shows markedly
improved conditions for the same scene.
A set of three photographs taken to the west of the EPA Trailer
also helps document a sharp change near 1045. The time of these three
photographs is 0930, 1022, and 1100. Note the sharp change in improved
visibility between 1022 and 1100 shown in Fig. 9.
The time-lapse photography at Buckley ANGB also confirmed the
rapid change which took place as the pollution layer was moved away
from the city toward the east.
A set of 13 stations was used to plot same-hour winds to show air-
flow patterns near the surface in the Denver metropolitan area. The
sequential order of the wind measuring stations used is shown in Fig. 10.
The first six stations are on the left shoulder of the Platte
River with the Overland and Welby stations being very near the river.
Only three of the nine observation points within the Rocky Mountain
Arsenal are listed. The seven stations on the right shoulder are
arranged sequentially from lowest toward highest elevations.
The hourly sequence of airflow at the 13 stations shown in Fig. 11
illustrates the abrupt change which took place between 1000 and 1100
-------
-17-
Fig. 8. Contrast between polluted air on a weekday morning and same
scene at noon following one hour's linkage with synoptic airflow from
west to east and elimination of the temperature inversion, November 14,
1973.
-------
-18-
Fig. 9. Sequential photographs looking west from EPA Trailer during
forenoon of November 14, 1973.
-------
-19-
Fig. 10. Sequential order of thirteen wind measuring stations for which
hourly data are used to identify airflow patterns. Left shoulder of
Platte River includes six stations from Rocky Flats through Welby. Right
should group includes stations from ARS. 9 through Arapahoe County.
-------
HOURLY WIND DATA
HOUR D<** JEFF.CO. ARV. CARIH OVE.
-\ - t—~ - MKG
•II—. -
02 -13
04- --! " !*
^ .17 - 6 - " - »*•
' i - ^« - ^-5 W - **-
/o .^—. /c -
- vi
•W
«X X
«~ - "*v
7^
\ffc- "^ _
^_ ije
«^» "*?-
y ' ~^. ,
H— ' -~ -
NOV. 14
ARAP.CO.
Fig. 11. Hourly sequence of surface wind measurements for November 14,
1973.
for six stations, most of which were in the western portion of the city.
The abrupt change for the five wind measuring stations in the eastern
portion of the metropolitan area took place between 1100 and 1200. The
wind is plotted as follows at each station. The direction indicator
ends at a center point for each hourly plotted value. The number at
the outer end of the direction indicator shows the wind speed. For
instance, a wind from the east at 10 mph is shown as • to. A wind from
the northwest at 15 mph is shown as /5x«. . "C" stands for calm. "G"
followed by a number indicates that the peak gusts in the previous hour
reached the level indicated.
Hourly temperatures in °F. at Stapleton Airport are plotted to the
left of the Stapleton winds. Relative humidity, if above 70 percent,
is plotted to the right of the Stapleton winds.
All the hours from noon through midnight on November 14 show a
nearly constant flow path over the metropolitan area while strong winds
prevailed. This is a good illustration of across-town ventilation.
-------
-21-
Episode Days November 16-17
The pollution episode for November 16-17 was primarily concentrated
on Friday, the 16th. The airflow patterns on the 16th showed a reversal
from downslope to upslope occurring midforenoon and a gradual thinning
of pollution density during afternoon hours as the temperature inversion
was eliminated.
The set of three photographs in Fig. 12 gives a brief visual
sequence of changes in dimension and location of pollution on that day.
The photographs at the top and bottom of Fig. 12 show the scene to the
northwest from the roof of a tall building in southeast central Denver.
At 0840 the upper edge of the visible pollution layer is not yet as
high as the tops of the taller buildings in downtown Denver. Note that
there is more pollution at the right edge of the 0840 photograph than
at the left edge. By 1305 the depth of the polluted air is much greater
but the density is less.
The center photograph in Fig. 12 was made at the EPA Trailer
toward the west at 1015. An elevated plume of more dense particulate
material can be identified just above the trees in the near foreground.
The temperature at Stapleton Airport was still only 40 at 1000 and the
top of the stable layer was at least 300 meters above the ground.
Forced vertical mixing would have been limited to somewhat less than
200 meters at that time.
The hourly sequence of winds at 13 stations is shown in Fig. 13
for the two-day period of November 16 and 17. The beginning of airflow
from north to south starts first on the 16th at the lowest elevation
stations near the north end of the metropolitan area. The flow of air
in the early forenoon hours of the- 16th moves the air from over the
metropolitan area toward the nearest mountains to the west. During
late afternoon of the 16th the airflow gradually changes from north-
easterly to southeasterly for most stations on the right shoulder of
the Platte River Valley. The airflow shifts around to a westerly com-
ponent on the left shoulder of the valley near 1700 and 1800. Both the
morning and evening changes of wind direction take place during periods
of low wind speeds. The nighttime drainage period between the 16th
and 17th extended from near 1800 on the 16th through 1000 on the 17th.
The surface temperature inversion was eliminated before 1100 on
November 17. (See Fig. 3.) There was a three-hour period of synoptic
linkage during the afternoon of November 17 across the entire area.
Synoptic linkage occurs when the pressure pattern airflow identifiable
on synoptic weather maps is sufficiently strong to override all local
terrain control features. This midafternoon period of strong winds
ventilated the entire metropolitan area moving air away from the city
toward the east on the 17th.
-------
-22-
084O 11/16
Fig. 12. Photographic record (top and bottom) of contrast from early
dense pollution layer and disappearance toward west near noon. Middle
photograph shows elevated plume in stable air west of EPA Trailer at
1015, November 16, 1973.
-------
-23-
HOURLY WIND DATA
NOV. 16
NJH BUCK. ARAP.CO.
24-< —
02-1— '
NOV.17
-Vt
t :
-«-
- W-
r '"<" «!" ~v< "!« T »T ;(
-**. -,- -i/t -,- - v -,-*-,- -,- -,
\ t\ i i\ x« *': t7; 'I *i
"Yz ^T4 c^" "fvj "(/ ",|" *" ^i!7 "T* y
-^ -Y1' <^^ "|; "y< ;/"*'"N ;/" ;j
^^ " i"1 «"/' ." V" /.^" V!" 2'" t" ^" ",?
"r* 9
- 47 - , - - Vt
ft -/
-! ,x!
A . -\
.
,/ SYNOPTIC LINKAGE
»:
N -
vx. .!..„
~r -s..
1" t*~ ~ir ~r T' "v* T**"!*1 ~r "N :N"
-11^. -yl/ _y^ -y€ . . - -y - i - f, - ^- -y - . . - ^s.n-
T
- V -r- «• -/ •
•V V'1 t/r
-y+
Fig. 13. Hourly sequence of surface wind measurements for Nov. 16-17, 1973,
-------
Episode Day November 21, 1973
November 21 was the second day of a two-day air pollution episode
period following a nighttime snowstorm on November 19-20. Airflow after
0800 of the 20th remained from a southerly direction across the metro-
politan area. The flow continued to move downslope to the north-
northeast from the metropolitan area throughout the night of the 20th.
On the morning of the 2lst light wind speeds prevailed throughout most
of the metropolitan area. Stronger winds were recorded only at Rocky
Flats and Jefferson County Airport. The central portion of the city
showed a dense layer of pollution in the early forenoon hours. However,
this dense layer of pollution was moved horizontally toward the west
with the onset of stronger easterly winds beginning between 1100 and
1200 in the eastern part of the city and sweeping across the city by
1400. Figs. 14, 15, and 16 all confirm this type of air motion.
The upper two scenes of Fig. 14 were taken toward the west from
the Sheraton Inn-Denver Airport. The first scene at 1100 shows a dense
layer of pollution and very little of the skyline of higher buildings
in the central part of the city. The second photograph in this figure
clearly shows the downtown buildings as being visible and the foreground
essentially clear at 1200.
The lower portion of Fig. 14 shows a scene from the roof of a high
building in southeast central Denver looking toward the northwest.
This photograph was made at 1240. There is much more dense pollution
in the left edge of the photograph as compared to the right edge where
pollution is being eliminated by stronger winds moving from the east.
Three sequential scenes of pollution to the west of the EPA Trailer
between 1125 and 1405 on November 21 are shown in Fig. 15. The scene
at 1125 shows a diffuse layer of visible pollution across the Platte
River Valley at both edges of the photograph. By 1215, the middle
scene, the foreground is much clearer than at 1125. By 1405 the moun-
tain profile directly west of the EPA Trailer is quite distinct and the
path of view is quite clean compared with the more dense pollution
identifiable at the left edge of the photograph. Soon after the onset
of the stronger winds from the east a deck of stratus clouds moved over
the metropolitan area. This cloud layer is very noticeable in the
lower photograph of Fig. 15.
The sequential pattern of hourly winds on the 21st is portrayed
in Fig. 16. When the direction was highly variable a small "v" was
used as an indicator. Note that in the nighttime period from 0100
through 0800 there are several stations without distinct indications
of direction. Airflow from a northerly component beginning near 0600
and continuing through 0800 at several stations indicates inflow from
the northern portion of the city to feed the chimney effect of a rising
layer of pollution over the heat island in the central part of the city.
The abrupt change in wind speed and the nearly constant direction from
-------
-25-
1200 11/21
1240 11/21
Fig. 14. Set of upper two photographs taken one hour apart from Sheraton
Inn toward west-southwest, plus single photograph to northwest from
central southeast point which shows progress of polluted layer from east
toward west, November 21, 1973.
-------
-26-
•> »»f ^f
Fig. 15. Sequential photographs looking west from EPA Trailer from
1125 through 1405, November 21, 1973.
-------
-27-
02 jj . - ift-
i/>, vx>
."\. ,J. ' -v
1 ' ^/!
oe - - . - **?- - it-
10 - - 1 - - . - - ' -
" "\ \ ?v
,N; W *v,
12- - .-' -,- V
-,-/- --' 1 -^
I4--X3 -t-s, -JN, -
- - \v - \ - r' '
• H • i i
"--(; -M -?- -
' "I* ^N« •>! '^
"1 8N1 *X IN
22-l2X 'X. -/' -
V1 'v
24- X - - , A -
'% -VI*- - i'* tx
J. - **- - I -
/V - ' - - vt
;V ! ^
\; >(. ^ * k* !» !<
, ^ < ^ vf !( !<
°N - X -•»-/- - VJ. -.- -OS
> : !> ,' * : < , o
X - . - -/-»-/- -M. -6- ->^,~ .
.' ^\ . 3(, _ X *- v* - , - - Vj, • 10
. ' 37 -x* - :/^ Jv .^ .
4 Xt. > f ' .v , X, . ,N*,
~~~^~IO ^^10 \i i i ' 1 \
SYNOPTIC i^™ r~* : i '" '
^10 EASTERLY X^ N» \^ ^,0 ~^lf
^ ._,. -^Jf-e- -«, -p. -^,-i.
^ fc^ „'< * ^f > Ns Nf
^r* -^_ ^._3i/_ _y^L _.- - v - -
, fc >' N N' ; !3 N
^ - , - -.,-31 - >_7 -Ut- -£- -j- - 20
x' : ' i : i i i
2! ' ' C
X - 1 37 - /' - J - -£- N, -22
» - N Is/ 1 - » ^ - - ^- - 24
Fig. 16. Hourly sequence of surface wind measurements for November 21,
1973.
an easterly component is easily traced between 1200 and 1400 as the
different wind measuring units show the changes in both wind direction
and speed. On this day the synoptic flow from an easterly component
ventilated the central part of the city and moved the polluted layer
toward the southwest corner of the metropolitan area where it continued
to be visible through mid-afternoon hours.
SUMMARY
During pollution episode days in Denver, Colorado, it is typical
that airflow nearly reverses from downslope to upslope during late
forenoon hours. A return to drainage flow takes place gradually in the
early evening hours. These multiple changes in general airflow elimin-
ate any possible use of a Steady State Plow concept for dense pollution
days.
-------
-28-
Depth of the visible polluted layer is directly related to surface
heating during daylight hours. The top of the polluted layer matches
the upper end of a dry adiabatic lapse rate line as it progresses up-
ward along the morning temperature inversion profile with warmer surface
temperatures.
During early afternoon hours there is frequently a detectable air
motion carrying polluted air from over Denver toward the west and/or
southwest. This air moving toward the mountains replaces air which has
already been vertically mixed upward along the relatively warmer moun-
tain slopes. The inversion is broken first over the higher terrain
where less solar heating is required to eliminate the more shallow
early morning temperature inversions.
There is good evidence that during forenoon hours "Heat Island"
effects are often observed to produce an elevated plume moving beyond
the downtown area toward the north-northeast. This "Heat Island Plume"
continues until surface heating eliminates the coldest layer near the
ground.
There is some evidence that a "chimney effect" is produced near
the north end of the "Heat Island" in the hours from 0600 through 1000.
This "chimney effect" can help draw air from north to south near the
surface in the lower part of the valley a few miles north-northeast of
the downtown area.
ACKNOWLEDGEMENTS
This work was sponsored by the Chemistry and Physics Laboratory/
Office of Research and Monitoring of the U. S. Environmental Protection
Agency.
Many cooperating agencies and interested persons collected data on
an intermittent basis during November, 1973. Only a limited number of
individuals who were particularly helpful in collecting meteorological
data are listed here.
Personnel from the Colorado Department of Health, Air Pollution
Control Division, who very willingly worked on weekend days included
Arthur Adams, Charles Bray, Richard Fox, William Retallack, L. Todd
Reynolds, Norman Weigel and David Wickham.
Students from Metropolitan State College who assisted in data
collection were Steven Geohegan, Ronald Rutherford, Timothy Smith and
Dennis Tarrin.
Aerial photographic missions were conducted by photographer
Charles Grover.
-------
-29-
Personnel from the Regional Office in EPA included William Cogger,
James Harris and James Lewis.
Clarence Everson of the Bureau of Reclamation and Sgt. James Summey
of Buckley ANGB generously volunteered their time to operate the time-
lapse cameras.
Many surface photographs were taken by Fred Crawford of Dow Chemical
Company at Rocky Flats and David Netherby of the Sheraton Inn-Denver
Airport.
The author acknowledges with thanks the willing cooperation of
many other persons and companies who collected wind and temperature data
from various locations throughout the Denver metropolitan area.
REFERENCES
1. Riehl, H., & L. W. Crow, A Study of Denver Air Pollution, Atmos-
pheric Science Technical Report No. 33, Colorado State University,
1962.
2. Crow, L. W., Airflow Related to Air Pollution at Colorado Springs,
Denver, Durango, Grand Junction, and Pueblo, for State Department
of Public Health, State of Colorado, LWC #44, 1964. (Not published.)
3. Reiter, E. R., N. Djordjevic, W. Ehrman, G. Swanson, Further Studies
of Denver Air Pollution, Department of Atmospheric Science, Colorado
State University, ASP #105, 1966.
4. Air Pollution in the Denver Area, pamphlet published by Public
Service Company of Colorado, text prepared by L. W. Crow, Consulting
Meteorologist, 1967.
5. Riehl, H., D. Herkhof, Weather Factors in Denver Air Pollution,
an abridged version of the final report to the U. S. Department of
Health, Education and Welfare, Department of Atmospheric Science,
Colorado State University ASP #158, 1970.
-------
-31-
THE ANALYSIS OF AMBIENT DENVER AIR FOR ORGANIC VAPORS
INCLUDING CARCINOGENIC POM COMPOUNDS
Peter W. Jones
Battelle
Columbus Laboratories
Columbus, Ohio 43201
ABSTRACT
This paper describes the collection of atmospheric vapors using a
chromatographic adsorbent and the subsequent analysis of the sample by
gas chromatographic-mass spectrometry. Samples were collected from dif-
ferent locations in the city of Denver, Colorado; comparisons with other
urban atmospheres were made where appropriate.
Initial studies concerned a general survey of the nature of com-
pounds found in the ambient Denver atmosphere. Many pollutants most
probably arise as a consequence of fossil fuel combustion, but others
appear to be a consequence of specific industrial operations. The for-
mer compounds are found in every other location studied.
The second phase of this study involved the analysis for polynuclear
organic materials in Denver air, particular emphasis being given to the
search for carcinogenic species. Comparison is made with ambient Colum-
bus air at a location near an asphalt roofing operation.
INTRODUCTION
The analysis of organic vapors in urban atmospheres has attracted
surprisingly little attention, in contrast to the steadily increasing
data pertaining to organic particulates. However, a number of collec-
tion and analysis techniques have been reported recently. Several cryo-
genic sampling methods have been reported, but this method invariably
lacks sensitivity due to the poor sample concentration which may be ob-
tained because of the presence of large quantities of water and low-
-------
-32-
molecular weight hydrocarbons present in the atmosphere ^.
The more efficient methods described for the collection all involve
solid absorbents, principally: activated charcoal , carbosieve ,
Poropak Q^, Chromosorb 102",14 ^ jenax GC^-. Charcoal and Carbosieve
are the least satisfactory for high sensitivity sample collection, since
both retain appreciable quantities of water; furthermore, charcoal has
a very low sampling capacity, while both are difficult to desorb unless
solvents are employed. Desorption with solvents precludes high sensitiv-
ity because of sample dilution and the risk of contamination. Chromosorb
and Tenax GC appear to be the most promising adsorbents which have been
used to date. Dravnieksl3 and Crittenden1^ have described efficient collec-
tion systems using Chromosorb 102, the former carrying out analysis by
gas chromatography following hot desorption of the sample into a low-
temperature syringe needle and the latter attempting direct high reso-
lution mass spectrometric analysis of the desorbing sample. In both
cases it appears that relatively small volumes of air were sampled and
high sensitivity was not sought. Crittenden collected both particulates
and vapor simultaneously, and while the high resolution mass spectro-
metric analysis of particulates is particularly elegant, no evidence is
presented which would suggest that this is an effective means for analy-
sis of organic vapors. There is no doubt that Tenax GC is also an
efficient absorbent, as demonstrated by Zlatkis for a variety of systems.
The high-temperature limit of Tenax may prove more useful than Chromo-
sorb in some systems. The sampling of large volumes of air has not been
attempted with Tenax, but it should be comparable to Chromosorb. In
this program, Chromosorb 102 has been employed as adsorbent, and up to
30,000 liters of air has been routinely sampled as described in the
following section.
The present study is primarily concerned with the determination of
compounds with six or more carbon atoms; it is particularly directed
towards the detection of oxidized hydrocarbon species and the detection
of hazardous polycyclic organic materials (POM) . The principal urban
location studied was Denver, Colorado. Limited analyses were also
carried out on vapor samples collected at St. Louis, Missouri, and
Columbus, Ohio, for comparison with the former data. Lastly, a simple
diurnal study of pollutant concentrations was made in Columbus, Ohio,
in an attempt to estimate their possible origin. Some evidence for am-
bient photo-oxidation of gasoline-type hydrocarbons was obtained in this
experiment.
EXPERIMENTAL
Preliminary experiments were carried out with Chromosorb 102 chroma-
tographic columns in which retention time data for typical anticipated
pollutants were obtained at various temperatures. Optimum dimensions
were then calculated for ambient atmospheric sampling columns (15 cm
long, 1.5 cm radius), and several were fabricated and packed with approx-
imately 35 g of Chromosorb (Figure 1). The sampling columns were
-------
-33-
O
O
(D
oo
O
cc
LU
I
CL
00
O
OJ
u.
3
01
-------
-34-
initially conditioned at 200 C with a nitrogen flow rate of 500 ml
min for 48 hours.
The design of the sampling columns permits the incorporation of a
filter holder at the entrance of the column, by means of a 15 mm glass
and Teflon "Solv-Seal" joint. A 4-inch glass fiber filter was routinely
employed. Filtered ambient air may be drawn through the sampling
columns at approximately 12 liter min~^ by means of a Cast 0522 rotary
vacuum pump. In this study the filter was discarded.
Prior to sample desorption, traces of water (and some lower molecu-
lar weight hydrocarbons) were removed from the sampling columns by purg-
ing with prepurified dry helium (liquid-nitrogen trap in series with an
activated 3A molecular sieve column), in the same direction as the origi-
nal sampling procedure, for 30 minutes at 2 liter min . During this
procedure several internal standards were added during some experiments
in order to facilitate the computerized gas chromatographic-mass
spectrometric quantification of POM compounds, as described later. De-
sorption of the collected sample is achieved by heating the sampling
column in an oven to 170 C and back-flushing the column with pre-
purified helium (liquid-nitrogen trap) for 30 minutes into a 12 cm x 1 mm
stainless steel loop cooled with liquid nitrogen. Repeated desorptions
demonstrated the sufficiency of this time period. The desorption/chroma-
tograph injection valve connections are shown schematically in Figure 2.
Injection of the sample onto the gas chromatographic column is achieved
by firstly directing the carrier gas through the cold stainless steel
loop and then flash-heating the loop to 250 C when the sample is imme-
diately swept onto the head of the gas chromatographic column. When
methane carrier gas is used for chemical ionization mass spectrometry,
the cooled stainless steel loop is initially warmed to -80 C (isopropanol-
solid carbon dioxide) prior to diverting the carrier gas through it.
The gas chromatographic columns routinely used in this program were
an 8 m x 2 mm 3 percent Silar 5CP column, programmed at 2 C min~l
from 60 C to 180 C, and 3 m x 2 mm 3 percent Dexil 300, programmed
at 4 C min~l from 170 C to 350 C, in a Finnigan gas chromatograph. Gas
chromatographic-mass spectrometric analysis is readily achieved since
the chromatograph is interfaced with a Finnigan 1015 quadrupole mass
spectrometer (equipped with a chemical ionization source) through appro-
priate heated stainless steel capillary tubing and sampling valve. The
characteristics of this mass spectrometer do not require any molecular
separator at the interface. The mass spectrometer is controlled by a
System Industries 150 data acquisition system and Digital PDP8 computer
equipped with a very wide variety of programs for manipulation and dis-
play of data.
RESULTS AND DISCUSSION
Preliminary experiments were carried out in order to establish
optimum sampling column desorption times at 170 C, with a helium flow
-------
-35-
LU
_l
Q-
to
O
Q.
LU
cc
o
O
H
O
5
O
° g
LU —
a |«
S 1
"• <
Q CO
2 -
z Q-
9 OC
t 8
2^
I- I
Q
O
H
Z
<
CO
u.
O
Z
g
h-
0 CC
LU UJ
z
<
O co
£ Q-
< <
w S
u- O
^ O
-?
§5
c/2 co
LU <
Q O
CN
O)
^
3
g>
u_
-------
-36-
rate of 500 ml min . Furthermore, by using the columns in series it
was established that the compounds of interest did not break through one
column under normal operating conditions. It was found desirable to
remove water from the sampling columns prior to sample desorption (see
earlier) to avoid plugging the cooled desorption loop with ice; it was
established that the helium purge used for this purpose did not affect
the integrity of the collected sample. By means of duplicate sampling
and chromatographic analyses, it was established that sealed samples
could be refrigerated for at least 10 weeks without apparent change.
Survey Analysis of Vapor Contaminants
Duplicate atmospheric collections of approximately 17,000 liters
were made in Adams City (Denver, Colorado), St. Louis, Missouri, and
Columbus, Ohio. The Denver sample was collected towards the end of a
temperature inversion episode on December 12-13, 1973. The St. Louis
and Columbus samples were both collected towards the end of November,
1973; intermittent rain and overcast skies were encountered at St. Louis
while clear skies persisted during the Columbus sampling period. All
of the samples were subject to gas chromatographic-mass spectrometric
analysis (Silar 5CP column) using chemical ionization with methane (CI),
and the compounds identified are listed in Table 1. A typical recon-
structed gas chromatogram is shown in Figure 3, the major peaks identi-
fied are indicated on the figure. Identification was based upon mass
spectra and relative chromatographic retention times, but specific iso-
mer assignment was not always possible. Many of these compounds appear
to be representative of partially burnt fuel from internal combustion
engines; it is possible that the naphthalene derivatives are formed by
combustion of lower molecular-weight hydrocarbons in the manner proposed
by Badger-^. Tolualdehydes and phthalaldehydes are presumably formed by
partial oxidation of xylenes, either photochemically or in combustion
processes. However, some compounds such as methyl methacrylate, chloro-
benzene, bipiridyl, dimethyl styrene, benzonitrile, and pyridine, would
appear to have originated from a local industry since it is difficult to
envisage other reasons for their presence.
Diurnal Variation Studies
In order to attempt to demonstrate the possible origin of ambient
atmospheric pollutants, a simple diurnal variation study was carried
out in Columbus, Ohio. Seven sampling intervals were selected during a
24-hour period, the Chromosorb 102 sampling columns were subsequently
desorbed and analyzed for fourteen previously detected compounds which
were well resolved by gas chromatography (Silar 5CP column). The
results were quantified by calibration of the chromotograph with stan-
dard solutions of the compounds of interest at concentrations of 200 ng
per |i 1 in methylene chloride or acetone, and are shown in Table 2.
-------
-37-
Table 1. OCCURRENCE OF VOLATILE COMPOUNDS
IN DIFFERENT URBAN ATMOSPHERES
Location*
Denver
Compound (Adams City)
Hexene
Heptene
Toluene
Xylenes
Trimethyl benzenes
Tolualdehydes
Naphthalene
1-Methyl -naphthalene
2 -Methyl -naphthalene
Dime thy 1-cumene
Phthalaldehydes
Phenol
Methyl methacrylate
Allyl phenol
Chlorobenzene
Tetrahydronaphthalene
2 ,3-Dihydro-methyl-benzofuran
Methyl-allyl-phenol
Bipyridyl
Dimethyl- styrene
Tetrahydro-me thy 1 -naphthalene
Benzonitrile
Propiophenone
Dimethyl-naphthalene
Methyl-benzyl alcohol
Hydroxy-biphenyl
Amino-methyl-acetophenone
Tetramethyl benzene
Pyridine
Benzoquinone
Benzophthalaldehydes
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Columbus
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
St. Louis
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X indicates >0.01 ppb detected.
-------
-38-
m
5
CM
^
CC
UJ
oo
UJ
O
UJ
Q
CC
LLJ
>
z
UJ
Q
I-
Z
UJ
CD
X
O
CO
O
O
UJ CO
Z U
O >
K X
w v> v> ° >
UJ UJ UJ 5 X
Z Z Z Q uj -i _i
uj uj uj > O < <
NNNX_IX_J_J
ZZZ<4l-^_iSz
aoninawv
-------
-39-
I-1*
*
L8
LS
LS
LS
o
.5
IS
o
t?
LI?
s
D
S
F8
-------
-40-
ptj
H
<
W
&
S
rH"
O
U
H
S
M
B
,
Hi,
H
H co
-I"1
<<
•
CM
01
rH
43
CO
H
CO
13
H
CO
!
O
.
E
•
cx
o
o
1 — 1
c
•H
CO
rH
K^
1^>
>
CO
d)
42
f\
•
cx
o
0
o^
rH
•r-l
JJ
(2
CO
0)
•H
k>
r-M
CO
M
CO
01
,.L[
o
r\
Pr
M-l
O
LO
o
4-1
LO
<(•
o E
1 1 rrj
LO O
t-H CO
O v£>
r-l
0 E
4J CX
O LO
0 rH
r^ CD
1— 1
o E
4-1 CX
0 0
CM (^
r-.
rH
0
CM
CO
O
vO
CO
0
oo
<)-
o
o
LO
0
r^
^
o
VO
00
0
0)
c
N
fi
o>
43
t-H
>,
CO 42
0) 4J
C 0)
oj B
rH T-l
>, M
X EH
-d- co
0 O
o o
VD CTi
i— 1 rH
O O
1^ CM
r-H CM
0 O
VO i-H
rH CM
0 O
LO O
rH CM
0 O
.
42
4-1
^
r-H
r-H
O
0
O
-d"
o
o
O
CM
o
o
CO
o
o
ro
O
o
vt
o
o
i-H
o
o
0)
13
>>
42
01
13
t-H
CO
3
r-t
O
H
E
CM i-H
0 O
0 0
O CO
r-H O
O O
l-H r~
i-H 0
0 0
r-l ,
13 42
>i 0)
42 T3
O> rH
T3 CO
i-H r-H
CO co
3 42
rH 40
0 42
H PM
1 1
cx o
0
O CO
o o
0 O
1 — 1
O CM
O i-H
o o
CM
O CO
0 rH
0 0
t-H
O CO
O i-H
o o
o
O rH
0 rH
0 0
0
O CT\
o o
0 0
o
o r^
O 0
o o
0)
13
>>
42 0)
CJ C
13 O
rH C
CO 0)
rH 42
CO CX
42 O
4-1 -ri
P4 O
cx pd
CM
0
O
CTi
rH
O
LO
0
o
LO
o
o
0
0
o
CO
o
0
-
CO O
O 0
0 0
LO
CM i-H
o o
O 0
LO LO
•* 0
O 0
a o
o
•d- o
O 0
O 0
0
CO O
O 0
o o
0
,
42
fi
CO
o
•H
r-H i-H
O to
C 42
01 4-J
J2 42
PM PM
-------
-41-
The low ambient concentrations during the final sampling period are
presumably due to the heavy and continuous rain which began during this
time. All of the hydrocarbons exhibit maximum concentrations during
the early daylight hours, and all except 1-methyl naphthalene show a
lesser peak later in the day. The variations in the concentrations of
the alkyl benzenes may be related to the density of automobile traffic,
although if this is the case it is surprising not to see a peak in the
early evening. The concentrations of the naphthalenes are higher than
might be expected, in view of their lower volatility than the other
hydrocarbons; this high concentration may be caused by a large nearby
stationary source, such as fossil-fuel heating unit.
The concentration profiles for the four aldehydes analyzed suggests
that the tolualdehydes and phthalaldehydes are formed by sequential oxi-
dative degradation of the corresponding xylenes. Within the limitation
of the number of measurements made, the tolualdehydes appear to peak
shortly after noon, while the more highly oxidized phthalaldehydes peak
in the late afternoon. This would appear to be consistent with stepwise
photo-oxidation of xylenes. The concentration profile for phthalic
anhydride appears to be parallel to those of the phthalaldehydes, but
the reason for this remains obscure.
The concentration of propiophenone, benzonitrile, and phenol
appear to be independent of both time and the fluctuations of other
compounds measured, which would suggest that such compounds may origi-
nate from relatively constant stationary sources such as manufacturing
industries.
Analysis for Potentially Carcinogenic Polynuclear
Aromatic Compounds
The earlier described survey analyses indicated the abundance of
naphthalenes in the ambient atmospheres studied, and thus a detailed
study of polynuclear aromatic compounds in Denver air was made using
an identical collection technique.
The saturated vapor pressure for commonly occurring POM under
ambient conditions is estimated to be approximately 50 ppb, based upon
the only available data, which has been obtained for anthracene. The
concentrations of specific POM's found in ambient atmospheres are
expected.to be several orders of magnitude lower than this, thus, the
probability of these species being present in the vapor phase is high.
While some POM compounds may be adsorbed on the surface of particulate
matter to varying degrees, even this may be partially or wholly
reentrained as vapor following filtration, impaction, or precipitation
of particulate matter in a sampling train. Thus, it is essential to
ensure that volatile POM vapors may be efficiently collected since the
majority of these species will be collected in this state. Our previous
work in this area has demonstrated that the majority of ambient POM
material may only be effectively collected as vapor.
-------
-42-
Samples for POM analysis were collected at two locations in Denver
(Table 3) and also at a Columbus location in close proximity to an
asphalt roofing operation. While the latter sample is of course atypical
of urban ambient air as a whole, the results clearly demonstrate that
such operations are significant contributors to ambient POM (Table 3).
Quantitative analysis for POM species was carried out by GC-MS
analysis of the desorbed vapor samples. Prior to desorption, appro-
priate quantities of 9-methylanthracene, 9-phenylanthracene, and
9, 10-diphenylanthracene were injected into the samplers during water
vapor removal, to serve as internal standards for identification and
quantification purposes. Identification of the POM species present was
made on the basis of their mass spectra, and retention time relative to
the internal standards. Quantification was made upon the basis of
specific absolute ion current integration, with reference to the quanti-
ties of internal standards present. We have previously demonstrated
that this quantification procedure has a reproducibility of better than
7 percent-*-' > *-°. In an ongoing modification of the computer program,
the entire procedure is being automated, with the aid of data bases of
mass spectra, retention times relative to the internal standards, and
response factors relative to the internal standards-'-". The quantitative
analyses for POM compounds from each location studied are shown in
Table 3, the possible error in these values should not exceed 20 percent.
Reconstructed gas chromatograms are shown in Figures 4A and 4B,
which show reconstructed gas chromatograms for both ambient Denver air
samples. The peaks identified are shown in the respective figures.
The chemical ionization (methane) mass spectra of POM compounds are
simple and characteristic; the base peak is always the protonated molecu-
lar ion (M + 1) and adduct ions are usually seen at (M + 29) and (M + 41),
due to addition of 02^ and C-jH^ species in the mass spectrometer.
Fragmentation is not usually observed.
It is not surprising that by far the highest concentration of POM
species was found in the vicinity of the asphalt roofing operation
(Table 3), but it is also interesting to note that significant quanti-
ties of these species were present in the ambient Denver samples.
The suburban and industrial locations monitored in Denver exhibit
relatively small differences, although the total POM loading is almost
twice as high at the industrial location. The total loading of known
carcinogenic POM (Benz(c)phenanthrene, benz(a)anthracene, methylchry-
senes, benzpyrenes, and methylcholanthrenes (—as rated by the National
Academy of Sciences) is almost four times higher in the industrial area,
which may be partly explained by the presence of an asphalt blowing
operation half a mile upwind of the sampling site. However, the benz-
pyrenes alone are significantly higher at the suburban location. These
limited POM data clearly indicate the importance of vapor phase POM
compounds, since several of these materials are known carcinogens.
Further data at a wide variety of sites are evidently required in order
to establish truly significant comparisons.
-------
-43-
Table 3. ANALYSIS OF AMBIENT POM VAPORS
AT DIFFERENT URBAN LOCATIONS
Concentrations at
Each Location, ppb
Compound
Anthracene/phenanthrene
Methylanthracenes
Fluoranthene
Pyrene
Methyl pyrenes and fluoranthenes
Benz (c)phenanthrene
Chrysene/benz (a)anthracene
Methyl chrysenes and benzanthracenes
Benz (k + b ) fluoranthenes
Benz (j )f luoranthene
Benz (a + e) pyrenes
Perylene
Methylcholanthrene
Suburban
Denver-L
0.73
0.49
0.086
0.09
--
0.012
0.09
0.09
0.09
0.022
0.083
--
0.004
Industrial
Denver 2
0.37
0.008
0.048
0.012
0.11
0.083
0.46
0.48
0.025
0.0046
0.018
0.002
0.002
Columbus 3
64
51
4.5
7.3
c
o
t-han
l_LlCt LI
01 Dt>b
• J. ^J F+S
QQ pVl
CCl \~-Ll
1
4>
1 Aurora.
2 Adams City, adjacent to northern perimeter of oil refinery.
3 Adjacent to asphalt roofing operation.
-------
-44-
3N3UH1NV10H01AH13W -
S3N3HAd(3 P"B "IZN3S —_=- .
3N3HlNVaOmj(l)ZN38
11ZN3B —
/
AVTH3/\0 Nl
S3N3SAHHD 1AH13W
S3N3DVUHlNVZN3aiAH13W
AV1H3AO NOI EfrZ ^
3N3SA«H3 AV1d3AO NOI 6ZZX
3N3UHlNVN3Hd(3)ZN3a
~\
I dNV S3N3dAd TAH13W
3N3HiNvaOn~ld
S3N33VHH1NV1AH13W
3N33VHH1NV
001 06 09 Oi. 09 05 0V 0€ OZ B'.
aoninawv
-------
-45-
801 06 08 OL OS OS W C£ K 01 0
acminawv
-------
-46-
CONCLUSIONS
The results of these studies demonstrate the suitability of the
sampling and analytical methods described to problems associated with
the analysis of polluted atmospheres. While survey analysis gives an
overall picture of an atmosphere, diurnal variation studies may give
insight into the origin of pollutants, as well as being an aid to mecha-
nistic studies. These methods are particularly amenable to the detection
and quantification of trace quantities of polynuclear aromatic compounds
in ambient atmospheres, and highlight the importance of considering the
presence of carcinogenic species which are present in the vapor phase.
ACKNOWLEDGEMENTS
We wish to thank Dr. L. Todd Reynolds of the Colorado Department of
Health for collecting the ambient air samples in Denver in December,
1973, and March, 1975.
REFERENCES
(1) Rasmussen, R. A. A Quantitative Cryogenic Sampler. Am Lab.
19, July 1972.
(2) Bellar, T. A., M. Brown, and J. E. Sigsby. Determination of
Atmospheric Pollutants in the Parts-Per-Billion Range by Gas
Chromatography. Anal Chem. 55:1924, 1964.
(3) Belsky, T., and I. R. Kaplan. Light Hydrocarbon Gases, C , and
Origin of Organic Matter in Carbonaceous Chondrites. Institute of
Geophysics and Planetary Physics. Publication Number 736.
September 1970.
(4) Cramers, C. A., and M. M. Von Kessel. Direct Sample Introduction
System for Capillary Columns. J Gas Chromatog. 6:577, 1968.
(5) Gordon, R. J., H. Maysohn, and R. M. Ingels. ^2~^5 Hydrocarbons
in the Los Angeles Atmosphere. Environ Sci Technol. 2:1117, 1968.
(6) Lonneman, W. A., T. A. Bellar, and A. P. Altshuller. Aromatic
Hydrocarbons in the Atmosphere of the Los Angeles Basin. Environ
Sci Technol. 2:1017, 1968.
(7) Rasmussen, R. A., and M. W. Holdren. Analysis of 05-0^0 Hydro-
carbons in Rural Atmospheres. (Presented at 65th APCA Meeting.
Miami Beach. June 18-22, 1972.) Paper Number 72-19.
(8) Stephens, E. R., and E. R. Burleson. Analysis of the Atmosphere
for Light Hydrocarbons. J Air Pollut Contr Assoc. 19:929, 1969.
-------
-47-
(9) Stephens, E. R., and E. R. Burleson. Distribution of Light Hydro-
carbons in Ambient Air. J Air Pollut Contr Assoc. 17:147, 1967.
(10) Grob, K., and G. Grob. Gas-Liquid Chromatographic-Mass Spectro-
metric Investigation of C -C Organic Compounds in an Urban
Atmosphere. J Chromatog. 62:1, 1971.
(11) Zlatkis, A., H. A. Lichtenstein, and A. Tishbee. Concentration
and Analysis of Trace Volatile Organics in Gases and Biological
Fluids With a New Solid Adsorbent. Chromatog. 6:67, 1973.
(12) Jeltes, R. Sampling of Nonpolar Air Contaminants on Poropak Porous
Polymer Beads. Atmospheric Environ. 3:587, 1969.
(13) Dravnieks, A., B. K. Krotaszynski, J. Burton, A. O'Donnel, and
T. Burgwald. High Speed Collection of Organic Vapors From the
Atmosphere. (Presented at llth Conference on Methods in Air
Pollution and Industrial Hygiene Studies. Berkeley, California.
April 1970.)
(14) Scheutzle, D., A. L. Crittenden, and R. L. Charleston. Application
of Computer Controlled High Resolution Mass Spectrometry to the
Analysis of Air Pollutants. J Air Poll Control Assoc. 23:704, 1973.
(15) Robertson, F. M., and A. C. Neish. Production and Properties of
2,3-Butanediol. Can J Research. 25:491, 1947.
(16) Badger, G. M. Mode of Formation of Carcinogens in Human Environ-
ment. Nat Cancer Inst Monogr. 9:1, 1962.
(17) Particulate Polycyclic Organic Matter. National Academy of
Sciences, Washington, D. C. 1972.
(18) Jones, P. W., R. D. Giammar, P. E. Strup, and T. B. Stanford.
Efficient Collection of Polycyclic Organic Compounds From Com-
Bustion Effluents. (Presented at 68th Annual Meeting of the Air
Pollution Control Association. Boston. 1975.)
APPENDIX
In view of the apparent importance attached to the monitoring of
POM emissions into the atmosphere, we have recently developed a novel
sampling system for high temperature stack emissions of these and other
organic species. The Adsorbent Sampler (Figure 5) is based partly upon
data from the present study, and has been shown to facilitate quantita-
tive collection and analysis of POM species, in contrast to the less
efficient aqueous impinger methods. The Adsorbent Sampler for stack
gases have been shown to be between 2.3 and 30 times more efficient than
-------
-48-
1 ^
J
I , /-
-?. l._ - ., ^
1 1
[i|i!i|i|
•^'M^.-
«•«
k i
1 5l
^
Z
Z
u
LU
cc
O
LU
CO
CO
<
CO
LU
CO
CC
LO
LU
CC
-------
-49-
conventional impinger methods, including EPA Method 5, depending upon
the nature of the combustion process. This development was sponsored
by the Electric Power Research Institute and the U.S. Environmental
Protection Agency."
NOTE ADDED IN PRESS
In the section entitled "Analysis for Potentially Carcinogenic
Polynuclear Aromatic Compounds," it has been brought to our attention
that the Industrial Denver Sample (Table 3) was in fact obtained near
to asphalt production and asphalt blowing operations. This appears
to explain an apparent annomally in Table 3, in that the concentration
of methyl pyrenes, benz (c) phenanthrene, chrysene/benz (a) anthracene,
methyl chrysenes, and methyl benzanthracenes appear to be abnormally
high. We have shown that all of these compounds are relatively abun-
dant in asphalt production in previous studies for the U.S. Environ-
mental Protection Agency.
* Jones, P. W., R. D. Giammar, P. E. Strup, and T. B. Stanford.
Efficient Collection of Polycyclic Organic Compounds From Combustion
Effluents. (Presented at 68th Annual Meeting of the Air Pollution
Control Association. Boston. 1975.)
-------
-51-
LIDAR OBSERVATIONS OF ATMOSPHERIC PARTICULATES NEAR
DENVER, COLORADO
V. E. Derr, G. T. McNice, N. L. Abshire, R. E. Cupp,
R. F. Calfee and M. J. Ackley
Wave Propagation Laboratory
NOAA, Environmental Research Laboratories
Boulder, Colorado 80302
ABSTRACT
Lidar observations were made by WPL-(NOAA) of the cloud of pollu-
tion often observed near the Platte River Valley north of Denver,
Colorado, during the EPA sponsored Denver Air Pollution Field Study of
November 1973. The lidar was located at the Adams County Fair Grounds,
near Henderson, during the period 15-29 November. During this time
four days of data were obtained on smog conditions, namely 16, 21, 26,
29 November. This data is reported, analyzed and compared with other
observations taken simultaneously. From the lidar, estimates may be
made of total aerosol content up to the inversion level, given a
ground calibration.
INTRODUCTION
In the fall of 1973 the Wave Propagation Laboratory (WPL) of the
National Oceanic and Atmospheric Administration (NOAA), U.S. Department
of Commerce, was invited by J. Durham, Environmental Protection Agency,
Research Triangle Park, North Carolina, to participate in an observa-
tion with EPA and other organizations on the location, concentration
and character of pollutants in the atmosphere over and near Denver,
Colorado. Personnel of WPL had just then brought into operation a new
major lidar system and agreed to join the observation at NOAA's
expense.
Essential parts of the lidar system were operational, but the data
recording was not yet complete, complicating the processing and
analysis. Calibration procedures were not then complete, hence only
-------
-52-
relative signals could be measured. In spite of these inflexibilities
the lidar could give valuable information on the presence of aerosols
in the atmosphere above the ground. It could provide profiles of
aerosol backscatter as a function of time for comparison with other
observations. Further, it could, aided by auxiliary data and under con-
ditions found generally to occur, allow determination of aerosol
concentrations as a function of time and altitude.
It is probably not a law of nature that expected atmospheric
events often fail to occur when a considerable effort is made to
observe them, but atmospheric scientists will not be surprised to know
that during the period 15-30 November 1973, when the lidar was on sta-
tion, only certain periods during four days occurred in which the lidar
and most other observation stations were operating and the meteorologi-
cal conditions contributed to significant smog formation. It is the
record of these four days and the correlation of the lidar soundings
with other observations that are the subjects of this report. Data
from other days will be used to establish "clear air" or background
calibration. During most of the other days smog was light or nonexis-
tent, sometimes arising briefly, to be swept away by clean wind from
the west.
OUTLINE OF EXPERIMENT
The persistence of a brown cloud of pollution which has caused a
steady degradation of visibility in the area over and surrounding
Denver, Colorado has become the object of concern of the Atmospheric
Aerosol Research Section, Chemistry and Physics Laboratory, Environment-
al Protection Agency, National Environmental Research Center of Research
Triangle Park, North Carolina. This agency conducted an aerosol
characterization program during November 1973 in the Denver area to
determine the composition and source of this pollution cloud. This
effort involved monitoring by aircraft as well as the use of a number
of ground stations.
The investigation was an attempt to determine the importance of
photo-chemical reactions, automobile exhaust emissions, industrial emis-
sions and wind-blown soil in the formation of the pollution cloud. The
instrumentation used and measurements taken during the experiment are
presented in the accompanying Table 1.
In addition to the Atmospheric Aerosol Research Section of the
Environmental Protection Agency other participating groups included
NOAA, National Center for Atmospheric Research (NCAR), the University
of Denver, the University of Washington, General Motors, Thermo-Systems
Inc., IIT Research Institute and the Meteorology Research Institute.
The geographic area of greatest concern was a segment of the South
Platte River drainage extending from a few miles up-river to a few
miles down-river from the center of Denver. The accompanying map,
-------
-53-
Table 1. PROPOSED MEASUREMENTS AND THEIR APPLICATIONS
Instrumentation
Aerosol Samplers:
Hi-Vol/Andersen
Lundgren Cascade
Andersen Cascade
Total Membrane Filter
Aerosol Monitors:
Type of
Analysis
u
• H
f-,
co
*j
0
u
in
v
(H
O
u-
^^
CS
^1
1
X
4-"
O
n)
^
4-i
<4-l
>H
a
^.^
ct)
i
X
0
• H
4-1
ft
0
in
Xi
o
• H
e
o
**
in
•H
in
rt
e
^
|-p*
u
X
a,
o
u
in
0
fi
u
•H
2
i— (
to
u
•t-t
4-1
o
°
croscopy
•H
S
4->
U
0)
tu
.
c
CS
o
CO
Aerosol
Distributions
I-H
ctj O
U O M
•H CS V
Oj to (H 6
J= CO 3 3
U S CO Z
Source
Inference
,— t
C8
O
• H
e
f
o
o
o
f
i-H
CS
^1
4->
I/)
3
T3
C
h—H
4->
«
3
a
c
f
o
t— i
CO
' f-<
*o D
C X
•H 4-J
2 0
Condensation Nuclei
Counter:
May relate to:
(a) aerosol formation
rate
(b) age of aerosol
cloud
Electrical Mobility
Analyzer
Single Particle Optical
Counter
Integ. Nephelometer
with Rel. Hum. Control
Lidar
Gas Detectors:
Dew Point
Total Hydrocarbon
Methane
Ozone
Oxides of Nitrogen
Carbon Monoxide
Sulfur Dioxide
Weather Monitors:
Wind Speed
Wind Direction
Temperature Gradient
Solar Light Intensity
Ultraviolet Light
Intensity
-------
-54-
Fig. 1, identifies the area and shows the location of the data collec-
tion points used in the study. The 5200 and 5500 ft contour lines
indicate the downward slope of the terrain on both sides of the South
Platte River. The drainage basins formed of Sand Creek in the northwest
direction from Buckley Air National Guard Base past Stapleton Interna-
tional Airport into the Platte River along with the combined drainage
of Clear Creek and Ralston Creek from the west past Golden and Arvada
help influence the local windflow patterns in these portions of the
metropolitan area. The EPA CAMP* station is located nearly in the
center of the area. Pibals were released at the locations shown in
Arvada, at the Yellow Cab Company and at the EPA Trailer site. Other
observation sites were located at Stapleton, Buckley, Jefferson County
(JEFFCO) and Arapahoe County (ARAPCO) Airports, the Federal Center, Rocky
Flats, Welby (WEL), Overland Park (OVE), National Jewish Hospital (NJH),
Children's Asthmatic Research Institute and Hospital (CARIH), the Denver
Trout Farm (TRA) and at Adams County Fair Grounds near Henderson. The
NOAA Lidar equipment was located at the Adams County Fair Grounds, approx-
imately one mile north of Henderson.
Aircraft flights were made over the routes indicated on the map,
Fig. 2 as Green 1, Green 2, and Green 3. These flights were by MRI
carrying the TSI instrumentation. Spiral flights were also flown over
Stanley Lake, Henderson and the EPA Trailer site. Flights were made on
the days of November 14, 15, 16, 17, 20, and 21.
DESCRIPTION OF THE LIDAR SYSTEM
A flexible remote sensing system has been developed in NOAA con-
sisting of multiple wavelength laser transmitters and receivers with
polarization sensitive detectors, a microwave radar, and a radiometer,
mounted paraxially.
The transmitters, receiving telescope and detectors, are mounted
in a trailer requiring a tractor for transport. The microwave radar,
radiometer, sighting telescope, and a closed circuit TV viewer are
mounted on the telescope mount.
The control center and data processing components are located in a
van normally parked parallel to the trailer. The system is transport-
able but not mobile. It can be transported wherever roads are
adequate. See Fig. 3.
Lidar Transmitter Characteristics
During the November 1973 experiment, only the second harmonic of
the ruby laser radiation was emitted (A = 1/2 x 697.3 nm).
*
Continuous Air Monitoring of Pollution
See Appendix for a brief discussion of light scatter from particles.
-------
-55-
Fig. 1. Map of area covered in experiment.
-------
-56-
5 W ,/^x
Fig. 2. Map showing general experiment area and flight paths.
-------
-57-
•H
o
Cfl
(50
C
to
C
t/1
a)
O
O
nS
a!
i
oo
•H
U.
O
Q
O
u
O
10
rt
-------
-58-
Lidar Receiver Characteristics
The following are the most important characteristics of the
receiver systems:
Telescope (Newtonian Mount)
Mirror Diameter 70 cm
Focal Length 203 cm
Telescope Mount - Converted NIKE-AJAX antenna mount.
One mil pointing accuracy, step and fast drives.
Detectors - Photomultipliers for optical and ultraviolet
detection. Interference filters are used to limit receiving
bandwidth and eliminate background radiation from the sky.
Data Processing Capability
The data processing system is shown in Fig. 4. The signal returns
from the two photomultipliers are digitized and stored in two Biomation
8100 digitizers. When the laser is fired each unit digitizes and
stores 2048 8-bit words at sample intervals as short as 10 nanoseconds.
The contents of the digitizers are then transferred to the NOVA compu-
ter for processing and storage on a 9 track digital tape. The computer
has a 24K word storage capacity, model 33 teletype terminal and a fast
paper tape reader. A disc unit and CRT display will be added later.
Along with the data, the system also records a number of bookkeeping
quantities (laser power, time-of-day, telescope position, etc.). A
total of 48 such quantities may be recorded.
Paralleling the computer data path is a standby system which can
transfer raw data and bookkeeping quantities to tape in the event of a
failure in the computer path.
DATA PROCESSING
The pollution study of November 1973 was a first trial of the new
lidar equipment and the incomplete data recording system had several
problems:
(a) Only six binary digits (instead of eight) were available and
occasionally the digitizer "hung up", that is, a straight line
would be recorded until a relatively large change (a few units)
produced a step.
(b) Interference occasionally produced very narrow, large "spikes"
on the record unrelated to the data.
-------
-59-
-------
-60-
(c) The signal-to-noise ratio was adequate only up to 1200 m
because of lack of circuitry to suppress the large signals
from short ranges while raising the gain at longer ranges.
(The problems have all been removed in later experiments.) Because, in
spite of these problems, much useful data are contained in the records,
the data were processed to remove the first two problems. No way exists
to recover the data when the signal is less than one integer of the full
dynamic range of the digitizer (256 units), as is generally the case
beyond 1200 m. Because, in most cases, the height of the temperature
inversion, when it existed, was below 1200 m, there was little interest
in the greater range.
Removing Instrumental Effects
Figure 5 shows examples of the problems mentioned above. Figure 6
shows the data after processing. The spikes of instrumental origin
were always due to a single point having a large value compared with
two adjacent points. They were removed by setting that "out-of-line"
point to the average of the adjacent points.
The steps were caused by the tendency of the digitizer to "hang up",
i.e., to retain a value until the signal changed by several units.
They were removed, replacing them by a sloping straight line.
In addition to removing the instrumental noise, the baseline, due
to instrumental bias, was subtracted, using an average of the last 200
points (the complete record contains 2000 points, sample interval
.01 us). The data were then smoothed by a sliding average extending
over 6 points or 60.0 ns, equivalent to the time constant of the receiver
circuitry. The results of this partial data processing to remove in-
strumental effects are shown in Fig. 6. Complete data processing is
shown, e.g., in Fig. 7.
Another instrumental effect, the cross-over effect, was not
removed. The effect arose because the laser beam and the receiving
telescope aperture are not coaxial; their axes are offset by 54 cm.
Data was used only beyond the range where cross-over was complete
(approximately 75 m). In some records noise of instrumental origin
caused broad peaks near 900 m range. These have been removed, where
identified.
One further instrumental effect that must be accounted for arises
from the fluctuations of laser power output. Solid state lasers
typically vary in power output for each shot by 5 to 10%. The power
emitted on each shot was monitored and each signal (i.e., each record)
divided by the approximate quantity. After this normalization and the
other procedures outlined above to correct for instrumental effects,
the data are recorded on a tape and are ready for scientific
processing.
-------
-61-
. .0
•s i
• -o
(D
XI
«
o
4-J
c
(D
e
0) (D
-o o
3 C
u o
C
O O i
^ X
CQ -P
W>
!H C
01 -H
> 3
C O
o j=
O c/>
03
•O
CD
O
O
^
DH
M
•H
U-
-------
-62-
-------
-63-
Scatter Cross Section as a Function of Range
The lidar signal from an extended target (say aerosols) is given
by:
S = P x (1/2) CT x 3" x -^-y x £(R) x t x exp (-2aR),
where: S = signal power at detector
P = transmitted power
c = velocity of light
T = lidar pulse width or receiver time constant,
whichever is larger
R = range of aerosol scatterer (m)
-• 23
3 = volume backscatter coefficient (m /m )
2
A = area of receiving telescope (m )
f(R)= cross -over function
a = total extinction coefficient (m~ )
tn = receiver efficiency
K
The quantity 3^ is the volume backscatter coefficient. We may
neglect the exponential factor; at the small ranges and high visibility
levels encountered in this experiment, the factor is nearly one. In
order to determine 3^ it is necessary to know the other factors. The
factors c, T, P , A, and R are determinable with satisfactory accuracy.
The quantity S requires a calibration of detector sensitivity and
electronic gain. These are made uncertain by temperature, aging and
equipment drift and modification. The function f(R), due to the beam
cross-over geometry, is difficult to evaluate because of variation of
temporal and spatial power density across the laser beam. Further, tR,
the receiver efficiency, is difficult to measure satsifactorily and may
vary with time as optical surfaces age and become dirty.
There are a number of ways lidar signals may be calibrated, e.g.,
(a) absolute calibration against a target whose scatter
characteristics are known.
(b) evaluate and stabilize the characteristics of the optical
and electronic systems.
-------
-64-
(c) use the Rayleigh scatter from the "clear air" for a target
as in (a) above.
(d) determine the aerosol concentration at short range by a
ground-based instrument, and estimate aerosol concentrations
at greater ranges from the relative intensity.
(e) if a clear day occurred, assume it was Rayleigh scattering,
neglecting a small error, and for each value of the range
determine the ratio of all other signals to that chosen
signal.
We here choose a combination of the last two methods because of
their simplicity and because the quality of the data does not warrant
more complicated procedures. Specifically, the signal from clearest
days as shown by ground based aerosol characterization facilities was
chosen to represent Rayleigh scattering. (An estimate, based on particle
counts, showed less than 15% error arising from this assumption at
ground level.) Ratios were formed, for each range, of other signals
to the chosen standard. Methods (d) and (e) are useful in estimating
the number concentration of particles only if the size distribution and
chemical constitution do not change radically over the range of
interest. When data from aircraft flights were available, the dis-
tributions were found to be approximately constant with height, as
discussed in the next section.
A possible flaw in this method is that data for choosing the
standard "clear" day was available only from ground based nephelometer
units. Thus, even if the air were properly chosen as "clear" at ground
level, it may not have been "clear" at higher levels. This source of
error was minimized by choosing records whose ratio to a calculated
Rayleigh scatter signal changed minimally with height.
Averages of the backscatter coefficient over space and time are
described in the following section.
OBSERVATIONS
Observations were made by lidar, by ground based instruments and
by aircraft. We present here the results of lidar observations and
summarize the available data from ground units located near the lidar
and aircraft runs in the vicinity. The section concludes with a dis-
cussion of the lidar measurements compared with other data.
Lidar Observations
Many of the lidar signals show significant aerosol content above
temperature inversions. Superficially, the inversion is considered to
"put a lid" on the lower atmosphere, preventing vertical mixing of
pollution. That picture is, of course, often accurate, but in the
-------
-65-
Denver area, during these observations, the^atmosphere immediately above
the inversion often had aerosols with as much backscatter intensity as
below, after correction for 1/R^. There has been some evidence of mix-
ing through inversions, but in this case it is suspected that the aero-
sols were advected from distant sources, emitted at earlier times, and
were either in place before the formation of the temperature inversion,
or it is possible that a wind above the inversion level may have brought
the aerosols over the lidar area at night, when the lower level winds
are calm, from other parts of the city. Because of the limited range of
the lidar as employed at that time, the upper level of the aerosols was
not always observed. In some models of aerosol concentration and diffu-
sion, the temperature inversion level is chosen as the top of the volume
in which aerosols are mixed. Clearly, this could lead to inaccurate es-
timates of total pollutants in the atmosphere. Those remote sensing de-
vices which detect the temperature inversion levels (e.g., acoustic
sounder, FM/CW radar) thus do not always provide accurate information on
the height of particulates. The lidar is backscattered principally by
particulate matter.
The relative backscatter radiance of the aerosols is presented as
a function of height in Figs. 7 to 10. Each graph results from a single
lidar shot. Each is marked with the approximate time of the shot (MST)
and the date. Shots with the same time indication were taken at differ-
ent but closely spaced actual times and are among those averaged for a
given entry under that time in Tables 2 to 5. The horizontal axis is
the height (m) and extends from 75 to 1115 m. The backscatter radiance
on the vertical axis has been multiplied by a convenient factor to accom-
modate it to the available space; the factor is constant for each day.
When multiplied by the appropriate factor the graphs are approximately
the backscatter radiance for each day; because of the lack of absolute
standards and the possibility of calibration drifts, the backscatter ra-
diance during the day may be considered consistent, but the values are
only approximately related between days.
The height was divided into four sections of 260 m and the average
backscatter radiance obtained for each. These averages and the average
over the total height are presented in Tables 2 to 5. No scale factors
are presented in this data and, within the limitations mentioned above,
the numbers in the tables represent the average relative backscatter
radiance.
By utilizing aerosol measurement data from nearby observations we
may approximately calibrate the lidar signals in terms of the number
concentration if the size distribution and index of refraction distribu-
tion remain approximately constant. An examination of aircraft data
shows that the relative size distribution varies little as a function
of altitude during the course of any one day. (The number concentration
varies considerably as a function of altitude.) More precisely, based
on data from this and other observations, if the meteorological condi-
tions are calm and remain approximately the same, during the course of
a day or a few days, the relative size distribution does not change much,
*
See Appendix
-------
-66-
F4RS 07:17:14
Nov. 16, 1973 Figure 7 A
F6R4 07:25:51
F8R1 07:35:50
F8R3 07:35:50
F9R1 07:40:41
F10R3 07:46:50
F11R4 07:50:47
F12R4 07:57:44
F13R1 08:00:53
FISRS 08:00:53
RANGE (m)
(75 to 1115 m, each graph)
-------
-67-
F14R1 08:23:24
Nov. 16, 1973 Figure 7B
F14R3 08:23:24
F16R1 08:51:28
F17R1 09:12:06
F16R3 08:51:28
F17R4 09:12:06
F16R4 08:51:28
F18R1 09:42:28
F18R3 09:42:28
F19R2 10:00:53
F19RS 10:00:53
RANGE im)
(75 to 1115 m, each graph)
-------
-68-
Nov. 21, 1973 Figure 8
F2R2 07:30:55
F3R4 07:36:13
F4RS 07:40:46
F10R2 08:09:39
F1JR1 08:15:53
F14R5 08:31:33
F2R2 09:05:36
F7R5 09:56:00
F12R4 10:25:45
F16R3 11:00:39
F4R3 12:05:55
RANGE (m)
(75 to 1115 m, each graph)
-------
F1R1 09:44:04
F3R2 09:55:04
F5IU 10:05:00
F7RS 10:15:00
-69-
Nov. 26, 1973 Figure 9A
H2R1 09:50:57
F4R2 10-00:00
F6R1 10:10:30
F8R3 10:21:00
FMU 09:55:04
F4R3 10:00.00
F6R4 10:10:30
F9R5 10:25:00
RANGE (tn)
(75 to 1115 m. each graph)
-------
-70-
HOK4 10:30:00
Nov. 26, 1973 Figure 9B
F11R2 10:55:00
F12R1 10:40:00
F12R3 10:40:00
F13R2 10:45:00
F14R1 10:50:00
F15R3 11:01:30
F16R1 11:06:00
F17R1 11:10:00
F17R5 11:10:00
F18R1 11:21:20
F18R3 11:21:20
RANGE imi
(75 to 1115 m, each graph)
-------
-71-
F1R3 07:13:10
Nov. 29, 1973 Figure 10
F2R3 07:25:30
F4R3 07:36:00
F9R3 08:00:00
F12R5 08:15:00
F17R1 08:40:00
F1R1 09:00:00
F11R2 10:10:00
F13RS 10:20:00
F14R1 10:25:00
F17R3 10:40:00
F1R1 11:00:00
RANGE (m)
(75 to 1115 m, each graph)
-------
-72-
Table 2.
AVERAGE RELATIVE BACKSCATTER RADIANCE
(Nov. 16, 1973)
HEIGHT INTERVALS (M)
7b- 33b- b^b- dbb- 7b-
UAY TlMt 33b 5>»b H-jb Hlb 1115
320 07:01!14 14.00 13. <*b 14.lb Ib.Ul i**.lb
3<;0 U7:0b:4ti 12.97 11.9d 12.5b 14.b9 13.Ob
U7:iO:bJ 19.33 17.14 Ib.bti 17.2b 17.bO
O7:i7:in 14.77 13.3 f Lt.dti 14.bB l^+.db
07:5 U.43 13.^0 13.i/ 13.91
3<:0 07:^b:bu in.30 Id.45 12.97 12.92 13.^1
07:bO:4/ lb.^7 14.12 13.?9 14.U9 14.?b
07:b7:4<4 19.bb lb.9b lb.ll 17.76 17.42
Utt:o05bJ ^.92 7.bb b.b7 t.b1 b.94
Od:23:24 ic.bb o.9^ 9.bH 1-^.3d 10.d/
0«:3i:iu ^O.d7 tu.t'f 22."b 3J.ti^ 24.37
OB:bi:2d il.09 d.VD o.ti7 3,b7 t).7b
09:i2:0fc H.32 9,dl 9.31 3.3D b.bO
09:42:dc 10.3u S*.lb S.02 ^.33 b.-^b
10:oo:bJ 17.17 14./d Ib.JU 13.40 Ib.bfc
-------
-73-
Table 3.
AVERAGE RELATIVE BACKSCATTER RADIANCE
(Nov. 21, 1973)
HEIGHT INTERVALS (M)
Jcb
J<:b
3cb
Jcb
1 1Mb.
07:^b:Hi
0 7 : J 0 : b 2
07:3b: 1 J
07 :HU :HC
Ob:oo: lc
06509:3^
0«: lb:bJ
ucj:20:bl
Ob:2b: j
.01
. 'Jb
.03
. JV
.*4
.it
.23
.dd
.*b
.3H
.tl
.3b
,3b
. iy
.*£
• CL-3
.19
.v7
.Ob
. 7b
JJ3- 3S»b- b3b-
b
b
7
7
0
l)
U
u
0
0
i
u
u
1
0
0
1
1
1
1
1
1
i
1
i
bS3
.££
.s>*
.«H
. Jb
.do
.^0
.y/
.H3
.bi
.^^
.12
. 74
.So
. lo
.S7
. (I
.lH
.27
. /H
. JH
. 13
. ^o
. 10
• £H
,-so
.On
.bo
o
7
7
7
0
u
0
0
0
0
1
0
0
1
u
0
i
1
1
1
1
1
1
1
1
abb
.37
• O1?
• ye
.30
.ti3
. He1
.93
.7d
.be
• td
. l /
. DO
.0 7
.27
• y i
• 7y
. 1H
.^7
.ob
• H£
• 11
..53
.2^
. JH
• ya
. Ob
.o9
1
b
7
0
7
0
0
0
0
0
u
i
0
0
1
0
0
0
0
2
1
0
1
1
1
i
1
113
.JO
.2d
. 3n
.90
.bl
.70
.bb
.bn
. 3o
.JH
.01
.iy
.6b
.HU
.7d
. J^
.bJ
.yy
.10
.HH
.bb
.32
.19
.37
. Ob
.On
.73
1
b
/
b
7
0
0
0
0
0
u
1
0
u
i
0
0
1
1
1
1
i
1
1
i
1
73-
llb
.bU
.32
.23
.74
.03
,0V
.92
. /y
.7j
.71
. 17
. OH
• VI
.£ /
.yd
.72
.11
.2n
.01
. jy
.07
. JH
.^1
.29
.yy
.Ob
.71
-------
-74-
Table 4.
AVERAGE RELATIVE BACKSCATTER RADIANCE
(Nov. 26, 1973)
HEIGHT INTERVALS (M)
DAY
330
330
330
330
330
330
330
330
330
330
J30
330
330
330
330
330
J30
330
TIME
09 '.44:0^
09:50:b/
09:55:0*
10:00:00
10:ob:ou
10:iO:3u
10:15:00
lo*'2i:oo
I0:2b:oo
101: 30:00
10:3b:ou
10:40:00
10:45:00
10:50:00
li:oi:3o
11:06:00
11 :io:oo
li:21 :do
73-
335
9.<+fc
9.^6
10.72
10.82
B.*3
11 .02
10. J7
11.85
10.^3
11 .oO
8. Vo
5.e9
6.0B
b.09
b.tl
O.Hd
b.5
335-
595
7. by
7.35
7.73
7.41
7.96
7. fed
8.65
b.tl
b.77
7.5^
^.70
3.^0
4.eo
4.00
<+.03
4.41
^.16
3.91
595-
655
7.53
7.4fc
7.59
7.0fe
7.dd
b.ti9
b.91
a. bd
cj.Bb
7.U7
/.db
3.72
5.11
4.18
-------
-75-
Table 5.
AVERAGE RELATIVE BACKSCATTER RADIANCE
(Nov. 29, 1973)
HEIGHT INTERVALS (M)
DAY
333
333
333
333
J33
333
333
333
3J3
333
33J
333
333
333
333
333
333
333
333
333
333
333
J33
333
333
333
333
3J3
J33
333
333
J33
riMt
0 7 : 1 3 : 1 U
07:25:30
07:30:4b
07:36:0u
O7:40:ou
O7:4b:oi>
07:50:oo
07! 55: 00
OB:OO:OU
08:o5:oo
OB: 10:00
08 •* Ib.'Ou
08:20:ou
OB:2b:Oo
Ob:30:0o
08:35:20
08:40:uu
08:<+5:30
085bO!UO
08:b5:oo
09: 00: Ou
09: ib:ou
09:30:uo
oy:3b:ou
09:40:00
09:4b:ob
09:50:0<:
09:5b:ou
lo:oo:oo
io:05:ou
10:iO:OU
I0:i5:ou
7b-
33b
0.97
l.il
0.90
1 .02
O.b4
1 ,U9
l.i/
O.,al
0.91
O.o9
1. Ob
Lib
O.^'f
O.b?
0.17
I.e./
0. '9
l.HB
1 .04
U .I'd
^.b5
O.^b
l.iv
i .Ob
1.10
\.tit
1.13
1 .09
U.^U
l.u^
l.^
O.d9
33b-
b9b
0 .o9
I.Ob
O.d3
0.93
0. 75
i .01
1 .Ob
0. /3
O.ci
-------
-76-
Table 5. (cont.)
AVERAGE RELATIVE BACKSCATTER RADIANCE
(Nov. 29, 1973)
HEIGHT INTERVALS (M)
7b- 335- bvt>- eibb- YD-
UAY TlMt 335 Sso obb lllb 1115
333
333
333
333
3J3
333
J33
333
333
333
J33
J33
333
333
3J3
10:20:00
1 0 : 2b : Ou
10:30:00
lo:3b:o«J
1 U :^0 : oo
I0:4b: u*j
10:50:00
10 :55: oo
11 : oo: ou
li:ob:uo
1 1 : 1 0 : Ou
11 : 15: Ow
11:20:00
1 1 : 25 : OH
11 : jo:ot
i
1
1
1
I
1
1
1
1
I
1
1
1
I
1
. 10
.32
• 3£
.Jl
• £H
•£4
• £ib
»0ci
.13
.2e
.12
.26
.07
• 2b
. Ifc
1
1
1
i
1
i
1
1
1
1
1
1
o
1
1
.Ob
.J j
.33
.^0
.23
.^b
.^3
.06
. 14
,C\
.4^
.23
. yb
.22
.10
1
1
1
1
i
1
1
1
1
1
2
1
0
i
1
. 10
.^»b
.H2
. JC
.37
.3^
. JH
. 1H
.17
• 2e
• lb
.22
.33
.27
.13
1
i
1
1
1
1
1
i
1
1
3
1
i
1
1
•
*
«
•
•
*
•
•
•
*
•
•
•
•
•
02
bb
Hb
J2
3o
31
HI
ib
11
27
34
21
16
31
IH
1.07
1.41
i « 3o
l.Ji
1.30
1 .CO
1.31
1.12
1.14
1 .£6
2.01
1.23
U.S'j
1.27
1.14
-------
-77-
in a given location, as a function of altitude, to the top of the mixed
layer. Under these circumstances a calibration of the lidar backscatter
signal is permissible by reference each day to ground based particle
measuring equipment. The total loading in micrograms per meter at the
ground should be applied to the backscatter radiance in the 75-335 m
height interval. Correction for the volume of the lidar beam may be
deduced from its initial diameter (5.08 cm) and its divergence (2 mr).
The total particulate loading of the atmosphere up to 1115 m may be
calculated from this data. See Appendix for the method. Calculations
are not included here because of the unavailabilability of ground based
data to the author.
Ground Based Observations
Although the lidar van was located at the Adams County Fair Grounds,
the nearest location where meteorological data of interest was taken on
days when the lidar was operating was at the Denver Trout Farm (DENTFARM)
on the 16th and 21st of November 1973. The significant measurements
were the wind data (speed and direction) and backscatter data measured
by an integrating nephelometer. Plots of these data as a function of
time for these two days are shown in Figs. 11 and 12. Along the left
ordinate is the scale for the wind speed in km-hr~l. The scale for
scattering in units of 10~4m~l is along the right hand ordinate. The
measurements on the 16th cover the entire 24-hour period. On the 21st
the wind became strong at about 1600 hrs blowing the pollutant cloud
away so measurements were discontinued. The data were given as
averages over one hour intervals so horizontal lines were drawn on the
graph covering the one hour interval. Light solid lines connect the
wind data points; dashed lines connect the backscatter points. The
wind direction is written in at each point. The total pollutant load
for November 16 (0000-2400 hr) was 41.3 yg/m3; for November 21 (0000-
1600) the load was 32 yg/m3 as measured by an 8-stage Anderson sampler.
Aircraft Observations
In addition to the ground based measurements, observations by
an instrumented aircraft over designated flight patterns were made.
The one of particular interest to the lidar observations is the one
referred to as the Henderson spiral. The aircraft circled over the
Henderson site taking data as it spiraled, thus getting an altitude
profile for the various measurements. The profiles of particular
interest for analysis of the lidar measurements were the backscatter
coefficients obtained by an integrating nephelometer (lO'^m'l), the
temperature (C) and condensation nuclei concentration, CN, (lO^cm-l).
The available data for 16 November and 21 November are presented in
Figs. 13, 14, and 15 at the times shown. Appropriate scales for the
three quantities are indicated by bsca^. for the scattering coefficient,
T for temperature, and X for condensation nuclei. The altitude in
hundreds of feet (MSL) is plotted along the ordinate starting at 5000 ft
(ground level is 5151 ft).
-------
-78-
C
; X
*
-3.0
-1.0
lACKKATTEIt
VMNDSKfD AND DlRICTION
OO O20O O400 0«OO
IOO IOOO IJOO 1400 14OO
TIME (HK5. MST)
Fie. 11. WIND AND SCATTEKINO DATA AT DtNTtAIM 16 NOVtMH* 1»7»
WOO 2MO MOO
-------
-79-
.6.0
0200 0400 0600 OMM 1000 1200 1400 1600 1100 2000
2.0
00
2200 2400
FIO. 12. WIND AND SCMTMINO DATA AT MNTFAKM 11 NOVIMUR 1773
-------
-80-
I-
P.
I
Cisw 'isad jo
-------
-81-
f « X
to
X
to
'e
4J 4-1
•i-l «
*J U
C «
C4
3 -0
^
0)
P.
>
o
z
-
•
a
4->
CO
•O
(TSIl 'J33d 10 spaapunH)
-------
-82-
R"
M
B
eg
*
fe
o-
I
L
(1SU '
spajpunH)
-------
-83-
At the time of the morning flight on 21 November there was a. very
distinct layer of pollution between 5500 and 6000 ft and a less pro-
nounced one starting at about 6500 ft (Fig- 14). These layers are just
barely distinguishable in the data from the noon time flight (Fig. 15).
Discussion of Lidar Observations
Four days of data are discussed below. During those days the lidar
was active. On 16, 21, November, data from the aircraft and DENTFARM
were available. On 26, 29, November only lidar data was available at
the time of preparation of this paper. The agreement between the lidar,
the aircraft and DENTFARM observations is qualitatively good. No more
can be expected since the air mass sampled by the three devices was not
the same.
November 16, 1975--The lidar signals (Figs. 7A, B) show large,
fluctuating signals from aerosols beginning at 0717, continuing to 1000.
A denser layer occurs often near 275 m, corresponding approximately to
a maximum in backscatter (bscat), condensation nuclei (CN) and tempera-
ture (T) observed by the MRI aircraft (Fig. 13), at a somewhat later
hour (% 1300 MST). Data from DENTFARM (Fig. 11) show a maximum in
backscatter there at 1030. The maximum occured in the lidar data
(Fig. 7B) between 0823 and 1000, with large fluctuations. Rising wind
at DENTFARM (Fig. 11) at 1200 was accompanied by decreasing particulate
backscatter. The layers observed by the aircraft in the Henderson
spiral (Fig. 13) appeared to persist in spite of the increased surface
wind. No records of the wind aloft are available for this period and
place. The lidar was not in operation after 1000.
November 21, 1973--In the early morning (0700-0800) Strong aerosol
signals were observed, with denser layers at 75 to 100 m (Fig. 8). These
occurred during a period of low wind velocity at DENTFARM (Fig. 12).
Rising wind at DENTFARM, after 1000, was there accompanied by increasing
backscatter, but at the lidar site cleaner air was observed. The
aircraft data (Fig. 14) for the period ending 0915 shows large back-
scatter and CN from ground to 300 m; the lidar shows large backscatter
generally below 250 m. The flight ending at 1234 (Fig. 15) shows much
cleaner air, in agreement also with the lidar. This general agreement
between the aircraft and lidar data cannot be extended to precise
quantitative measures because of the different air masses and times of
observation and because the lidar was not absolutely calibrated during
this period. However, the aircraft data showed that size distribution
did not vary greatly with altitude and thus, by the method outlined in
the Appendix the lidar data can be used to estimate total loading if a
single point calibration can be obtained from ground or air data.
Presumably this data is available in other sections of this report.
November 26, 1973--The lidar signals (Figs. 9A, B) showed strong,
variable, aerosol backscatter from 0944 to 1121. There was some evidence
of an inversion near 300 m. Data from other observations was not avail-
able at the time of preparation of this paper.
-------
-84-
November 29, 1975--Only weak and variable aerosol backscatter
observed (Fig. 10) with no noteworthy features.
SUMMARY
The results of lidar observations of particulate air pollution
at Adams County Fair Grounds have been reported. Other data, aircraft
and ground, have been compared with the lidar data where available.
Only on one day, 21 November, have the data been sufficiently concurrent
and sufficiently contiguous to fully compare the aircraft and lidar
data. Here the agreement is qualitatively good.
A method has been presented in the Appendix for estimating total
particulate loading of the atmosphere up to the top range of the lidar
(1115 m) under circumstances where the size-distribution and chemical
composition were not greatly varying with height.
Especially valuable, but not analyzed in the text, is the insight,
gained by lidar observations into the fluctuations of particulates.
Great variation (compared with a mean) can be seen in Figs. 7-10.
An analysis of such variations is required to arrive at proper spatial
and temporal averaging procedures.
Additional data expected in other papers of this report may be
further compared with the lidar data.
ACKNOWLEDGEMENTS
The authors thank J. Durham (EPA) for making funds available for
data processing costs.
-------
-85-
APPENDIX
The presence of particulates in the atmosphere is detected by their
properties when illuminated by light. Small particles scatter light in
all directions (Mie scatter) in complicated patterns depending on the
size, shape and composition of the particles and the wavelength of the
light. Molecules of the atmosphere also scatter light (Rayleigh scatter),
but effect is usually small compared with that due to particles in a
polluted atmosphere. A portion of the light scattered directly back
from the particles toward the source (laser) may be gathered by a tele-
scope and detected. The power of the light detected is an approximate
measure of the amount of particulate matter present. It is not exact,
since the amount does not depend in a simple way on the number of
particles nor on the mass of the particles. However, if the size dis-
tribution and chemical character remain approximately constant, the
lidar signal is proportional to the number of particles present. Thus,
if the mass density is known at the ground, its variation with height
is proportional to the lidar signal, and the total amount up to the
height recorded may be found by integrating under the lidar signal
curve. Tables 2-5 give numerical values of the lidar signal (or radiance)
from estimates which may be made quickly of the relative amounts of
particulates in the beam. Adjustment must be made for the changing
volume of the beam with range due to the 2 mr. The lidar signal received
is proportional to the radiance of the particles in the source beam,
which is proportional, under the circumstances cited above to the number
of particles. The backscattered radiance is the radiant energy emitted
toward the laser, by the particles, per unit time, per unit projected
area of surface, per unit solid angle. Because, at the time of the
experiments, no absolute calibration was available, the signal is
presented in relative backscatter radiance units.
A laser is used as a source (instead of a searchlight) because its
narrow beam width, monochromaticity and high peak pulse power permit
detection of signals from great ranges even in daylight when the back-
ground skylight is a source of interfering noise. Because of those
characteristics of the laser, the receiving telescope may be of limited
field of view (excluding much skylight) and the detector may be protected
by a narrow band optical filter centered at the laser wavelength. Pulse
operation permits ranging just as in radar, and the high peak power
permits high signal-to-noise ratios on a single pulse, simplifying signal
processing. (The acronym LIDAR arose analogously to RADAR: Light
Detection and Ranging.) Particles are regularly detected in the strato-
sphere at heights of 80 Km. However, for low lying pollution observation
only the lower atmosphere is normally examined.
-------
-87-
NUCLEATION CHARACTERISTICS OF DENVER AEROSOLS
Charles C. Van Valin and Rudolf F. Pueschel
Atmospheric Physics and Chemistry Laboratory
Environmental Research Laboratories
National Oceanic and Atmospheric Administration
Boulder, Colorado 80302
ABSTRACT
The aerosol physical parameters of Aitken, cloud, and ice nuclei
concentrations, and of atmospheric light scattering are reported here
for 16 days of measurements at two sites within the South Platte River
Valley northeast of Denver, Colorado. These two sites are within the
shallow basin that defines the urban area airflow patterns and repre-
sent locations fairly close to, and more removed from, the urban
emission sources.
Measurements show that the atmospheric pollution at the more remote
site is generally more dilute than at the closer site, but they consis-
tently demonstrate that automobile traffic is the primary source of
Aitken nuclei. Also consistently demonstrated is the conversion of
Aitken nuclei to cloud nuclei and light scattering particles. This
process proceeds slowly in the dark and is markedly accelerated in the
sunlight. The striking increase of atmospheric ice nuclei from high
winds is shown on two notable occasions. The measurements of ice nuclei
from traffic or other urban sources are interpreted to show small in-
creases in ice nuclei concentrations at both sites.
INTRODUCTION
The influence of cities, and particularly of automobile traffic,
on atmospheric processes has been the subject of considerable research
(e.g., Schaefer1, Barrett et al.2, Hidy et al.3, Weickmann4), and a
definite relationship between man-made cloud and ice nuclei and precipi-
tation effects has been demonstrated (e.g. Hobbs et al.5, Weickmann"*) .
-------
-88-
In a long-range program of the Nucleation Chemistry Group, Atmospheric
Physics and Chemistry Laboratory, ERL-NOAA, we had earlier conducted
studies on the significance of the Denver urban aerosol (Pueschel et
al.6); participation in the EPA-coordinated "Brown Cloud" experiment in
November 1973 presented the opportunity to pursue this important subject
fur ther.
Aerosol parameters measured during the study period of 6-23 Novem-
ber 1973 were Aitken nuclei (AN), cloud condensation nuclei (CCN), and
ice nuclei (IN) concentrations, and the atmospheric light scattering
coefficient (bscat). Chemical composition and surface characteristics
of the Denver aerosol collected on 14 and 21 November 1973 have been re-
ported (Van Valin et al.7), and these properties are not discussed here.
AN, CCN, IN, and bscat for those two days were also examined in detail;
discussion of these parameters is included in the context of the examin-
ation of the entire study period.
By strict definition, AN are those particles that serve as conden-
sation nuclei for the formation of droplets at the > 300 percent super-
saturation achieved during the rapid adiabatic expansion produced in an
Aitken nucleus counter; in common terminology, AN are all particles
between molecular size and 0.2 ym diameter. In terms of number concen-
tration in the atmosphere, AN are predominant. Most AN are smaller
than the size required to produce appreciable light scattering. CCN
are those particles that are suitable sites for the formation of water
drops from condensation of water vapor in the natural atmosphere; they
are necessarily wettable or soluble in water. Particles must be as
large as or larger than a certain minimum size to be active as CCN at
a given supersaturation, e.g., the minimum radius for wettable, but
insoluble particles at 0.1 percent supersaturation is ca. 1.6 ym, and
at 1.0 percent it is ca. 0.13 ym. The minimum radius for soluble parti-
cles is smaller by an order of magnitude or more. IN are those particles
that serve as nuclei for the formation of ice crystals. IN are insol-
uble in water; their ice nucleating capability apparently depends upon
the presence of surface sites that resemble the geometry of the ice
crystal. Light scattering is accomplished by particles larger than
0.2 ym diameter. In the real atmosphere, essentially all light scatter-
ing is done by particles 0.2 < D< 2 ym because there are so few larger
particles present.
EQUIPMENT AND LOCATION
AN concentrations were measured with the Environment-1 counter,
b with the Meteorology Research, Inc. nephelometer, CCN concentra-
tions with the MRI thermal diffusion cloud chamber operating at 0.5%
supersaturation, and IN concentrations with the N£AR counter. These
instruments are mounted in a medium-size travel trailer, with the air
sampling intakes at a height of ^4 m above the ground.
-------
-89-
From 5 November to 19 November 1973, our mobile laboratory was
located at the Adams County Fairgrounds, 16 km northeast of the Denver
city limits, and about 1.5 km west of the South Platte River. From
19 November to 23 November 1973 it was at the State of Colorado Humane
Society about 6 km northeast of the Denver city limits and approximately
1 km east of the South Platte River. Both sites were within the river
valley; the Fairgrounds and Humane Society sites were approximately
0.5 km from the west and east edges of the valley, respectively. Accord-
ing to Riehl the Crow8, the Denver urban plume is commonly carried
northeast along the South Platte River valley during the night and early
morning hours when light downslope winds occur along the eastern slope
of the Front Range of the Rocky Mountains. During the daytime, when
sufficient surface heating takes place, the direction of air flow
reverses. Thus, one can imagine an oscillating flow of air within
the valley, with urban pollutants being added at the southwest end of
this system and removed and diluted toward the northeast.
OBSERVATIONS AND DISCUSSION
For the purpose of this report, we consider observations for
the periods 6-17 and 20-23 November 1973. Figures 1 through 4 show
AN, CCN, IN, b wind speed, and wind direction. To include enough
time so that patterns are apparent, each figure covers four days. Wind
speed and direction were taken from measurements of the EPA Mobile Field
Laboratory at the Humane Society site, from the Welby monitoring site of
the EPA Field Monitoring Program (Crow9), and from the National Weather
Service-NOAA, Stapleton International Airport, in that order of prefer-
ence. In addition, comments that we made in our project notebook have
proved invaluable in the interpretation of our data in the light of local,
on-site air movement, and in refreshing our memories.
Our notebook remarks for 6 November 1973 indicate a mostly cloudy
sky, wind from the northeast at about 1.5 m sec"1, and a distinctly visi-
ble pollution cloud to the northeast at 0730 MST. This weather situation
continued, with alternately light northerly winds and calm air and visi-
ble pollution, until mid-afternoon when cloud cover decreased and visibi-
lity improved. The visual observation of an aged pollution cloud is
corroborated by the AN, CCN, and bscat measurements. AN continued to be
supplied in rather high concentrations throughout the day from traffic
on nearby roads; the greatest CCN concentration and bgcat occurred during
mid-afternoon, when insolation had increased. In late afternoon and
early evening the CCN concentration and bscat declined markedly.
On 7 November 1973 (Figure 1) the large increase of AN concentra-
tion coincided with the daily increase in traffic, although CCN and bscat
did not increase, probably because of cloudiness. The sky cleared about
noon, but the accompanying increase in wind velocity was sufficient to
dilute any possible atmospheric pollutants. A power outage interrupted
our instrument operation from 1230 to 1430.
-------
-90-
12 IB 00 06 12 18 001 06 12 B 00 06 12 16 00
'£
12 18 00 06 12 18 00 06 12 18 00 06 12 18 00
6 Nov 73 Tuesday 7 Nov 73 Wednesday 8 Nov 73 Thursday 9 Nov 73 Friday
Figure 1 AN, CCN, and IN concentrations and bscat as measured by
the mobile laboratory at the Adams County Fairground
site for 6-9 November 1973. Wind speed and direction
were obtained from measurements of the EPA Mobile Field
Laboratory, from the Welby Monitoring site of the EPA
Field Monitoring Program, and from the National Weather
Service.
-------
-91-
Fog covered the site for the entire day of 8 November 1973 (Figure
1). The AN concentration did show a slight increase during the daylight
hours, but, obviously, most particles were washed out by the fog. The
apparent slight increases in CCN and bscat are probably the result of
instrument response to the fog, rather than to a solid aerosol.
Fog was also present on 9 November 1973 (Figure 1) until about noon,
when the sky cleared. It is interesting to note that the AN concentra-
tion increased slightly as the fog (and relative humidity) decreased.
The minor increase in bscat at about 1100 is not readily explained, but
it is obvious that the increasing numbers of AN were not appreciably con-
verted to CCN and scattering particles, especially after sundown.
The variation of IN concentration between limits of approximately
0.4 and 1 £-1 appears to be related to wind direction, with the higher
concentrations measured when the air movement was from the urban area.
It is known (Morgan and Allee10) that engine operation using leaded gaso-
line produces many IN, but that typical polluted atmospheres deactivate
the IN. The measurements shown here and those taken on subsequent days
indicate, however equivocally, that the urban area is a minor source of
IN.
The following day, Saturday, 10 November 1973, (Figure 2) was clear
and warm, with the usual increase in AN corresponding to the daily in-
crease in traffic. The lack of conversion to CCN and scattering particles
may be explained by the linkage with synoptic airflow during mid-day
(Crow9). This served to limit the AN source concentration during that
time of the day when solar irradiation could have accelerated conversion.
Sunday, 11 November 1973, (Figure 2) was clear and warm, with air
flow from the south southwest (urban area) until about 1000, when the
wind direction shifted to northerly. The modest supply of AN from 1100
to 1400 came from the traffic east and north, but the source was too
weak to produce measureable CCN and bscat. After 1800 the wind shifted
to southerly and moved an aged pollution plume across the site. This
was dispelled when another wind shift occurred about 0200 on 12 November.
During the morning of 12 November 1973 (Figure 2), there were blus-
tery west winds closer to the mountains west of the site, but within the
valley pollution was visible. Monday morning traffic produced an ample
supply of AN, but the lack of direct sunlight retarded conversion to CCN
and scattering sized particles, and the comparatively high wind velocities
served to dilute the air mass before significant non-irradiated processes
could take place. During this time the comparatively high IN concentration
may have been produced partly by wind-raised dust. A lessening of wind
velocities and a shift in direction to southwesterly in the early evening
permitted the city plume to be carried to the site.
-------
-92-
~ 00 06
18 00
06
00 06 12
10 Nov 73 Saturday
06 12
Mov 73 Sunday
06 12
12 Nov 73 Monday
06 12 18 00
13 Nov 73 Tuesday
Figure 2 Parameters (see Figure 1) for 10-13 November 1973, Adams
County Fairground site. The notation "G12" in the wind
speed record shows the peak gust velocity during the
indicated hour.
-------
-93-
On 13 November 1973 (Figure 2), a modest increase in AN correspond-
ed to the daily traffic increase, but the changeable and sometimes blus-
tery winds continually diluted the valley air. At 1400 the first example
was seen during this study of the effects of high wind on the IN concen-
tration. This order-of-magnitude IN increase is from wind-raised dust;
we estimated gust velocities to be as much as 15 m sec-•'at the site at
1330.
The observations of 14 November (Figure 3) have been examined and
reported in detail (Van Valin et al.7), but the essential factors are as
follows: The maximum CCN, AN, and bsca^ occurred only about 1.5 h after
sunrise. Photolytic aerosol production was therefore not a factor; CCN
and bscat increases were from normal coagulation and reaction processes
taking place in the urban plume as it was carried northeastward from the
city. The sudden onset of the high-velocity "chinook" winds just before
1100 swept away the man-made pollution. AN, CCN, and bscat abruptly
dropped to very low levels, but the IN concentration just as quickly in-
creased by nearly two orders of magnitude. This is a vivid example of
natural aerosol production that was merely hinted at in the brief episode
of the previous day.
The expected AN increase appeared, coincident with traffic, on 15
November (Figure 3) and was accompanied, after a minor time lag, by a
small increase in bscaf Unfortunately, our CCN and IN instruments were
out of operation for all or part of this and the succeeding two days.
However, during most of this study period we noted a fair correlation
between CCN concentration and bscat, and we feel justified in assuming
a similar minor increase in CCN, followed by a return to low concentra-
tions after about 1100. The sky was mostly cloudy all day, preventing the
sunlight-accelerated particle production, and the fairly vigorous wind
flow was from the southeast, preventing most of the urban plume from
reaching the site.
The 16th and 17th of November (Figure 3) were mostly cloudy, with
the more-or-less expected AN supply. At 1000 on 16 November a minor
pollution concentration was visible west of the site, but the easterly
winds supplied fresh air and during the day no pollution build-up was
measured. At about 1900 the wind became southwesterly, with lower veloci-
ty, and the urban plume was carried across our sampling site; the maximum
concentration, according to bscat, was at 0300 on 17 November.
On 19 November we moved our mobile laboratory to the Humane Society
site, next to the EPA mobile field laboratory. A snowfall began during
the afternoon, continued through the night, and ended the following morn-
ing. No observations worth reporting were made until late afternoon of
the 20th (Figure 4), when a modest increase in CCN and bscat indicated
the presence of the urban plume.
-------
-94-
00 06 12 18 00 06 12 18 00 06 12 18 00 06 12
00 06 12 18 00 06 12 18 00 06 12 18 00 06 12 18 00
14 Nov 73 Wednesday 15 Nov 73 Thursday 16 Nov 73 Friday 17 Nov 73 Saturday
Figure 3 Parameters (see Figure 1) for 14-17 November 1973, AJ
' Adams
County Fairground site. Peak gust velocities are noted
in the wind speed record.
-------
-95-
The observations of 21 November (Figure 4) have been analyzed in
detail and reported (Van Valin et al.7), but a summary is given here.
A heavy fog from 0400 to 0540 produced elevations in bscat and the appar-
ent CCN concentration, but pollution-produced increases in these para-
meters and in AN coincided with the morning traffic increase. Solar
irradiation accelerated the conversion, which resulted in a relative de-
crease in the AN concentration. At about 1200, the wind shifted to the
east and a cloud cover moved in. The CCN and bsca^. immediately decreased
to essentially "clean air" levels, whereas the AN concentration continued
at fairly high levels, varying only in response to the varying rate of
dilution by the east wind. The important role of sunlight is illustrated
here. Because of the location just west and northwest of heavily travel-
led roads we would have expected some increase in CCN and bscat if the sun
had been shining. We believe that the indicated increase, during the pol-
lution episode, of the IN concentration was real and is an indication that
engine-produced IN were not completely deactivated before reaching the
site; nearly complete deactivation seemed to be the case at the more re~
mote Fairgrounds site.
Beginning at about 0200 on 22 November (Thanksgiving Day) (Figure
4), the city plume was carried across the site, but traffic was very light
during the day and the easterly winds around midday supplied clean air.
A tripped circuit breaker dropped our CCN counter off the line, but there
is no reason to believe that CCN measurements would have contradicted the
bscat roeasurements. Late in the evening a southwest wind carried an aged
aerosol cloud across the site. The density of this pollution cloud de-
creased in the early morning hours of November 23 (Figure 4) and then was
somewhat replenished as the morning workday traffic began. At about 1000
the wind shifted to southeasterly and became stronger, but under the in-
fluence of sunlight the bscat declined only gradually. This observation
further supports our previously stated thesis (Pueschel et al.6) that sun-
light is a powerful factor in the creation of CCN and light scattering
particles.
In Figure 5 we show an aerosol size distribution for membrane fil-
ter samples collected on 21 November at the Humane Society site. This
was done by a replicating technique with the transmission electron micro-
scope. The dashed line represents six samples collected from 0610 until
0830; the solid line represents four samples collected from 0900 until
1126. The aerosol populations are the same at the lower end of the size
distribution, but in the particle size range from 0.08 ym to 0.25 ym the
particle concentrations are significantly different and show the impact
of automobile traffic. There were too few particles larger than 0.7 ym'
to form a statistically significant sample.
The ability of particles to serve as CCN depends upon both particle
size and chemistry; at supersaturations of less than 0.5 percent, such
as are found in nature, water insoluble particles of diameter less than
approximately 0.6 ym will not be active as CCN (Pueschel and Weickmann1*).
It is apparent, therefore, that either particle solubilization or parti-
cle growth must take place before the automobile engine-derived particles
can enter into the liquid water nucleation picture.
-------
-96-
00 06
18 00 06 12 18
20 Nov 73 Tuesday 21 Nov 73 Wednesday
06 12 18 00 06 12
22 Nov 73 Thanksgiving 23 Nov 73 Friday
Figure 4 Parameters (see Figure 1) for 20-23 November 1973,
Humane society site.
-------
-97-
rO
'E
o
10
o
en
o
o
"o
0)
o
c
o
o
o
10
10
i i i i i i i
i i i i i i i
1 I
0610-0830 (6 Samples)
0900-1126 (4 Samples) J
I I I
0.01
0.1 1.0
Particle Diameter D (/xm)
Figure 5 Aerosol size distribution at the Humane Society site
on 21 November 1973.
-------
-98-
CONCLUSIONS
Examination of the aerosol parameters measured during this study
in the vicinity of Denver leads to the following points:
AN are produced primarily by automobile traffic.
AN are converted by coagulation and chemical reaction
to CCN and particles of light scattering size.
This conversion of AN to CCN and light scattering particles
occurs slowly in the dark and much faster in the sunlight.
Although it has been shown that operation of engines on
leaded gasoline produces large numbers of IN, most IN
are deactivated in the polluted atmosphere.
Automobile traffic in Denver constitutes a low-level source
of IN.
Natural pollution in the form of dust raised during high
winds introduces orders-of-magnitude more IN to the atmos-
phere than do Denver area emissions.
ACKNOWLEDGMENTS
We take this opportunity to thank Richard A. Proulx for his part
in this study. His willingness to work early and late, and his effective-
ness at all times were important and at times decisive factors in the
success of the project.
-------
-99-
REFERENCES
1. Schaefer, V.J. The Inadvertent Modification of the Atmosphere by
Air Pollution. Bull. Amer. Meteor. Soc. 50: 199-206, April 1969.
2. Barrett, E.W., R.F. Pueschel, H.K. Weickmann, and P.M. Kuhn. In-
advertent Modification of Weather and Climate by Atmospheric
Pollutants. ESSA, Boulder, Colorado, ERL 185 - APCL 15. September
1970, 103 p.
3. Hidy, G.M., W. Green, and A. Alkazweeny. Inadvertent Weather
Modification and Los Angeles Smog. J. Colloid. Interface Sci.
39: 266-271, April 1972.
4. Weickmann, H.K. Man-made Weather Patterns in the Great Lakes
Basin. Weatherwise 25: 260-267, December 1972.
5. Hobbs, P.V., L. Radke, and S.J. Shumway. Cloud Condensation Nuclei
from Industrial Sources and Their Apparent Influence on Precipita-
tion in Washington State. J. Atmos. Sci. 27: 81-89, January 1970.
6. Pueschel, R.F., C.C. Van Valin, and F.P. Parungo. Effects of Air
Pollutants on Cloud Nucleation. Geophys. Res. Letters 1: 51-54,
May 1974.
7. Van Valin, C.C., R.F. Pueschel, F.P. Parungo, and R.A. Proulx.
Cloud and Ice Nuclei from Human Activities. Atmospheric Environ-
ment. In press.
8. Riehl, H. and L.W. Crow. A Study of Denver Air Pollution. Dept.
of Atmospheric Science, Colorado State University, Fort Collins.
Technical Paper No. 33, June 1962, 15 p.
9. Crow, L.W. Airflow Study Related to EPA Field Monitoring Program,
Denver Metropolitan Area, November 1973. Prepared for Chemistry
and Physics Laboratory, Environmental Protection Agency. LWC No.
128, February 1974, 64 p.
10. Morgan, G.M. and P.A. Allee. The Production of Potential Ice
Nuclei by Gasoline Engines. J. Appl. Meteor. 7: 241-246, April
1968.
11. Pueschel, R.F. and H.K. Weickmann. Pollutant Scavenging by Aero-
sols and its Effect on the Cloud Nuclei Budget. Proceedings,
Precipitation Scavenging - 1974 Symposium, October 13-18, 1974,
AEC. In press.
-------
-101-
ATMOSPHERIC AEROSOL DYNAMICS: THE DENVER BROWN CLOUD
P. B. Middleton and J. R. Brock
The University of Texas at Austin
Austin, Texas 78712
ABSTRACT
Pollution conditions which arise in the Denver area are studied
using a numerical model for the evolution of the aerosol size distribu-
tion. The time variation of the particle size distribution, total
particle number, total particle mass and light scattering coefficient
provide valuable insight into the mechanisms of particle growth and the
color of the "brown cloud."
INTRODUCTION
In order to propose rational control policies for the Denver "brown
cloud," a knowledge of the dynamics and relative importance of the vari-
ous pollutants is required. To aid in the interpretation of the field
study data on particulate matter for such "episode" conditions, we have
developed a computer model for the evolution of the Denver aerosol.
Although the model represents a gross simplification of typical Denver
"episode" conditions, it does provide valuable insight into the dynamics
and produces several useful predictions about the mechanisms of particle
growth and the color of the "brown cloud." Details of the model and
analysis of the results follow.
DYNAMIC MODEL
The dynamic model includes advective and (implicitly) convective
flows, meteorological conditions, and aerosol processes assumed to be
operating during a typical "winter episode" in Denver. Basically, the
evolution of the particle size distribution in a parcel of air travers-
ing the city is considered. The parcel itself is bounded by the ground
and the top of the mixing layer. Initially the parcel contains a dis-
tribution of particles. As a parcel moves across the city the aerosols
within the parcel evolve according to a number of mechanisms.
The description of air movement is highly simplified. "Heat
island effects," the "chimney effect" and complex air flows such as
horizontal dispersion and advection other than drainage flows are
neglected. The assumption of relatively small mixing heights and
-------
-102-
cyclical drainage flows is supported by several previous meteorological
studies ' . The evolution of aerosol itself depends upon coagulation,
dry deposition, primary source input, condensation (as a function of
ozone concentrations), vertical mixing, and the advective drainage flows
associated with the Platte River Valley. This basic model is schemati-
cally depicted in Figure 1.
The mathematical expressions which describe the various processes
are only outlined in this paper since complete derivations are avail-
able in the literature-^'^»->. First we present the general equation
for the evolution of the particle size density and then we discuss
each growth term separately. The parameters particularly relevant
to the Denver aerosol are noted.
We believe that all current evidence, to be discussed below,
supports the view that the Denver aerosol during winter episode condi-
tions, is dominated by primary sources. A very accurate description
of the dynamics of such an aerosol is afforded by the evolution equa-
tion for the particle number density function. The following relation
is assumed to describe the evolution of the density function n(x,r,t)
for an aerosol with convective transport-1' !
-. (x,r,t) + y.vn(x,r,t) = 7«K. Vn(x,r ,t) (1)
ot
x _,._,.
+1/2 / dx'b(x-x',x')n(x-x',r,t)n(x',r,t)
0^ oo
-n(x ,r,t) / dx'b(x' ,x)n(x' ,r,t)
4,(x)n(x,r,t) + --J *(x)n(x,r,t)
G(x)-Vn(x,r,t) + £v (x,r, t)+£\>N (x,r,t)
P P
where n(x,r,t) is the number of aerosol particles having mass, x, in
the range x, dx at point r in space at time t. V is the fluid velocity.
The first term on the right hand side of this_equation accounts for
the turbulent dispersion of particles, where K is the eddy diffusivity
tensor. The second and third terms on the right hand side of the equa-
tion represent the change in n(x,r,t) due to coagulation ''. The
-------
-103-
MIXING BASE
Drainage
Flow
T
I
I
I
-I
V
Volume- Averaged
Air Parcel
1
Dry
Deposition
Primary
Sources
r
I
i
i
i
i
i
i
i
FIG, 1: SCHEMATIC DIAGRAM OF MODEL
-------
-104-
fourth and fifth terms describe the change in n(x, r, t) owing to
densation of trace gaseous substances. In the next term G(x) is the
sedimentation velocity of a particle of mass x. ^ (x,r,t) represents
the rate of production of particles of mass x at r, t from primary
sources. \>N (x,r,t) is the rate of production of particles of
mass x at j r,t by homogeneous nucleation of the j-th chemical
species. It should be noted that heterogeneous nucleation of con-
densable species is accounted for by the condensation terms, four and
five. The form of the coagulation coefficient b(x,x'), the condensa-
tion coefficients ijj(x) a(x), and the details of the other terms are
discussed below.
Equation 1 is, of course, for the general case of an aerosol in
the atmosphere, or any fluid in general motion, coupled to the energy
and momentum balance equations. It is, of course, also coupled
closely to the continuity equations for the various chemical species
undergoing homogeneous nucleation and condensation with the aerosol
phase-*. For any turbulent fluid, equation 1 is not in a useful form.
By analogy with the usual procedure for turbulent dispersion of
non^reactive gases these quantities can be expressed in terms of time
averaged quantities. Subsequent to this time averaging, we carry out
a volume averaging procedure, where the volume average of the density
function is defined as
= i / n(x,r,t) dr. (2)
where for brevity is written for . The control
volume V may be thought of as the volume of a "box" whose vertical
bounds are determined by the ground and the top of the mixing layer.
The box covers a surface area of 5,3 km^ which is set by the estimate
of the average wind speed (see below) .
Through this volume averaging procedure we will, in effect, fol-
low the evolution of the particle size distribution in a particular
control volume as it moves across the city. This model, obviously,
involves gross simplifications in that the pollutants in the box are
assumed to be well mixed from ground to the mixing height and diffusion
out of the box is not considered. In particular, a moving control volume
was chosen in order to study explicitly the evolution of the aerosol in
the cyclical drainage flows associated with winter episode conditions
in Denver. A more comprehensive and conventional model could be
constructed through the use of a multiple box model in a fixed matrix
over Denver and its environs.
-------
-105-
Unfortunately, we had insufficient information for Denver to
permit such calculations. Information is now being acquired and it
is our purpose to carry out more detailed analyses. In this paper we
carry out the calculation described above for a single control volume
which represents an average for the entire city.
Following through with our present development, the volume
average of equation (1) is taken according to equation (2) . If the
resultant volume integrals are converted where applicable to surface
integrals, an evolution for <5n(x)> is obtained having the form:
Cl/2) / b(x-x',x')dx' (3)
0
/ b(x,x')dx'
0
^
- - <\Kx)>
9x
+ Z T.W.(x)
P
For daytime conditions the volume averaged density is
representative of n(x,r,t) throughout the mixed layer. However, at
night there may be a substantial vertical variation in n(x,r,t), and
therefore the meaning of may become less precise.
Several other changes between equation (1) and equation (3) should
be noted. The volume averaging procedure introduces explicitly the
unspecified boundary conditions of equation (1) into equation (3). For
the system under discussion, these boundary conditions appear in
equation (3) as the term which represents those processes
effecting removal of particles from the control volume,v • Homogeneous
nucleation is neglected since the time scale over which the nucleation
rate might be significant is small compared to the time scales of the
other rate processes. The last term on the right hand side of
equation (3) represents the volume average of the rate of production of
particles from primary sources and the summation extends over all
primary sources, i, , , and g(x) are explicitly functions
of time; T^ depends implicitly on time as will be indicated below.
-------
-106-
COAGULATION
For the coagulation coefficient b(x,x') we assume the particles
are interacting by Brownian coagulation only. Given this assumption,
b(x,x') becomes 3,4_
b(x',x) = 4u(R + R') (A + A')a (4)
x = 4/3iT R3p
u = (8kT/Tix)1/2
A = (kT/6uyR) {1 + f (A + B - ) }
a
9 71/7
+ 4 (A + A')/( R + R ) (u + u' V
where p is the particle mass density, y and fi are respectively the
viscosity and mean free path of the gas and A, B, C are the constants
in Millikan's empirical equation for particle mobility^. R, A, and u
are respectively the radius, diffusion coefficient and mean thermal
speed of a particle. kT is the thermal energy. Compared to other
coagulation mechanisms, Brownian coagulation appears to be dominant
for the Denver aerosol. This feature will receive future study in
our laboratory.
CONDENSATION
In equation (3) condensation is represented by two terms,
2 ,
-£i-j(<;a(x)><-n(x) >) - -r- (<\Jj(x) >. For the purpose of this paper,
ax
the first term which describes the dispersion of the particle size
distribution is neglected. Furthermore, the second term only accounts
for the diffusional transfer of chemical species to the surface of
a particle, Motivation for these assumptions is discussed elsewhere.?
By definition has the form;
1.21u.(S. - S. e** )x2/3
^00 > L.J 12 (5)
5 1/3
1 + O.aSCe"1'4'10 )x -1) + 5.43(104)x1/3
-------
-107-
where S is the mass concentration of component j corresponding to
the vapor pressure of j over a plane liquid (or solid) surface, u.
is the mean thermal speed of molecule j. K is the Kelvin coefficient
appropriate for the condensing substance associated with the Kelvin
effect:
K = 2yVJL/kT (6)
where y is the surface tension at a particle-gas interface and V is
the volume of a vapor molecule of j in the liquid (solid) phase.
S. is the mass concentration of the j1-'1 chemical species.
Since the conversion of H-SO, aerosol is assumed to be the only
significant mechanism for secondary source input, we need to consider
only one chemical species. Given this assumption the mass concentration
S. can be calculated from the concentrations of ozone, olefin, and
sulfur dioxide and from the known mass conversion rate of H_SO, in the
2. fy
following way. The mass conversion is just the rate of change of
particle mass due to condensation, dM /dt:
x
* / dx (7)
Q
Given the concentrations of olefin, ozone and sulfur dioxide during an
episode day in Denver", the volume conversion is deduced from experi-
mental results relating the H_SO, conversion rate to the chemical
concentrations". The concentration of olefin is assumed to be 8.2 ppm
and the concentration of sulfur dioxide is .066 ppm°. These values
remain constant throughout the calculations since the diurnal varia-
tion of these substances is not significant. The ozone concentration,
CO, however, varies in time by the following perscription:
C033 = ,0025, t=0 - 600 and 1800-2400 hours (8)
C033 - .00.25 ppm + (.09 ppm - .0025 ppm) sin(|- (t~QQ-°)). t=600-1400 hours
CQ33 => .0025 ppm + (.09 ppm-.0025 ppm) sinO| C^rM—)), t=1400-1800 hours
This model for the diurnal variation is illustrated in Figure 2.
DEPOSITION
The deposition coefficient, , is expressed as
= V(x,t)/H(t) (9)
-------
-108-
0.10
E
Q.
Q.
LU
z
o
M
O
0.05
0
0
6 12 18
TIME (hours)
24
FIG, 2: DIURNAL VARIATION OF OZONE
-------
-109-
where H(t) is the mixing height (to be discussed later) and V(x,t) is
the deposition velocity:
V(x,t) = v(x,t) + VQ(x) . (10)
V"0(x), the gravitational settling velocity, is given by:
VQ(x) = grr Rp/67TyR (11)
where g is the acceleration of gravity, p is the particle density
and u. is the viscosity of air. The density of quartz (a principal
constituent of the large particle mode) is taken in this calculation
as the particle density. V~(x) is constant in time but v(x,t) varies as
v(x,t) = U(x) L(t). (12)
U(x) is fit to the sedimentation velocity estimated by Sehmel for a
friction speed, U^, of 20 cm/sec and a surface roughness of 10.0 cm.
Sehmel's velocity estimates are illustrated in Figure 3. When U =0,
of course, the only operant removal mechanism is gravitational
sedimentation. The dynamics of the aerosol is very sensitive to the
choice of this deposition velocity. The values used in this simula-
tion were arrived at by trial and error and represent a best, but not
necessarily unique, choice. The relationship of v(x,t) and wind
speed is taken into account by the time factor L(t) — a monitorRof the
diurnal variation of wind speed for a typical Denver "episode" .
L(t), illustrated in Figure 4, varies in the following way:
L(t) - 0.35 cos (-J- ()) + Q.15, t = 0-400 hours (13)
L(t) * Q.85 sin (- (')) + 0.15, t = 400-1400 hours
L(t) = 0.65 cos 0? (n)) + °-35, t = 1400-2000 hours
2
L(t) = 0.15 sin (()) + °-35> t = 2000-2400 hours
In the volume average context, the deposition process at the surface
is converted into an effective volumetric sink term. Hence, during
night-time meterological conditions, some imprecision is introduced
through this procedure.
-------
-110-
o
o>
CO
\
E
o
O
O
_l
LU
0
CO
O
I I I [ I I I I! I I I | 1 I I
" =0
i i I i i n i i i 1 1 1 i i i i I i 1 1
10
-2
10"' I 10
PARTICLE DIAMETER (jim)
10
-2
FIG. 5: DEPOSITION VELOCITY AS A FUNCTION
OF PARTICLE SIZE
FRICTION VELOCITY = 20 CM/SEC,
ROUGHNESS HEIGHT = 10 CM,
-------
-Ill-
.00
a:
o 0.75
o
<
U.
LU
Q_
C/)
O
0.50
? 0.25
0
0
6 12 18
TIME (hours)
24
FIG, 4: DIURNAL VARIATION OF WIND SPEED FACTOR
-------
-112-
PRIMARY SOURCES
The primary source term is the sum of all the source contributions:
)2) (14)
F(t) is the valley flow factor which accounts for advection and
H(t) is the mixing height. Since no information on source size
distributions is available, each of the five sources are assumed to
be lognormal distributions in mass. The magnitudes A. are calculated
from emission inventories for Denver, I., in the following manner.
For a lognormal distribution the first moment, which is the total
mass emission flux, I., is given by,
The sources and their emission fluxes, I., are listed in Table 1.
These rates were calculated from the reported tons per year values
given by the PEDCO study . In Table 2 are listed the particle „
diameters D., corresponding to the mean mass and the variance, (#ng.) ,
of the lognormal distributions of the primary sources. Size distribution
data for industrial sources were obtained from a report by Shannon and
German-^. The data for dust were calculated from a paper by Schutz
and Jaenickel3. Transportation is taken from an estimate of
residence times in an exhaust system^. Each source magnitude A.
is constant over a one hour period, but varies from hour to hour.
This time dependence of sources reflects the diurnal change in
emission rates. As illustrated in Figure 5, the point sources and
fuel combustion are constant whereas traffic dust and traffic exhaust
have maxima during rush hours. Construction dust is a constant
during working hours and zero otherwise.
VERTICAL MIXING
Vertical mixing is accounted for by the time variation in mixing
height H(t) and volume correction to the density. The time variation
in mixing height for a typical episode day^-% illustrated in Figure 6,
is given as;
-------
-113-
Table 1. ESTIMATED EMISSION RATES FOR PRIMARY PARTICULATE SOURCES
I , Eq. 15
Emission Flux
24 hr. Average
Sources g/km^sec Fraction of Total
1. Traffic dust 0.815 0.428
(unpaved roads, sand
and salt on roads)
2. Construction dust 0.23 0.12
3. Point sources 0.14 0.075
4. Fuel combustion +
process losses +
solid waste disposal 0.18 0.095
5. Transportation 0.34 0.18
TOTAL 1.90 1.00
Table 2. SIZE DISTRIBUTION PARAMETERS FOR PRIMARY PARTICULATE SOURCES
JParticle Diameter
0 D.(ym) 3 . . Eq. 14
Source i i
1. Traffic dust 5 5
2. Construction dust 5 5
3. Point sources .2 2.5
4. Fuel Combustion +
process losses +
solid waste disposal .2 2.5
5. Transportation .10 2
-------
-114-
io-101-
(VJ
e
o>
X
LU
O
o
(S)
10-" I—
10
-12
Traffic
Dust
Fuel
Combustion + Etc
Point
Sources
-^-Traffic
Exhaust
Construction
Dust
8 12 16
TIME (hours)
FIG. 5: DIURNAL VARIATION OF PRIMARY PARTICIPATE SOURCES
-------
-115-
2000 -
0>
cu
X
C£
UJ
X
CD
500
6 12 18
TIME (hours)
24
FIG, 6: DIURNAL VARIATION OF MIXING HEIGHT
-------
-116-
H(t) = 200 meters, t = 0-600 and 1800-2400 hours (16)
H(t) = 200 meters + 500 meters sin (?r (t~ . )), t=600-1400 hours
2. oUU
H(t) = 500 feet + 1500 feet cos ( ()), t=1400-1800 hours.
When the mixing height changes, the volume of the region over
which the density is averaged changes. As a mixing height increases
the density decreases assuming there is no substantial particulate
concentration above the mixing height as it rises. As the mixing
height decreases, the density increases due to the convection but
also decreases due to the loss of particles above the mixing height.
The density corrected for the increase of mixing height is
(1) _ H (1)
n t + At ~ HfTAi n t (17)
and the density corrected for the decrease of mixing height is
(at the appropriate time step)
(1) (18)
t + At H - AH t •
These modifications are illustrated in Figure 7.
ADVECTION
The advective drainage flows are taken into account by the
valley flow factor F(t) in the source term. A background source
input is always assumed. However, as the air parcel moves out of
the city the source input decreases. When the air parcel is over
the city the sources are full strength. Figure 8 illustrates the
diurnal variation of F(t). We chose an air parcel which was north
or the city at midnight, moved south until the afternoon and then
reversed its flow and drained down the valley in the late afternoon
and evening. Assuming an average wind, speed of 0.9 meters per second,
the air parcel would complete the cycle of moving up and down the
valley in one day.
-------
-117-
H
n1
Mixing Height
Increases
H
AH
n
Mixing Height
Decreases
FIG. 7: VOLUME CORRECTION TO
PARTICIPATE NUMBER DENSITY
-------
-118-
F(t)
.0
0.5
0.0
TT
Drainage
Flow
! Reverse
! Flow
0
I
6 12 18
TIME (hours)
24
FIG, 8: DIURNAL VARIATION OF VALLEY FLOW FACTOR
-------
-119-
INITIAL SIZE DISTRIBUTION
For the initial size distribution we assumed a typical night-time
aerosol distribution-"-^. The distribution is expressed as the sum
of two lognormal functions:
= exp (-(Inx-lnx)2/2(lne)2) (19)
exP(-(lnX-lnx)2/2(ln£)2)
The values chosen for the £., e., and x. are discussed below.
RESULTS
From the results of the computer calculations we are able to
comment on the role of sources on aerosol growth, the relative
importance of photochemical production of aerosol, and the cause of
the "brown cloud" color. The main inadequacies of the program —
the volume averaged deposition and uncertainty of input data for the
sources — are studied. We also include suggestions for improvements
in the computer model and recommendations for further field study
measurements.
The computer studies, outlined in Table 3, are divided into
f,our categories. Runs 1 through 4 study the effects of modifying the
deposition term,. Run 5 examines the importance of photochemistry in
aerosol production. Run 6 and run 7 show the influence of varying
the source input distributions. For all of these runs advection
and vertical mixing are constant. This means the mixing height is
constant and the air parcel does not flow down the valley. In each
run only one aspect of the model is varied. Since these are just
case studies, the runs take place between nine o'clock and noon.
Run 8 however, is a simulation over forty~eight hours and it includes
all of the time dependent mechanisms.
The results of the computer runs are illustrated in Figure 9
through Figure 16. For each run the size distribution at various
times is plotted as a function of particle diameter. In addition,
the moments, total number, total mass and light scattering coefficient
are plotted as a function of time. These moments are defined in the
following way:
-------
-120-
Table 3. COMPUTER STUDIES*
Processes
Computer
Run Advection
1
2
3
4
5
6
7
8
NV
NV
NV
NV
NV
NV
NV
V
Convection
NV
NV
NV
NV
NV
NV
NV
V
Conden- Time
Sources Deposition sation (Hours)
Dry
deposi-
tion
V V
V 0
V 0
V NV
V V
V+ V
V ~ V
V V
Gravi-
tation-
al set-
tling
NV V
NV V
0 V
NV V
NV 0
NV V
NV V
NV V
9-12
9-12
9-12
9-12
9-12
9-12
9-12
0-48
*NV = time independent, V = time dependent, 0 = omit
Source input distribution changed for combustion and point sources.
Source input distribution changed for dust sources.
-------
-121-
oo
Total Number Concentrations, N = / dx (20)
o
Total Mass Concentration, M = /dx
° ~2
Light Scattering Coefficient, b = TT/R Q(A/R,m)dx
s o
where Q(x/R,m) is the light scattering efficiency factor obtained
from Mie theoryl^. For the wavelength of the incident light,X ,
we chose a typical nephelometer setting of 500 nanometers. The
refractive index, m, is assumed to be (1.55-0.022i), a value that
is believed to be in the correct range for the Denver aerosol^?.
To study the color effects in the Denver "brown cloud", bscat is
calculated as a function of wavelength and refractive index for the
simulation run. For all the runs, the above values are assumed.
The parameter values for the initial distributions are given
in Table 4. For the simulation run, a low background distribution is
assumed-*- . For the shorter runs, the initial distribution is the
nine o'clock distribution from the simulation run. Modifications
of the source distribution parameters for run 6 and run 7 are listed
in Table 5. Unless otherwise stated the distribution parameters are
the same as those listed in Table 2.
Variation in deposition is studied because the results are
most sensitive to this process. In Figures 9 through 12 we present
the results of varying the form of the deposition term. In the
first run the complete deposition term is used as defined in equation
(9). In run 2, only gravitational settling is present. In run 3,
deposition is omitted completely. Last, in run 4, the time dependence
of dry deposition caused by changing wind speed is neglected. As one
might expect, the aerosol grows most rapidly in the absence of
deposition. The true deposition rate probably lies somewhere
between the two extremes of no deposition and the complete deposition
process.
The importance of the photochemical production of aerosol is
illustrated by run 5 for which condensation is neglected. A com-
parison of Figure 13 and Figure 9 shows that condensation has only
a very small effect on the growth of the aerosol. Although growth
of the "accummulation" mode (o,l ym to 1.0 urn diameter range) is
usually assumed to be evidence for photochemical activity, we find
that this mode grows at almost the same rate in the absence of any
photochemical input. This fact supports the idea that the aerosol is
primarily source dominated. The conclusion—that photochemistry is
not too important during the Denver pollution episodes—is reasonable
considering the typical temperatures. It has been shown that photo-
chemical reaction and subsequent aerosol growth are reduced by a
significant amount 1°"" at low temperatures such as are prevalent
-------
-122-
Table 4. INITIAL DISTRIBUTION VALUES
Mean Value of DM/DLOG(D)
Diameter (microns)
Variance
Magnitude
of DM/DLOG(D)
Run
8
1-7
Mode 1
.2
.2
Mode 2
3.0
9.0
Mode 1
5
5
Mode 2
5
5
Mode 1
.65
18.0
Mode 2
1.63
18.0
Run
Table 5. SOURCE TERM MODIFICATIONS
Particle Diameter
Source Changed Dj(ym) &i » Eq- 14
Point sources and fuel 0.6, 0.6 1.1, 1.1
combustion
Dust (construction
and traffic)
20, 20
5, 5
-------
-123-
* 4-1
T3 M
O> C CO
6 -H 4-1
•H CO
4J QJ
e 60
M -H C
O 4-1 -H
M-i i-H
T3 4J
CO Pi 4->
4-1 Cfl 0)
(H CO
0) C
B 0 rH
O -H cfl
B J-> C
cfl o
T3 CO -H
C (3 4J
cfl 0) cfl
13 4-1
C Cl -H
O O >
•H O CO
^_) J^
3 -T3 O
^ C
•H cfl T)
J-i C
4-1 « Cfl
CO fi
•HO«
13 -H C
4-1 O
01 -H -H
N CO 4-J
•H O 0
CO P- 01
0) > •
rH P ti CN
0 0 rH
CO >> C_>
O S-< frf
^-t f~^ ** p~]
tp a
Cfl •> O 4-1
CO -H cfl
14H 0) 4-1
O CJ CJ 60
S-J 01 C
C 3 > -H
O O rO ""O
•H CO <1 C
4-1 01
3 co en
r-H 0) 0) T3
O t-H iH fi
^ Q ^rt td
QJ C^ CO
•H -H a-.
a) J-i v-i
pd cfl CO Pi
EH > > K
-------
-124-
i
o
i
pi
I
o
I
XI
o
a
C/l
C/3
I
cs
Pi
Cti
X!
o
c_>
C/3
pq
O
I
•K
I
X!
O
O
Pi
W
a
4-1
C
CU
•X)
C!
Q)
P. CO
CU f 4-1
CU cfl O
cu
i-l 4-1
cfl -H
c e
o o
•H
4-1 CO
cd -H
4-1
•H C
> o
cfl TH
!-J 4->
60 *rl
CD
T3 O
C P.
cfl cu
Pi >.
O M
•H O
4-1
O
cu •
> CN
C rH
O Pi
60
'H CU -H
S>.g
cd
cu
-------
-125-
i
o
i
c?\
PS
PS
I
o
X
u
u
to
en
PS
XI
e
o
o
en
M
O
QJ
B
•H
4-1
M
O
M-t
CO
4-1
C
0)
e
o
6
13
C
cd
CO
C
0
•H
4-1
3
,£]
•H
S-i
4-1
CO
•H
13
O)
N
•H
CO
rH
O
CO
0
!-i
OJ
cd
4J
C
a)
13
C
0)
a.
cu
T3
a
•H
a)
e
•H
4J
13
C
cd
C
0
•H
4-1
cd
CO
C
01
T)
C
o
o
•a
C
cd
CO
CU
a
S-i
|3
O
to
•
CO
eu
4J
cd
60
C
•H
T3
a
eu
13
C
cd
CJ\
PS
K
4-1
cd
60
C
•H
4-1
>-l
cd
4-1
CO
C
o
•H
4-)
CJ
0)
>
C
0
u
13
a
cd
a
0
•H
4-1
•
T3
cu
4J
4J
•H
6
O
a)
!-j
cd
60
C
•H
rH
4-1
4J
0)
CO
rH
cd
C
O
•H
4-1
cd
jj
•H
>
cd
S-i
60
13
cd
fi
O
•H
4J
•H
o cd
•H
C rt
o cd
•H >
O CO
eu o
0)
3 4-1 co o
rH (3 0)
O 0) i-H
> T) ,0 •
a) C cd CN
CU -H rH
cu ex s-i
,c cu cd PS
H 13 > ffi
-------
-126-
i
o
i
i
E co
•H -H
ca
M K"
o
c
co cu
c c
cu cu
60
c
•H
4-1
}-t
ca
4J
CO
C
o
•H
4-1
•H
CO
O
01
13
>->
j_j
B CX 13
O 0)
B 'O
C
I"O *H
c
CO 01
B
CO -H
c! 4J
o
•H 13
4-1 C
3 CO
•H fi
t-l O
4-1 -H
CO 4J
•H ca
13 CO
C
Q) CU
N T3
•H C
co o
u
r-H
O 13
CO C
O CO
^_j
CU CO
CO 0)
o
t[ 1 J_(
0 3
O
C co
o
•H
4J •
3 CO
rH 01
0 r-l
O> CO
•H
CU V-l
f: ca
H >
"^3
C
ca
60
c
•H
1— 1
4-1
4J
CU
CO
iH
CO
C
0
•H
4-1
CB
4J
•H •
CO rH
^
60 fyj
ffi
pj 4J
O CO
•H
4J 60
0 C
CU -H
£> *T3
C C
o cu
o
13
- G
C tO
o
•H O*\
0 pt!
CU ffi
T3 4-J
-------
-127-
^^^
^^
••• d
: ^^,
^"S.
>
"
]*
: <»
< >
: j/
^^^^
s^z^^
^^C.
^^^r^
vk
; ^Nf,
; ' '
,,
- •
\\
f '
: o
_
o
cu
6
•H
1 1 4-1
°e = o + n
0=' 0
uj | | 4_j
UJ
cE Cg CO
^ C^ r-H 4-J
c
Dd Qd 1)
« « g
g
*"O
c
LO CO
~2 g co
5 c
o P5 0
C-J «H
J_J
33/Od ' [ 0)0010/UO •• 3
co ,n
i-H -H
i-t
O CO
M 'H
PCH T3
ni
i«
*
<
*
t
~
(>
1 1
T T
r *
1
T *
*
T
~L
1
«
1
T
1
T
*
T
t
,t
>
N
•H
M
III iH
O
O + * W
O
III M
- (1)
r^
i-l i-i O
f \ X
2 0 X fi
S rH 0 0
gU -H
X 0 ^ 4J
^ -H 3
O rH rH
O O
* fX-t £*
H W 0)
co c_> S a)
S PQ £3 H
0,
4J
a
a)
c
0)
p.
OJ
TlJ
0
•H
0)
g
•H
4-1
Tj
c^
CS
O
•H
4-1
Cfl
0
a)
T?
60
a
•H
j_.
cfl
4-1
CO
60
•H
i— 4
4J
4-1
0)
CO
Cfl
a
0
•H
4J
cfl
^J
•H
>
cO
&0
!>,"d
•d
T3
(3
cfl
CO
a)
a
}_i
p
o
CO
•
co
a)
rH
,0
CO
•H
^
Cfl
[>
4-1
a)
""CJ
|3
ex
•
13
<1J
4-1
4-1
•)-|
g
O
Cfl
'
G
O
•H
4-J
Cfl
CO
a
Q)
c
o
u
.
CN
i — |
Pi
»
4-1
CO
ox
(3
•H
T3
C
-------
-128-
during a pollution episode in Denver.
The last aspects of the model to be considered are the sources.
Since there are little data on the size distribution for the different
source categories we took to be operating in Denver, we could only
r-liyo-IO
estimate distribution values J»-LJ-»-LZ»-"-->. It would be particularly
useful to have such measurements available for Denver since the
selection of such parameters significantly affects the evolution of
the aerosol. In run 6 we changed the point source and fuel combustion
distribution by moving the peaks to a larger diameter. As seen in
Figure 14 this shift gives rise to a rapidly growing, unrealistic
third mode. In run 7 we shifted the dust distribution peak to 20 ym
diameter. This mode, as shown in Figure 15, decreases at a faster
rate than the coarse modes (D>1.0um) in Figures 9-13. In fact, the
coarse mode for run 7 can never grow, whereas the probability of
accumulation is much higher when both the dust source distribution
and the initial distribution are peaked at lower values of particle
diameter. The sensitivity of the model to source input should be
studied further. In addition, it would be desirable to obtain more
detailed information about the source distributions for the Denver area.
The simulation study, run 8, is perhaps the most interesting
run because we observe the slow build-up of aerosol over forty-eight
hours, noticeable diurnal variation during the first day, and then
follow the build-up of the aged aerosol during the second day.
These results are illustrated in Figure 16. The total mass, total
number, and bscat are all within the range of experimental observa-
tions^ 14, 15,23, we also see a definite correlation between bscat
and the fine particle mode.
Since this model produces reasonable results when compared
to field study measurements we can venture several conclusions
about the Denver aerosol based on our calculations. The Denver
episode aerosol is probably source dominated with photochemistry
acting only as a minor secondary source. This means that the fine
particle mode grows mainly by primary source input and coagulation.
Since the coarse particle mode also oscillates in time, both coagula-
tion and source input must be influencing its growth. The coarse
mode never becomes too large because the larger particles are
strongly affected by deposition. Condensation, which reflects the
secondary source input, also affects both modes, but not to as
great an extent as the other processes.
Several changes in the source term can produce variations in
the magnitude of total mass, total number, and light scattering
coefficient, These quantities are increased by raising the valley
flow factor when the air parcel is outside of the city. This causes
an increase in primary source input without changing the basic
source ternparameters derived from PEDCO study emission rates.
Actually varying the emission rate for a particular source would
-------
-129-
i
o
i
ffi
I
+
-*
1
0
1
-H
~H
i
0
iH
X
O
o
ft
C/3
OT
*^
S
1
+
1
1
0
rH
X
e
u
iH
EH"
•H
5-1
4J
CO
•H
13
01
N
•H
CO
rH
0
CO
O
)-(
0)
cfl
UH
O
C
o
•H
4J
3
(—)
O
£>
01
O)
&
H
4-1
£j
0)
T3
C
Ol
ft
QJ
13
fi
•H
CU
B
•H
4J
13
C
Cfl
C
0
•H
4-1
cfl
CO
C
0)
13
G
O
£j
13
C
cfl
fi
0
•H
4-1
•H
Cfl
o
ft
01
T3
£*•»
5-i
*"O
A
CO
01
0
5-1
3
O
•
Cfl
0)
i— :
f^
CO
•H
5-1
cfl
>
4-1
Cfl
60
fi
•rl
4J
^
Cfl
4-1
Cfl
60
C
•H
rH
4-1
4-1
CU
CO
iH
Cfl
C
0
•H
4J
cfl
4-1
•H
>
cfl
t-i
60
TJ
C
Cfl
*\
C
0
•H
4-1
O
0)
>
fi
O
CJ
*\
C
O
•rl
4-1
O
0)
>
<
•
CO
CU
rH
O
cfl
•H
cfl
>
5-1
O
<4-<
"0
01
60
fi
cfl
c~]
U
CO
•H
fi
O
•H
4-)
3
,0
•H
^4
4-J
CO
•H
T3
4-1
^
ft
G
•H
Ol
O
^_j
3
O
CO
•
CN
i-l
PH
PC
4-1
CO
60
C
•H
13
fi
CU
13
C
cfl
o\
p^
ffi
•
CO
cu
o
^_j
3
O
Cfl
4-1
fi
•H
O
ft
*"O
rj
cfl
f^
O
•H
4J
CO
3
f^|
e
o
o
-------
-130-
I
^^~-*^ T
< /
: %>. t
^^SL V
: ^<\,
^
*
• 3
I -ir^—2^^^^^
-Kl^. e!C~
^"^•fc^-^
^^s,
: ]
a
i
O
- ~
'or- ' 1
'_ o 5
!: " • o +
,: i=
- £ ' '
C
S CN
n <^ rH
: ~ ffi EC
i-
J. "o
: "
o £
3D/Od ' [010010/UO
cj
«
*
*
--J£
-- K
-- M
- );
— 3;
-- 3;
-- ii
> -- 3;
1 -- 3E
' -- Sc
> -- )i
-- H
1 I )\
1 1
0 +
1 1
"
r--
1
•H 0
iH iH
i s «
? i-H
e
XI 0
0 "**v-
G iH
O
n
H
co <
co cj
^ CO
S M
O"
C 0 4J
01 01 ca
C G 60
cu oj G
fX PU -H
CU 0) 4-1 M
T3 TJ M O
ti ca M-I
Ci) -H 4-1
S co -x)
•H O> 01
4-) g 60 60
•H _C G
O rH ^
Mn T3 4-1 U
(3 4J
to ca CD co
4-1 CO -H
C G
Q) O rH C
6 -H ca o
O 4J G -H
6 ca o 4J
CO -H 3
13 G -U rd
G 0) CO -rl
r*^ ca *xJ 4-i }-j
G -H 4J
25 CO O > CO
I — t G cj ca *H
Pi O !-i T3
•H 13 60
4-1 G 4->
3 cB -X) 3
m j3 c a
•H -H •> ca c
H G -H
4J O •>
O CO -H G Ol
H -H 4J o a
pL4 *XJ "H "H ^4
CO 4J 3
CU O CJ O
N ex a) co
•H CD >
CO T) G
O •
I rH >, a CM
O !-l rH
•K CO 13 "
0 C Pi
1 S-i •> O EXJ
01 CO T-(
<1" C8 CU 4J 4-J
O O O C3
rH 14-1 S-l 0)
O 3 > 60
X O 13 G
G CO rQ ,£1 C8 CO
W Ol CO cB
M -H -H 0\ 4-1
S Ol M M CO
P ,C CO C3 Pi 3
25 H > > W 13
-------
-131-
33/Od '[OlOtna/UO
I I
o +
I I
CM
O rH
erf erf
w pd
I
O
I
o
o
CO
CO
X
o
u
CO
pq
CM
Pi
K
•K
I
X
O
O
erf
w
i
X
I
oo
a
00
erf
O
c
o
4-1 -H
(3 4J T3
CD O (3
13 CU n)
(3 >
CU C O
ft O
CU O Crf
13 FS
13
CU C 4-1
0 cd cd
•H
4-1 "60
C C
!-4 O -H
O -H 4-1
M-i 4-J 5-J
O Cd
CO CU 4-1
4-> > CO
(3 -0
0) cd 60
€ C
O » -H
B C rH
O 4J
13 -H 4-1
C 4-J CU
cd cd co
CO
CO C r-l
C cu cd
O T3 C
•H C O
•U O -H
3 O 4-J
42 cd
•H "4-1
M C -H
4-J O >
CO -H cd
•H 4-1 H
13 -H 60
CO
cu
N
O CU
CXrH
•H CU
CO
o M cd
CO 13 >
o
S-i * 4-J
CU CO £
cd cu cu
O t3
U-l t-l C •
O 3 CU 00
o ex 43 -H 60
cu cd 4-1 C
•H
4J
0) M 13 13
J2 cd C C
H > ctf 0)
-------
-132-
would also modify the evolution of the particle size distribution.
For example, if we assumed lower emission rates for the fine particle
mode, we would expect the fine particle mode to be smaller and there-
fore, bscat also to be smaller. As we already illustrated in run 6
and run 7, changing the form of the source distribution mean and
variance affects the evolution of the size distribution. Since the
model is rather sensitive to the primary source input, further
investigation of source distributions through field analysis would
certainly help remove some of this ambiguity in the model.
We have studied the variation of bscat with wavelength and
refractive index. These results are shown in Figure 17. It repre-
sents the calculation of bscat at wavelengths 430 nm, 530 nm, and
640 nm for the 9 a.m. size distribution from the 48 hour simulation
run. The wavelengths chosen represent respectively the blue, green
and red portions of the electro-magnetic spectrum . Two refractive
indices for the particles were chosen: (1.5, - O.li) and (1.5-O.Oli).
The fact that bscat is higher for the smaller wavelengths sug-
gests the light scattering by particles can contribute to the
"brown cloud" effect. Although little is known in detail of the
absorptive properties of the particle, an increase in the imaginary
part of the refractive index (assuming that the imaginary part is
not wave-length dependent), of course, lowers bscat and therefore
would, within our assumptions, tend to diminish the importance of
attenuation by scattering to the "brown cloud" effect.
It is possible that there may be several mechanisms contributing
to the "brown cloud" effect. The wavelength dependence of attenuation
by scattering could be one of these. However, this type of wavelength
dependence is sometimes observed in other urban atmospheres24.
Therefore, the full explanation for the marked "brown cloud" effect
in Denver must lie elsewhere. One can speculate that the large mode
of the particle size distribution may exhibit wavelength dependent
absorptive characteristics (the reddish soils of certain regions of
Colorado, including the Denver area, have been noted by many visitors).
Thus, multiple scattering (at episode (high) particle concentrations)
and absorption may play some role in the "brown cloud". More detailed
analyses than performed here may implicate differences in forward and
back scattered light from the "brown cloud". In addition to the
participation by particles, wavelength dependent light absorption
by trace gaseous species could be of importance.
From the various simulation runs reported here (and many others
which we have carried out) for Denver, we arrive at several conclusions.
The ambient particle size distribution is very sensitive to the choice
of parameters for the dry deposition rate and the primary source
input rates; relatively small changes in these parameters results in
values of mass concentration which disagree strongly with field data
during episode conditions. This leads us to believe that our assess-
-------
-133-
'o6
X
^
O
H
O
CO
QQ
.01
' K-,
. ^X
J [_
500 600
WAVELENGTH.NM
Fig. 17 Variation of Bscat with wavelength
Refractive index (1.5, - O.Oli)
Refractive index (1.5, - O.li) —X-
-------
-134-
ment of these parameters does not contain appreciable error. We
also conclude, as a result of our calculations plus various experi-
mental observations^>14,15,23^ that the Denver episode aerosol is
probably source dominated with photochemistry acting only as a
minor secondary source. Thus the submicrometer particle mode grows
mainly by primary source input and coagulation. The large particle
mode is also influenced primarily by primary source input but also
by deposition with some lesser but still appreciable effect of
coagulation with combustion nuclei. Finally, the wavelength
dependent light scattering by the episode aerosol has been shown to
be a possible contributor to the "brown cloud" effect.
As noted earlier, a comprehensive model for the Denver aerosol
is now under development. We believe that such a model can be a
useful tool in the planning of field studies and interpretation of
field study data, In addition, even with out present model (intended
primarily to illustrate the basic features of the comprehensive model)
it should be possible to test various control strategies for the
Denver "brown cloud" and thereby to improve the air quality of this
great city.
-------
-135-
ACKNOWLEDGMENT
This work was supported by a research grant from the Atmospheric
Aerosol Research Section, Chemistry and Physics Laboratory, National
Environmental Research Center, U.S. Environmental Protection Agency.
REFERENCES
1. Crow, L. W. "Airflow Study for Major Air Pollution Days During
November, 1973, Denver, Colorado". (This proceedings).
2, Fiehl, H. and D. Herkjof. "Some Aspects of Denver Air Pollution
Meteorology." J. Applied Meteorology. 11: 1040-1047, 1972.
3. Hidy, G. M. and J. R. Brock, The Dynamics of Aerocolloidal Systems,
Oxford, Pergamon Press, 1970.
4. Drake, R., "General Mathematical Survey of the Coagulation
Equation," in Aerosol Research, (G. M. Hidy and J. R. Brock, Eds).
Oxford, Pergamon Press, 1971.
5. Brock, J. R., Processes, Sources, and Particle Size Distributions.
In: "Fogs and Smokes," The Faraday Division, Chemical Society,
London, 1973,
6. Brock, J. R,, "Condensational Growth of Atmospheric Aerosols,"
J. Colloid and Interface Sci., 39: 32-36, 1972.
7, Middleton, P. B. and J. R. Brock, Simulation of Aerosol Kinetics.
(to be published).
8. Ferman, M. A., R. S. Eisinger, and P. R. Monson, Characterization
of Denver Air Quality. (This proceedings).
9. McNelis, D, N., "Aerosol Formation from Gas-Phase Reaction of
Ozone and Olefin in the Presence of Sulfur Dioxide." Environmental
Protection Agency, Publication number EPA-6501 4-74-034. August,
1974.
10. Sehmel, G. A, and W. H. Hodgson, "Predicted Deposition Velocities".
In: Proceedings of the Atmosphere-Surface Exchange of Particu-
late and Gaseous Pollutants 1974 Symposium, Richland, Washington.
September 4-6, 1974.
11. Amich, R, S., K. Axetell, and D. M. Wells, "Fugative Dust Emission
Inventory Techniques". 67th APCA, Meetings, Denver, Colorado,
June 9-13, 1974.
-------
-136-
12. Shannon, L. J. and P. G. Gorman, "Particulate Pollution System
Study," Vol II, Fine Particles Emissions, MRI Project Publication
Number 3326-6. August 1, 1971.
13. Schutz, L. and R. Jaenicke, "Particle Number and Mass Distributions
above 10~^cm Radius in Sand and Aerosol of the Sahara Desert".
J, Applied Meteorology 13:863-870, December, 1974.
14. Anderson, J. A., D. L. Blumenthal, and G. J. Sem, "Characterization
of Denver's Urban Plume Using an Instrumental Aircraft".
(This Proceedings).
15. Willeke, K., K. T. Whitby, W. E. Clark, and V. A. Marple,
"Size Distributions of Denver Aerosols—A Comparison of Two Sites".
Atmospheric Environment 8:609-633, 1974.
16. Van de Hulst, H. C., Light Scattering by Small Particles. New
York, Wiley and Sons, 1957.
17. Waggoner, A. P. and R. J. Charlson, "Measurements of Aerosol
Optical Properties". (This proceedings).
18. Tuesday, C. S., Ed. Chemical Reactions in Urban Atmospheres,
New York, American Elsevier, 1971.
19. Altshuler, A. P. and J. J. Bufalini, Photochemical Photobiology,
4:97, 1965.
20. Physical and Chemical Nature of Photochemical Oxidants, "Air
Quality Criteria for Photochemical Oxidants". National Air
Pollution Control Administration. Publication Number AP-63,
March, 1970.
21. Bufalini, J. J. and A. P. Altshuler, "The Effect of Temperature
on Photochemical Smog Reactions:. Intern. J. Air Water Pollution,
7(8): 769-771, October, 1963.
22. Alley, F. C. and L. A. Ripperton, "The Effect of Temperature on
Photochemical Oxidant Production in a Bench Scale Reaction
System". J. Air Pollution Control Assoc. 11(19): 581-584,
December, 1961.
23, Durham, J, L., W. E. Wilson, T. G. Ellestad, K. Willeke, and
K. T. Whitby, Comparison of Volume and Mass Distributions
for Denver Aerosols. Atmospheric Environment. (To be published).
24. Severs, R. K. and Peng, J. C., "Multi-Channel Radiometry as a
Unique Air Pollution Measurement", Presented at the Air
Pollution Control Association Specialty Conference on Ambient
Air Measurement, Lakeway, Texas, March, 1975.
-------
-137-
NOMENCLATURE
Italic Letters
A Millikan constant for particle mobility
A. Magnitude of source i distribution
B Millikan constant for particle mobility
b(x,x') Coagulation kernel
bs Light scattering coefficient
C Millikan constant for particle mobility
D. Particle diameter corresponding to mass x.
DM/DLOG(D) Particle mass distribution function
F(t) Valley flow factor
g Acceleration of gravity
H(t) Mixing height
I. Emission flux for source i
x
K Kelvin coefficient
L(t) Wind speed factor
M Total mass concentration
m Particle refractive index
N Total number concentration
n(x,r,t) Density function
Volume averaged density function
Q(X/R,m) Light scattering efficiency factor
R Particle Radius
r Space vector
-------
-138-
S. Mass concentration of j chemical species
J
S. Mass concentration of component j corresponding to
vapor pressure of j over a plane liquid (or solid)
surface
T Temperature
t Time
U(x) Function fit to Sehmel's sedimentation velocity
u Particle mean thermal speed
V. Volume of a vapor molecule of j in the liquid
L (solid) phase
V(x,t) Deposition velocity
V , ,. Gravitational settling velocity
o(x)
v(x,t) Time dependent component of the deposition velocity
W(x) Source distribution function
x Particle mass
x. LnX. is mean for lognormal source distribution i
Greek Letters
ot(x) Condensation coefficient
Volume averaged condensation coefficient
2
3. Input for source i distribution variance (£n g.)
Y Surface tension at gas-particle interface
A Particle diffusion constant
-------
-139-
2
£. Input for initial distribution mode i variance (£n e)
£. Magnitude of initial distribution mode i
A Wavelength of incident light
y Gas viscosity
p Particle density
Volume averaged deposition coefficient
T Time factor for source term
ijj(x) Condensation coefficient
Volume averaged condensation coefficient
Gas mean free path
-------
-141-
X-RAY FLUORESCENCE ANALYSIS OF DENVER AEROSOL
T. G. Dzubay
M. Garneau
0. Durham
R. Patterson
T. Ellestad
J. Durham
Environmental Sciences Research Laboratory
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
ABSTRACT
Aerosol samples were collected in Denver during November, 1973, and
were analyzed to determine which sources might contribute to the Brown
Cloud phenomenon. A set of Low Volume samplers for collecting particles
in the fine (less than 2 microns) and total size ranges was deployed
at six sites, and an automatic device for making 2-hour average measure-
ments was located at a site north of the city. The samples were analyzed
for elemental composition with an x-ray fluorescence spectrometer. Most of
the sulfur, bromine and lead was observed in the fine particles. By
deduction, it was found that the crustal elements such as Al, Si, K, Ca,
Ti and Fe occur mainly in the coarse particles (larger than 2 urn)• The
concentrations of several elements including lead, bromine, silicon,
calcium, and iron and the gas, carbon monoxide, were found to be highly
correlated with each other. Measurements of the light scattering coeffi-
cient were found to correlate best with a linear combination of the lead
and sulfur concentrations. It is concluded that the automobile is a major
source of the lead and bromine and causes crustal elements to be suspended,
whereas the major sulfur sources are non-automotive.
-------
-142-
INTRODUCTION
The scattering of light by atmospheric aerosols may be a major cause
of the brown color of Denver's urban plume. As a means of determining
the sources of these aerosols, a complete morphological classification
as a function of size is needed. In the present work, an important part
of the needed information is provided by x-ray fluorescence analysis of
particulate matter collected on filters. X-ray fluorescence analysis is
suitable for determining elements which have atomic numbers above 12;
these include Al, Si, K, Ca, Ti and Fe, the major metal constituents of
soil dust. It also includes S, whose major source is often the burning
of fossil fuel, Br and Pb from auto exhaust and several trace elements
from a variety of sources.
Since the scattering of light by particles is strongly size-depen-
dent, it is important that the analysis be carried out on size fraction-
ated samples. In the present experiment, aerosol samples were collected,
during November of 1973, in two size ranges, using a Low Volume Sampler
developed by Marple (1). For this device, the filters were changed manu-
ally as frequently as once every eight hours. An additional automated
sampler which changed filters every two hours was also used. The two-
hour sampling duration was short enough for investigation of the correl-
ations between elemental concentrations measured on the filters and the
light scattering coefficients measured with an integrating nephelometer.
PROCEDURE
Size fractionated aerosols were collected with the Low Volume sampler
(1) which is illustrated in Figure 1. Fine particles, with diameter less
than 2 microns, were collected on a filter preceded by a jet impactor.
The impaction surface was coated with silicon grease. Particles of all
sizes were collected on a second filter. Both samplers were oriented
upwards without a rain shield in order to collect with high efficiency
the fine and coarse particles. The elemental concentrations of the
coarse (diameter greater than 2 microns) particles were deduced by taking
the difference between the measurements on the pair of filters. The
coarse particles collected on the impactor were not analyzed.
For the Low Volume Samplers, the flow rates were controlled with
critical flow orifices rated at 14 liters per minute. Millipore Type AA
filters of 37 mm diameter were used in plastic filter holders shown in
Figure 1. After the sampling was complete, the holders were covered and
sent back to the laboratory for analysis.
In order to make a sequence of measurements over two-hour intervals,
an automated sampler was also used. The device operated at a flow rate
of 50 liters per minute through the inner 32 mm diameter of Millipore
Type AA filters. The filters were premounted in 5.1 x 5.1 cm square plas-
tic frames to facilitate automatic manipulation. Although no specific
-------
-143~
TO PUMP
FIGURE 1. Cross-sectional view of Low Volume Sampler
-------
-144-
size fractionators were used, the inlet plumbing caused particles larger
than about 15 urn in diameter to be lost on the walls.
To obtain a representative measurement of Denver aerosol, the Low
Volume Samplers were deployed at six sites located in a north-south line
along the South Platte River. The locations of the sites are shown in
the map on Figure 2 and are specifically indentified as Henderson Fair
Grounds (HF), Trout Farm (TF), Welby (WB), General Motors Assembly Plant
GM, EPA CAMP Station (CP) and Denver Research Institute (DR). The auto-
mated sampler was placed inside the AARS mobile laboratory located at
the Trout Farm site, where several other parameters, including wind
speed, carbon monoxide and light scattering coefficients were measured.
A summary of the measurements which were made from November 6 to 21, 1975,
is given in the Appendix in Tables Al and A4.
After the field sampling was complete, the collected aerosols were
analyzed for elements with atomic numbers above 12 with an energy-dis-
persive x-ray fluorescence spectrometer. In the spectrometer, an x-ray
tube with a tungsten anode was used to excite a secondary fluorescer,
which in turn excited the sample with its nearly monochromatic charac-
teristic x-rays (2). Each sample was excited for 5 minutes using both
titanium and molybdenum secondary fluorescers in order to excite elements
with atomic numbers in the intervals 13-20 and 19-38, respectively.
Lead was also measured using the molybdenum fluorescer. Several of the
samples were also analyzed using a samarium fluorescer in order to ex-
cite elements with atomic numbers between 38 and 57. Measurements of
the spectrum of x-rays from the sample were made with a lithium-drifted
silicon detector. The spectra were analyzed using a stripping procedure
to determine the contribution from each element. A correction for the lead
M x-ray interference with sulfur was made based on the strength of the
measured lead Lg intensity. A small computer was used to control the
spectrometer and to perform the analysis.
The analyzer was calibrated with vacuum-deposited foils for each
element using a procedure developed by Giauque et. al. (3). The
deposits were prepared by Micromatter, Inc., Seattle, Washington, and
they ranged from 50 to 150 yg/cm in mass per unit area. For airborne »
particles uniformly deposited onto a filter, the concentration (inng/cm )
of a specific element was determined by comparing the observed count rate
with that of the known foil of the same element. For light elements,
an additional correction was made for self absorption effects using a
procedure developed by Dzubay and Nelson (4).
In the above manner, the spectrometer was calibrated to measure
aerosol deposits in units of ng/cm . The measurements were later con-
verted to units of ng/m by dividing by the air volume sampled per unit
area of filter. For 8 hours of operation of the Low Volume Sampler at
a flow rate of 14 liters per minute through a deposit area of 9 cm , the
conversion factor is 0.74 m /cm ' For two hours of operation of the au-
tomatic sampler at a flow rate of 50 liters per minute through an area
of 8 cm2, the conversion factor is 0.75 m3/cm . These conversion factors
are based upon the assumption that the deposits are uniform across the filter.
-------
-145-
C
STANDLEY
LAKE
/
CHERRY CREEK
RESERVOIR
FIGURE 2. Map o£ Denver area. The sampling sites are shown as solid
squares (•). Major highways and elevation contours (in feet above
mean sea level) are also shown.
-------
-146-
RESULTS
A tabulation of all of the required measured results is given in the
Appendix in Tables Al and A4. The tabulated elemental concentrations
were deduced from the procedure described above. Whenever the concentra-
tion of an element was smaller than the detection limit (2), the results
were reported as being less than the detection limit as shown in Table
Al. Table Al shows only those elements which frequently occurred at con-
centrations well above the detection limits. The concentrations of P, Cl,
Cr, Co, Ni, Ga, Ge, As, Se, Rb, Pt, Au and Hg were usually at or below
the detection limit. The accuracy for each element is estimated to be
either the detection limit or 15 percent of the concentration, whichever
is larger. Each sample was examined for uniformity of deposit and for
integrity of the filter. In a few instances, the deposits were nonuni-
form or the filters were cracked, and such cases are noted in Table Al.
The elemental concentrations for the 2-hour averaged measurements
at the Trout Farm are given in Table A4. Also given are 2-hour averaged
values of the light scattering coefficient (BSCAT), carbon monoxide (CO)
and wind speed (WSPD) measured by Durham et al (5).
DISCUSSION
There is a striking difference between the composition of the fine
and coarse particles. According to Table Al, at least 75 percent of the
S, Br and Pb occur in fine particles (smaller than 2 pm). On the other
hand, Al, Si, K, Ca, Ti, Mn and Fe occur predominantly in the coarse
particles (larger than 2 ym). This is identical to an effect observed
previously in St. Louis by Dzubay and Stevens (6).
It is suggested that the various elements originate from different
sources. Lead and bromine would most likely come from automobiles as
components of exhaust. The elements such as Al, Si, K, Ca, Ti, Mn and
Fe are expected to come from wind-blown minerals and fly ash. The sul-
fur in the fine particles probably originated as S02 from power genera-
tors, motor vehicles and petroleum refining and was later converted to
sulfate in the atmosphere.
Interelement Ratios
The ratios of measured concentrations for pairs of elements are
given in Table 1. The ratios of Al, Si, K, Ca and Ti to Fe are similar
to the values for the average earth's crust (7) and to the values for
the soil samples from Denver (8), which are also given in Table 1. The
similarity of these ratios suggests that possibly this group of elements
in the Denver aerosol are associated with wind-blown mineral dust. How-
ever, microscopic examination showed that the fly ash content varied
-------
-147-
from several percent to greater than 25 percent, indicating that the
fly ash also has elemental ratios similar to the average earth's crust
(5). In relation to the crustal elements, the relative concentrations of
lead, bromine and sulfur in the aerosol are elevated by large factors.
Automotive emissions are likely the major source of lead and bromine in
the fine aerosol fraction. The measured lead to bromine ratios of 0.26
to 0.30 are typical of values seen in another study (6) and are compara-
ble to the value of 0.39 for PbClBr. Automotive emissions are also the
major source of the carbon monoxide. The presently measured carbon mo-
noxide to lead ratio of 1.2 ppm per yg/m^ is comparable to the value 1.1
ppm per yg/nr which is an average measured at freeway sites in various
parts of the country (9). On the other hand, the automotive emissions
cannot be the major source of the sulfur in the aerosol. The average
measured sulfur concentration of the atmospheric aerosol exceeds that of
lead by a factor of about 3, whereas the sulfur to lead ratio in gaso-
line is typically only 0.5 (10). Moreover, automotive sources would not
have emitted the sulfur as an aerosol, but rather as sulfur dioxide gas
(for the non-catalyst-equipped cars of this 1973 study).
Table 1. RATIOS OF ELEMENTAL CONCENTRATIONS (a)
Elements
All Lo
Vol Sites
Trout Farm
2-hr Data
Earth's
Crust(b)
Denver
Soil(c)
Al:Fe
Si:Fe
S:Fe
K:Fe
Ca:Fe
Ti:Fe
Pb:Fe
S:Pb
Br:Pb
CO:Pb
2.8
5.2
2.9
0.7
0.8
0.11
1.3
2.3
0.26
4.8
5.4
6
0.6
0.8
0.16
2.1
4.1
0.30
1.2 ppm per yg/m
1.6
5.5
0.01
0.52
0.73
0.088
0.0003
2.9
21(c)
0.9
0.014
Measurements below
(a) Averages of the ratios for each measurement.
detection limits are not included.
(b) Ref. (7).
(c) Ref. (8) - The results for Si may be high due to the use of a sili-
con hammer for crushing the sample.
-------
-148-
Interelement Correlations
In order to examine the above suggestion that the various elements
originate from specific sources, correlation coefficients were calculated
for each pair of elements. For the Low Volume Sampler data of Table Al,
only the unfractionated size range was used, and the results are given
in Table A2. For the two-hour averages of Table A4, the correlation coeffi-
cients are given in Table AS. Coefficients of linear regression are given
in Table 2. Plots of pairs of elemental concentrations and linear regression
lines are given in Figure 3.
According to Table 2, there are very high correlations among elements
in the crustal group which includes Al, Si, K, Ca, Ti and Fe. There
are also very high correlations among Pb, Br and carbon monoxide of auto
exhaust. It is important to note that there are also high correlations
between crustal elements and components of auto exhaust. For example,
the correlation coefficients are 0.91, 0.94, 0.94 for the pairs Pb-Fe,
respectively, from the two-hour averages. These high correlations are
not explained by asserting that lead and bromine occur in the same parti-
cles as do the crustal elements. This is clearly not the case, because
the crustal elements occur predominantly within the corase particles,
whereas Pb and Br occur primarily within the fine particles. A more
likely explanation is that combustion aerosol is generated by the burning
of gasoline at the same time that the automobiles are mechanically
generating a road dust aerosol.
Sulfur is very poorly correlated with all of the other measured
species. Copper is also poorly correlated with other measured species
in the aerosol. The only exception is a high correlation between cop-
per and calcium at the CAMP Station site. The copper concentrations
were relatively high at this site, indicating a possible source nearby.
There is a weak negative correlation between wind speed and both
the crustal elements and the components of auto exhaust. The plot of
iron concentration vs wind speed in Figure 3 indicates that the highest
wind speeds are associated with the lowest iron concentrations and
vice versa. Hence, the wind may be serving as a means of ventilation
for pollutant removal. This is contrary to the familiar experience
of high winds introducing soil dust into the atmosphere. However, the
maximum wind speed in the present experiment for which correlations
were computed was only 20 km/hour. According to studies by Newman et
al, the wind becomes an effective dust generator only when the speed
exceeds about 20 km/hour (11).
-------
-149-
Table 2. CORRELATION COEFFICIENTS BETWEEN VARIOUS ELEMENTS IN UN-
FRACTIONATED AEROSOL SAMPLES. For the Low Volume Samplers, the number
of measurements is shown in parentheses.
2 HOUR
LOW VOLUME SAMPLERS AVERAGES
MINERAL GROUP
Si-Al
Si-K
Si-Ca
Si-Ti
Si-Fe
AUTO EXHAUST
Pb-Si
Pb-Fe
Pb-Br
CO-Si
CO-Fe
CO-Br
CO-Pb
SULFUR
S-Si
S-Ca
S-Fe
S-Pb
LIGHT SCATTERING
BSCAT-Si
BSCAT-S
BSCAT-S
BSCAT-Pb
BSCAT-[%(S) + (Pbjj
WIND SPEED
WSPD-Si
WSPD-S
WSPD-Fe
WSPD-Pb
OTHER
Cu-Si
Cu-Ca
Cu-Fe
Cu-Pb
All
Sites (50)
0.94
0.98
0.89
0.92
0.98
0.84
0.83
0.97
0.05
0.36
0.10
-0.02
Camp
Station
0.98
0.99
0.89
0.99
0.98
0.73
0.77
0.94
0.20
0.54
0.28
-0.49
Trout
(5) Farm (11)
0.91
0.99
0.98
0.93
0.98
0.77
0.82
0.94
-0.40
-0.42
-0.46
-0.34
Trout
Farm
0.94
0.98
0.83
0.91
0.96
0.89
0.91
0.97
0.94
0.94
0.95
0.97
-0.10
-0.26
-0.15
-0.11
0.21
0.48
0.24
0.40
0.67
-0.49
-0.07
-0.54
-0.49
0.55
0.83
0.59
0.46
0.78
0.98
0.72
0.30
0.21
0.29
0.30
0.15
-------
-150-
16000
12000
8000
4000
R=0.96
8000
6000
s
4000
2000
R » 0.91
800 1600
Fe, ng/m3
8 16
WIND SPEED, km/hr
2400
24
FIGURE 3. Linear regression pl.ots for various pairs of measured para-
meters. The data represent two-hour averages measured at the Trout
Farm site during November, 1973.
-------
-151-
Light Scattering
Since the scattering of light by aerosols is a major cause of visi-
bility degradation, valuable clues about the Brown Cloud problem can be
gotten by examining the light scattering measurements made with an in-
tegrating nephelometer. Of all the atmospheric particles, the fine par-
ticles in the 0.1 to 2 pm range are expected to have the greatest ef-
fect on light scattering (12). Hence, the sulfur, lead and bromine (as
well as unmeasured soot and carbon compounds in the fine particles) are
expected to correlate best with the light scattering coefficient. Ac-
cording to Table 2, the correlation coefficients are a meager 0.40 and
0.48 for lead and sulfur respectively. When the quantityf%(S) + (PbH
is compared with the light scattering coefficient, a much oetter cor-
relation of 0.67 is obtained. A plot of the above quantity is given in
Figure 4.
From these data it appears that the light scattering is caused both
by auto exhaust, for which lead is a tracer, and by some other source,
for which sulfur is a component. This observation is based on a rather
small number of measurements. Additional measurements would be needed
before a great deal of confidence could be placed on this observation.
CONCLUSIONS
The following conclusions can be drawn from the present analysis
of the Denver aerosol samples collected during November, 1973:
• Most of the S, Br and Pb occur in fine particles. The
elements in the crustal group consisting of Al, Si, K, Ca,
Ti and Fe occur predominantly in the coarse particles.
• For the coarse particles in the crustal group of elements,
the ratios of concentrations between elements are similar
to the values for the average earth's crust.
• The correlations among members of the crustal group are very
high.
• The correlations are very high among the auto exhaust com-
ponents such as Br, Pb and carbon monoxide.
• There is a poor correlation between sulfur and all other
measured species.
• The light scattering coefficient correlates fairly well with
the quantity Ps(S) + .(Pbj), indicating that the sources of
both the sulfur and the lead contribute to visibility de-
gradation.
• Additional measurements at more than one site using an in-
tegrating nephelometer, a carbon monoxide analyzer and an
automatic sampler are needed before the above conclusions
have adequate statistical validity for establishing policy
for control of pollutants.
-------
-152-
o
C/5
5000
[S] + [Pb] , ng/m3
10,000
FIGURE 4. Light scattering coefficient vs. sum of lead and
concentrations for two-hour averages at the Trout Farm site.
of sulfur
-------
-153-
REFERENCES
1. Marple, V.A. Minnesota Fine Particle Sampler. In: University of
Minnesota Particle Technology Laboratory Publication No. 228. 1974
p. 40-41.
2. Goulding, F.S. and J.M. Jaklevic. X-Ray Fluorescence Spectrometer
for Airborne Particulate Monitoring. EPA Publication EPA-R2-73-182.
April, 1973. 70p.
3. Giauque, R.D., F.S. Goulding, J.M. Jaklevic, R.H. Pehl. Trace
Element Analysis with Semiconductor Detector X-Ray Spectrometer. Anal.
Chem. 45:671-681, 1973.
4. Dzubay, T.G. and R.O. Nelson. Self Absorption Corrections for X-Ray
Fluorescence Analysis of Aerosols. In: Advances in X-Ray Analysis, Vol.
18, W.L. Pickles, C.S. Barrett, J.B. Newkirk and C.O. Ruud (eds.). New
York, Plenum Press, 1975. p. 619-631.
5. Durham, J.L., R.K. Patterson, T. Ellestad and W.E. Wilson. Aerosol
Characterization of Denver's Urban Plume, 1973. This Symposium Report.
6. Dzubay, T.G. and R.K. Stevens. Ambient Air Analysis with Dichoto-
mous Sampler and X-Ray Fluorescence Spectrometer. Environ. Sci. Tech.
9: 663-668, 1975.
7. Mason,B. Principles of Geochemistry. John Wiley and Sons, 1952.
8. Battelle Columbus Laboratory. Unpublished data. See further dis-
cussion in Reference 5.
9. Colucci,J.M., C.R. Begeman and K. Kumler. General Motors Corp.,
Warren, Mich. Research Publication GMR-773. June, 1968.
10. Shelton, E.M. Motor Gasolines, Winter 1973-74. U.S. Dept. of
the Interior, Bureau of Mines, Bartlesville, OK. Petroleum Products
Survey No. 85. June, 1974.
11. Newman, J.E., M.D. Abel, P.R. Harrison and K.R. Yost. Wind as
Related to Critical Flushing Speed vs Reflotation Speed by High Volume
Sampler Particulate Loading. (To be published).
12. Faxvog, F.R. Optical Scatter Per Unit Mass of Single Particles.
Appl. Optics 14: 228-229, February, 1975.
-------
-154-
APPENDIX
Correlation coefficients were calculated for all pairs of measured
concentrations according to the expression
Z (x - x) (y - y)
Coefficients of linear regression for a best fit to the data with the
linear equation
y = A + Bx
were calculated according to the expressions
& - y
~
.
2 2
£( x - X )
B =
2 - 2
Z (x - x Z)
In the above summations the subscripts i have oeen omitted;
x. and y. are individual measured concentrations, and
x and y are average values for the data sets (x.) and(y.)
respectively.
FOOTNOTES TO TABLES Al - A6
Units
Elemental Concentration: ng/m
Wind speed (WSPD): km/hr
Light Scattering Coefficient (BSCAT): 10 m
Carbon monoxide (CO): parts per million
Size Parameter
Size 0: all sizes; no fractionation
Size 1: fine particles; less than 2 ym
-------
-155-
Note Parameter
Note 0:
Note I:
Note 2:
Note 3:
highly nonuniform deposits, qualitative results
only
cracked filter
good quality deposits
fine and total size range labels seem to be inter-
changed or incorrect
Using the data sets in tables Al and A4, correlation coefficients are
given in Tables A2 and A5, and coefficients of linear regression are
given in Tables A3 and A6 respectively. Of the Low Volume Sampler
measurements, only the data with size range code 0 and note code 2 were
used in the calculation of correlation coefficients.
-------
-156-
o rx. oooi -o o
c»4 CM ^ f^ o* co
CM CM CMJ co CM -oo
ICM CM. CM CM c* —in -*i m — rxcM co
V V V — '
-• o- co
in o CM O- o
) TS» f) rM
— tsj _
O — 3-
r^. —• —« cO co 3- T- 3-
V V V
-o— -*
— in v o-v
0*01 r*jo*
vvv
vvvv
vvv
vvv-. vv
o
2
z
o -«
OO OO OO
OO P-O OO
in o 3- -. CM
oo -oo ~o o oo oo
i/»o oo oo
OO -000 -O* O1J1 OO OO OO f>*f>4 OO OO OO OO — r»-
i/i 1/1 -o —• tN« —• r<4Oi jiji uiin 1/1 di CO O
— o fn -o
-O "1 O- CO
O* O3 IN O-
V V V
V V V V
V VV VV
V VV
CM CM CM
CM CM CM i
CMCM CM CM CM CM CMfM
i CM CM CM CM CM
CM CM CM o CM
a.
u
a.
5
y
~$
O
•^
Q
_^
<1
0 0
• o
t- O
-*• ~c
— —
^ 0
r o
~t r&
h- O
LJ
in
0
o
o
•£
—
0
o
CO
o
CO
3-
0
o
o
3-
CM
ro
n
X
-0
o
rt
0
3-
CM
o
o
•4
CO
0* 0
o o
0 0
0 O
X CO
o o
o o
o o
0 0
0 0
0 O'
1^
o
0
0
-O
—
o
o
ao
o
o»
00
0
o
o
-o
—
0
0
CO
0
O"
if\ ~O
o o
0 O
0 0
3- 3"
- i^.
-* —
O 0
CO CO
0 0
O 0
o —
0 0
O 0
CD CO
o o
o o
O 0
O 0
O 0
f» a
— fM
O 0
0 0
X X
o o
o o
0 0
0 0
o o
-* y
- -
0 0
0 0
-o -o
— —
0 0
0 O
00 X
a o
rv o-
0 0
0 O
0 O
cc cc
0 O
0 0
0 O
0 O
a o
-
CM
o
<••»
CM
«*»
o
o
•o
fM
«
.-)
*•>
CM
-------
-157-
i* •* n* ot Q ^ •• *•• •• » »• o» ^ »• #• »• » -* o»- -• -• » »• oo o o
-
V V VVV V V
Of*. f» O
ta. «• v « m
QC
Ul
a.
z -
V V V
V —V V -» — V V -• V V V V •— V
-*-. V VN -* V V -* V V V V V — V
v v -• — -• -. -,
V V V V V V V V V V VV
£ in o -• o -* o** (M (M no 010 i/t oo oo oo oo oo oo oo ao oo jitn oo
3*- o n o oo oo oo oo ft*** oo oo oo oo o o oo oo oo ut m — — oo
Z f O> V *O *O y 9" -« -O tOCO -» -* O> - t-
-------
-158-
3" — O in 3T 01 I
o *• c> o f*
-O in Jl -• 3
mO* r» -O Pv.
04 — CM *>4
»• 3 o- o*
-O > *-x. -si
fM O CO -O
— — 04
03 W-n 30- ^O 3£ '
— 3- m a- CMO- •* *• **_ °1
•N 04 Jl <"» CO -•
O * >
3" m o» a> 04 CM
f» v -o 04 PS. m
m oa-
I 04 *« ^
P*.P* 03
vvvv
04 - co - •* «•
ul '
U
z
3
— V —
VV —
V _ V
* 3
04 O-
x m
v v
— o
— X
— V
V V -»V —
— — Ps. X 3 i
00 « X — * tM V
vv»ma-in
vvvv
in 3 o
a- co PS.
in v
o- o o
3- 04 —
— -. V
mm f-s ui -•
I
O »•
*» tn
o o
00*
PS. -O
O O
ji»n oo «>* » O3
in in rsjo4 »nin — — —
mm x» mm en p*.r-
0404 3O 0«04 O <-»*•>
O O
— -* OO — —
OO O O in J»
m <"» mm **J 04
co «o *O -Q mm
33 —-* *N(M
3O
33
« o>
33
3O O3
<*» «*l 33
3-3- «o x
040J 33
O3
O3
O3
33
3OO
O <•»<•*
33 -**Mf
3 O 33
33 33
*
-------
-159-
z
o
3- rx
co — 01
>o co
CO
in
m
01
CO
o
o
o
ce
CO
-o
Ul
rx
o
CN
rx
a-
(M
in
CO
fv.
m
CO
rx.
rx
O
«*»
CO O
ON O
rx o
UJ
NI
»-•
en
o
3-
3-
m
I
>o
ON
-0
Ul
01
-O
-O
3"
O
o
CO
CN
Ul
01
CO
:D -»
u -o
Ul ON
Ul *•»
a
01
u>
ao
Ul
Ul
CN
in
o
o
o
-o
•o
«n co
o ui
o
»
o
U)
Ul Ul
U. ON
Ul
Oi
01
Ul
ON
O -•
Z 3- -O
r rx -o
co
in
o
rx.
Ul O
«•> o
rx a
o un
o ft
o rx,
CN
Ul
•O
ON
Ul
0| CO
ix. fx.
rx rx
<-> rx.
rx. tn
CO
OJ
CO
o
o
m -» —
tn ON 3-
3- CO CN
O
3
O
OJ
in
3- m
ui —
co
rx.
Ul
rx.
in
H-
z
Ul
Ul 01
01 ON
ON CO
CO CN
ui «n
O GO
o —
ON
«n
x
CN
CO
Ul it
CN
CO
Ul
-O
01
01
O* O
«•> OJ
ON m
01
CN
•O
3-
-O
O
O
CO
5- 01
01 01
— I
I
Ul
oc
ir
o
CM
— 3-
«n ON
o
o
o
o
o
o
-* 01
Ul CO
CN
-• in
CO O
01 ON
01 m — in oj
ON 01 -o rx m
00 ON -O Cr- Ul
-• o rx
CN CN «n
-o rx co
ui ix,
01 <•»
o
3-
rv
tn rx
m —
3-
rx
~O 3-
Ul —
rx co
Ul
_l
en
M m
«n
•- z
Ul
oc.
CO
ca
o.
-------
-160-
z
o
z
o
U.
u
O
2.
cc
o
u.
»n
t—
in
z
•X
z
o
(A
in
IU
a:
(M
3- —
— in
ro CM
O 0
in —
rN. o*
co —
3- —
0 O
in O
m —
— oj
ro —
0 O
O CM
ro o
CO »•»
3- 0
O O
(M CM
O -O
ro CM
ro o
o o
+ «•
rN. r
in ro
in 3-
o
*•
o o
o o
o —
_l
•<
3- —
o o
ro o
O 3"
(M "»
3- 01
0 0
0 o
m CM
(M —
3- «M
0 0
-.CO
at o
<•» <•»
3" (M
o a
> 3-
-O 00
in ro
ro —
0 0
O CO
ao in
IN. 3-
3- ro
0 0
3- O
0> 0
CM —
3- CM
0 0
3- (M
Ul O
— 3-
3- —
a o
O» (M
a ro
•o 3-
ro —
c o
-O (M
«•> o
-O PN
3- O
0 O
in u>
IN. o
OO _
O
+
0 0
o o
o —
ro -.
0 0
* •»•
IN. m
f» a
— CM
»
in
3" —
0 0
IN, ,0
OfN.
*» 3-
1
3" O
O O
O —
3- <•>
ro o
t
3" O
o o
O CO
0 0
<•> IN.
1
3" CM
00
O CO
(M -•
3- 0
O 0
CO -O
in -•
CM CM
3" CM
O O
fs. f*)
r<- ^>
JO —
O O
— 3*
> CM
rv ao
3- o
o o
om
in -•
MOO
3- 0
O 0
o««>
30 CO
— m
o
O 0
O 0
o -.
3" —
o o
•f 1
in rv
rs. 3-
CM CM
3- o
O o
* +
-o —
in o
-•«•»
in
»•> o
00
IM -.
» r
-»
o o
GO -•
^o in
3- -•
m —
o o
in f>
•0 ».
-o rt
•n —
o o
CO —
CMr*.
— in
CM O
o o
* +
m rv
m m
3- —
1
— o
0 0
+ 4
O —
IN. rv
CM CN
*
—
O 0
••> m
>0 <0
in m
<*> (M
o o
<*> 3-
CO O
fN. -.
•n o
O 0
«•) CM
00>
CM ao
i
r«> fM
0 0
3- -0
CM —
— CM
<»» —
0 0
rv o
— in
CM CO
i
o
0 0
0 0
o —
CM —
0 0
O PI
O
0 0
* +
<•> 3-
^ CO
(M —
1
CO O
m r*.
(M O
o o
-O 3-
tM CM
CO CM
tM O
a o
<•> 3-
OD -O
0. «
••» —
O O
o —
3- O
•^ ~^
— 0
o o
3" -^
CM 0
00 —
CM —
0 0
«•» 3-
ji in
3- CM
O
0 O
O O
o —
(M O
O O
— 3-
-O O
r —
(M O
O O
in
OIN.
1*1 in
3-
-O CO
CM CM
CM CM
O 0
-0 3-
m —
in c»>
CM IM
o o
4 1
fN. >0
0 «*>
CM 3-
IM —
0 0
* I
in rN
IM a
z
s:
<«» o
0 0
CO 3-
-O 3-
CM fN.
n —
o o
in 3-
i> in
ro (M
<*» —
o o
fN. CM
i^l 3-
i/> rN.
3- —
0 0
o —
oco
— 00
0
0 O
0 O
o —
<•» CM
0 0
0 0
«•» 3-
!•» CM
<*» —
O O
CO CO
oo-
— 00
<«> o
o o
3- in
CM CO
3- «•
CM —
o o
o m
f 3-
3- IN,
1
CM -.
O O
+ 1
CMX»
3- fN,
IM 0
00
in o
3" fN,
— CM
O
00
00
o —
o< o
00
0* —
»• tN.
CM — •
oo
-O CM
— 0
-.00
oo
0 3-
•o m
CM CM
— o
0 O
<•» o
— IN.
CO
•« f*l
r*> —
(M O
O O
OCO
on
OO CM
f\ -.
0 0
3-0
4) IN.
— in
m —
o o
0-0
1 O
» CM
—
0 0
+ *
•o in
oao
in ao
3" -•
O 0
3- 00
CM in
— « fN.
ro O
O O
o CM
00 CN
IMO
fO CM
O O
3- 0"
•O (M
in
-------
-161-
O »• -4)
O O O ;
-• •* O-i
3
•
Q
04
fs. ''»»« 04 r»- O -4» in
a
a.
04
>
o o o o o OOOOOOOOOOOOQOOOOOOOOOOOOOOOOOO
• •••• **»»•••«*»»•*•**»»•**•*•******•
o 04 r* co 4> o4it>4o4<^3"-o(>«'^3">O3t'Oinr«»3"«^ino-*'1^'-t'^1*>iniA-oji-«inf^o
co
a.
.^ — 04
(*• Off* O O O »• — *»•
-* o* r^ -* oj o O O* *•
en ?>icoincoinoo»in
-• T r*l -* t)
*n o
04 04
>o r- a> <*> -• o < -o
*•«> —• o in in •* <^
04 (^ 04 -* — 04
»• -^
o* T O
O" -O 3*
in 3- o
O* ff' O O O O «•» — f-*
in -o in ^ o co u> *n ui
-^ o — or*. ^*n«n*o
04 04 — -, — -.
2- O ^ 3- 3- CO 3- -O — J> •*> r*. (O
z c^ r^i f*> (^ oj /•) x i/i r^ X^.(M
f*> fs. r^ —• o-
t» r^ 3- ^i o
3" O
O4 —
r>| —
»-» O •« 3)
r- o in 04
3- in -o r*.
-•r- -o 3- -* m -o
f*« O4 O •« 3- O < 04 rs. ^> rv. co in
r*. — in CD a> r-.
»• m **» — 04 to
in 04 *o in o -o
in oj »*» a> i
> J» O » O -O -O O* J1 (—__
i o * o \fi
•o —
v> r->
in s in -o **i
**> T ji i/i fr
n 04 14 T —
y — rs.
4) -O CO
O O eo
— 04 m 3~ Pi««in'^3'x
*nrs.o4-o Q 3- — r*. -o
l/tO4^Oin O- —
z < »*io> o-c-or*.a>f--»' ^ o —•
3-r- —• i*» o —-jirsiDO- x-«
O O Qin -O 04
o r^ rv 3- o o
3000OOOO
j v- ooooooo-
«I Z OOOOOQ^'
U OCCO—' — — —'—OJ04OOOO—-*-* — -.0-^— — — "
^ ^* aooooooooooooooooooooooooo;
« oooooo oo oa aooopaooapppooopii
OOOOOOO OOOOOOOOOOOOOTOTIOOO OOOOO
ooooo oooooooooooo oooooo o oo ooooo
*^in*ii>*i^'^ji^'inf*"<^-*^inri»^-— •^•j^^»'O»>—lA'^inr-' "> —• rt LTI r**
oooo——— —ooo — oooo— — — — — 0404000 o — — — —
aoooooboooooaoaboob ob boooo
o
o
X >•
^ fv T
ooooooooooooooooooooooooooooooooooooooooooaooooooooooo
L U. U.U.U.U.U.U.U. WU.U.U.U. U.lkl
-------
-162-
z.
o
t~~
•s.
^
3
H-
C»
•4
or
u.
UJ
">1
*-^
»n
o
2
O
i_»
>- Px
< rv
O —
in
CO
O <"»»
0. co
st in
-» •
.» •
ro
CD -O
a. co
3-
ac in
CD CO
•o
O
r* -o co
in
rO
9-
r*.
ft
fM
—
3"
"Jl
1
o r^
in o
— o-
0 CM
O» 3-
CO
m
<•»
ji
3-
rv
~o
>0
!N
"1
3"
3-
-O
O
CM
>O
-0 tM
o in
00 O-
SO CO
— ji
ro f»
-o in
m 3-
3- 3"
(•
•
m «o
CM r^
«o o-
o* a
— o
m o
-+
CM
r*
t>
r*>
O
3-
CO
CO
3"
|
O
0
o
-«
«n
rv
o»
CO
eo
3"
1
at
00
i
0
o
a
CO
00
3"
1
in
3-
3-
1
3-
in
rvl
in
o>
UJ
3" CO
Z 30 O
N 3- in
o —
o in
in
CO
3"
I
>o
o
ao
at
o
X
3- in
* ro 3-
•jj in
~o
CO
O
— »
co
3"
rs.
in
o —
m -o
— — CO CM
in in r-» -o
_ o ao co
I
in
O
o
o
o
(M
CO
O
• O
o o
_l O <*>
•* o o-
a
0
o
"••
O CO
o* r*.
1 o>
_.
X)
fM
1
O
•*»
CO
.-3
fv
_
t
r>.
o
t>
>
CO
(SJ
a
o-
CM
_
in
— ••
t
o
-o
O"
o
.^
CO
in
3-
m
CM
r>.
CM
1
CO
O
m
M
CO
i
l
f^
CO
-o
o
•— •
1
t
o
ao
r^
•a
l
tf-
CO
3"
CO
r^
3-
OO
O
CM
o
ao
co
-o
o
CO
=r
<•>
«
3-
CO
3-
3"
in
CO
«*>
-o
CO
n i^
CO IN.
in —
i
— in
in
-Z. U)
r u.
QC
CD
CD
0-
a i-
OL <
tn u
x vt
T)
II
*
-------
-163-
ooa-OMMMooo—a—o -•
OO OO OO OO OO OO OO O
»«»»»»»»»*»»»« »
Mcoo-cDin— —OCD«—j>M>v,cc»3
M»c0— QDo>--oaDinrvaDa'Oooo
a-— —JlflM—M—-Ofv-OMMCa —
•••»••••••••••••
I I
a- M oMa*r*iM-vi'*i'«i'n'*t—o — — o
>- O O O o O 3 OO OOOOOOOOOOOO O O O
««»»+»»» » + 4444 44444444 444
v» o a- -o f< a> -o in o •>>!•».—to iflc» — o r^ a- M — a- M a- a- oo o e»
01 a- — a- m rv co —««> rv-o—rxco— <•> — »> a- o 3- M a- M a-Ma-o.oMi'Ma-o — — — — CD
QOOOOOO OO OOOOOOOOOO O O O O O
— a. »»»»»» 44 4444444444 «»l«*
^ I/I -O ?• a- 01 a- ~o o CD 3- o «•! m -o—a~ —f» o o co a- oo — — - -••... . . ................
4 ill I i i i I I it
z a* o f> — a- o o. o «•» o *>i —• MM o o a* o M — 04 o — -»<••> — *n — M
O 00 30 OO OO OO 00 00 C3 O 00 00 00 O OO 00 OO
- 00 »» +» 44 44 44 41 * I 44 44 +1 »» » 4 t »l II
t- a. o- «•» (Ma- _,„. ox M ,03- -o— <«> rv rv f> — — -o CD rv 00 oo a- »• ma- o-o
vj -o ». go <>. ,0 — rv » o-o -o M a-o > M Jiao MO M a> oo a- in M o> o a-
< — »• rv — o o — — a- — 00 M in 3- — o — co a* a- oo o o— a* »• Ma- o--*
<*• i ill
a-— a--. a-o o o o o o — MM o o a--. M — — o — — M — M or*
u oo oo oo oo oo ao oo oo oo oo o oo oo oo oo
1X1 a-a- o<»i -O coo i»o
(/i o^ >->r». a--o a>r» o>.n oo^f~ -.mrs.-. r»!>- ooi>a- o—a-— a-.-N
(M*N -*a* m *o <*)ri Ln«*i —in i/iin r^rv —tsi -oco o— MM O»M M-* M»*»
o •• •• •• «• •• .. •• •• ». •• .. •« •• •• ••
Z I I
— a- —• a-M a~ —t (•>-» r»»-» rno (MO r> —• a- —• —• en— n —• -«M -^M —• —
OC3 00 00 00 OO 00 00 00 00 O 00 00 00 00 00
m ^ ++ ++ ++ -f+ +-f +^ «+ -4+ ++ + +4 ++ +1 +i it
uj NI r^a- in —• Oin 3*in i^rv - a-p^ xtj ^ng
•« tMto IM— a-o- n— nn -^m o— M>n rxffl r.fv Mtv —tv
X •> • . . . . . . • . . . . . . . . . . » . . . • « . • ••
LU I II
> a-o a~o a-— <*io r^,o t>a- — -o —• y oo a-a- -ox -or^ M&- oa* a-in
o —o -«(O — -. . •• . .
i i
o a-— ri_ a-— o M— M— — a-— MO <">o <">— MM — <•» —M
* oo oo oo oo oo oo oo o oo oo oo oo oo oo oo
t- Ul * + * + 44 * * +4 41 41 4 44 44 44 44 41 4t 41
a:
o
ti.
(A
in
l/l
(A
or
a:
o
X
*
k—
(J
it
in
T
i/i
•x
— M
a- —
o o
rv m
a- —
a- M
o o
4 4
— in
Fv M
1
3- —
O O
0- O
— -0
— 10 c
o —
4 4
O X
rv m
* "
y —
O O
4 1
3- .n
O O
3" O
o o
4 4
-O O
— 3-
r
o
4
O O
Q O
o in
i
a- M
O O
0 M
a- M
0 0
4 4
0- rv
-O M
in oo
i
o —
0 0
T Jt
CD in
3- rv
,
O — -•
0 O
4 4
-I" O
3- CO
0 0-
1
3" —
0 0
4 1
r o
0- 0
0 -O
1
o
4
O O
0 0
CD —
a- —
o o
4 4
a- o
CO CO
£r
i
4- -»
0 0
CO l/>
3- (N
O 0
+ -f-
00
3- —
1
3- —
O CD
co rv
-n N
i
_
^>
4
O O
3 O
a- o
0 0
4 4
CM —
o —
i
a- o
CD O
•4 4
rv CM
— -0
O CD
0"- in
o- in
rv -o
O CD
o o
0- -0
OK M
-• >0
_
o
4
o cr
"D CD
*
CM —
o o
4 1
a- 3-
03 rv
•o o-
M O
0 0
4 4
M CC
— • rv
— -o
0 CD
0 CM
M Jl
3 in
0 0
M O
1
— *
0
CD O
CD 3
3 —
O O
O O
4 4
— O-
O 'M
•NO
O —
O CD
4 |
-o a-
3- -v
a :M
1
o — •
CD O
4 |
•N O
-o —
CM 0-
O O
0 0
4 4
CO 00
3- m
rv ~o
•D CD
•a ji
l
o
o o
CD O
CD —
M —
CD 0
rv oo
V. -V
O "Si
0 O
4 1
in o
T —
— ">
1
CM O-
i/l — •
-O -•
M »N
CD ^
4 1
0 -0
J\ 0
I- tl
fM >vj
O O
4 1
rv c*-
-0 —
3- 0
CM O
CD 0
4 1
0 —
•o o
0 0
o a
0 —
D O
•O CM
•v 3"
0 M
O "D
0 0
-O X
-O — •
1
—• — •
0 0
3- X
•O — '
o -~
CD 0
4 4
CM Jl
D O
M —
O — •
0 CD
4 1
o cr
o o
rv •N
1
o o
CD C?
4 4
O rv
— in
~~* ~°
O CD
0 0
4 4
O 0
CM rv
a- —
— rw
a- M
•DO
— CD
CO 0
O M
0 O
> 0
M o
M3-
1
0 0
CD 0
r M
0 O
M -O
M3-
|
M —
O O
O CD
CD —
"v — • '
1
M —
O 0
4 4
o rv
:O O
-n -•
o —
o o
4 |
in o
•N D
m —
l
CM O
0 0
4 4
O CD
rv ,0
rv —t
1
o a
0 0
4 4
O -0
O 0
o 01
M M
3- — •
CD 0
— M
— m
3- M
O O
M O
1
— —
0 0
m in
3" 0-
^ VJ
o — *
CD a
4 4
CM rv
rv o
J) 0
3- — *
o a
4 1
rv M
-0 -0
— 0
1
M CD
O O
4 +
a o
0- —
0 3-
0 O
CD 0
4 4
3- O
— in
O M
1
"M M
O O
-O =0
M 3-
— i.n
1
— 3-
0 0
4 1
0- 3-
-O M
T> O
|
CM O
a o
4 1
M CO
o rv
— 3-
1
— O
0 0
4 1
-0 -O
CM O
M CO
— M
o o
-O M.
-o -.
3- -* i
1
— <"}
0 0
4 1
oj a-
3- -o
M •M
~* **?
0 0
4 |
3- rv
in N
M —
— O
0 0
4 |
rv oo
CO —
-o a-
— • a-
1
-^ m
0 0
4 |
-o a-
in in
— Pv
1
0_ — _ CMO rv— — — -0 — -OH — M —o
a. «
in u
Jt in
o
-------
-165-
AN ANALYSIS OF PARTICIPATES FROM THE DENVER URBAN PLUME
USING SCANNING ELECTRON MICROSCOPY AND ENERGY
DISPERSIVE X-RAY SPECTROMETRY
Philip A. Russell
Clayton 0. Ruud
Denver Research Institute
University of Denver
Denver, Colorado 80210
ABSTRACT
Millipore substrates and backup filter of an Anderson 8-stage im-
pactor operated at the Trout Farm sampling site on 16 November 1973
were intensively examined using scanning electron microscopy and X-ray
energy dispersive spectroscopy. Impactor stages from the Trout Farm
sampling site for 14, 17 and 21 November 1973 and Adams County Fair-
ground sampling site 1^, 16 and 21 November 1973 were also examined.
It was determined that there was little or no significant difference in
particle size distribution across the substrates or backup filter of the
Anderson impactors. However, there was large variation of particle
sizes on any stage and the pattern of size deposition on progressive
stages was species dependent. Large mineral particles were predominant
in the initial stages of the impactor while auto emission products and
sulfur were observed mainly on the last stages. Flyash was observed in
all stages. Most flyash produced X-ray fluorescence spectra very similar
to major mineral components of Denver's plume. Particles greater than
one micrometer made up the great buli of the particulates samples; they
were almost entirely mineral with clay, potassium aluminum silicates and
quartz predominating. Particles less than one micrometer were predom-
inantly auto emission and mineral or mineral-like material. The esti-
mated concentration of particles greater than one micrometer was 1.55 x
10°/m3, less than one micrometer was 5-89 x 10°/m3. Sulfur was pre-
dominantly associated with the smallest particulates, auto emission;
its source, however, was not automotive.
-------
-166-
INTRODUCTION
A major portion of urban air pollution is composed of particulates
which are products of (l) natural erosion entrained by wind or volcanic
activities (fugitive dust), (2) natural erosion entrained by man's
activities (automotive, farming, building and industrial activities),
(3) combustive processes which are primarily associated with power plants
and heavy industry, and (4) auto emissions. The last two categories
contribute metals to the atmosphere in concentrations greater than the
crustal composition (van Valin 197 ^ and Duce 1975); this contribution is
observed world wide and is assumed to contribute about 2.5$ to earth's
total aerosol mass, although the contribution can be as great as ^OQffo
of naturally produced aerosols near the source of production (Research/
Development, News Notes August 197^0•
Knowledge of size distribution and composition of particulates that
are in the environment of men is a very important health concern because
the lungs act as a selective filter, capturing particles less than 1 Mm
in diameter (Natusch 197^)• This is particularly important as some toxic
metals and organic compounds are found in relatively high concentrations
in respirable particles (Loh 197^, Natusch 197U, and Davison 197*0 . In
the lungs, extraction of toxic material can be as high as 60$ to 80$
(Davison 197*0 • Even relatively inert materials may be harmful if they
are deposited in the lungs, i.e., freshly dry-crushed quartz (Hanusiak
1975)' Particles less than 1.0 Urn are also responsible for visible light
scattering which contributes very significantly to reduction in visi-
bility (O'Brien 19?U).
The Environmental Protection Agency invited the Denver Research
Institute, University of Denver, to participate in the 1973 Field Study
of the Denver "Brown Cloud" and, as a part of its contribution, to use
scanning electron microscopy (SEM) with energy dispersive X-ray speetro-
metry (EDS) analysis to characterize particulates collected on an
Anderson 8-stage impact sampler. The SEM-EDS study was to (l) obtain an
estimate of the particulate composition of the Denver plume. (2) evaluate
the need for a fine particulate ambient air standard and (3) determine
the effectiveness of the 8-stage Anderson impactor using Millipore sub-
strates .
METHODS AND MATERIALS
The 0.8 Mm Millipore backup filter and stage substrates from the
Anderson 8-stage sampler operated on 16 November 1973 a* the Trout Farm
sampling location were selected for intensive examination to (l) obtain
particulate information of the Denver Urban Plume and (2) determine the
effectiveness of the Anderson impactor using Millipore substrate.
Pie-shaped segments (approximately 20°) were cut from the backup
filter with each of the eight samples including portions from the center
-------
-167-
to outer edge; the sections were attached to an appropriately sized piece
of clean Plexiglass which was attached to an SEM stub. The attached sub-
strate was then vacuum coated with approximately 200A of carbon.
Particulates in outer, middle and center portions of each stage and
the backup filter were examined in a scanning electron microscope for
morphology and chemically characterized using an energy dispersive X-ray
spectrometer. All examinations were conducted using a 30 KeV accelerat-
ing potential, ^5° specimen tilt angle and 15° specimen rotation.
For the backup filter and stage 0, 1 and 2 particles were dispersed
enough on the substrate to allow relatively easy in situ examination.
The procedure was to examine a portion of each segment at 1000X or 2000X
magnification and take a photograph which was subsequently divided into
quadrants. These were then examined thoroughly at higher magnifications.
Photographs were taken at high enough magnifications to allow size mea-
surements and morphological observations. Size was measured from the
photograph and converted to effective diameter in urn. For non-spherical
particles this was derived from the mean of two measurements taken at
right angles to a line through the particle's longest axis. Each
particle was X-ray analyzed for 100,000 total counts, all elemental peaks
were quantified for integrated counts and the spectrum plotted for future
reference. An optimum sample size of 30 particles was selected for each
portion of a segment; this figure was usually exceeded.
Stages 3~7 presented a problem as particle distribution became pro-
gressively more dense, producing a spot of concentrated material. Ex-
tensive efforts were made to representatively redistribute the particles
on an inert substrate; satisfactory results were not obtained. (We
later concluded that the substrate filters were probably composed of
Millipore Celatate which is impervious to most solvents). It was assumed
that particulates observed outside but near the compact spots would be
representative of particulates impacted on each substrate. Observations
of gross spot X-ray analyses suggested this was true, except for stages
6 and 7« The X-ray analysis procedure was also modified to permit more
rapid chemical classification of particles. The collection of X-ray
information from a particle was continued until the strongest peaks
were clearly evident (20,000 to 50,000 total counts) and then elements
were recorded as major, moderate or trace. Major compositional elements
were those that had an integrated peak height greater than background
including peaks that were > 50$> of the strongest peak's height; moderate
elements also had a well-defined emission peak but one that was less than
50$ the height of the strongest element present; trace elements had
poorly defined peaks with heights only slightly above background. The
charts produced for the backup filter and impact substrates 0-2 were
also examined in this manner so that their particulate data could be
compared with the later stages. With the help of a particle atlas
(McCrone 1973) and mineral/chemical standards prepared in our labora-
tory, particles were assigned to the various categories (Table l).
Distinctive morphology was also noted and used as an aid in determining
particle species.
-------
-168-
Table 1
Particle Classification, Energy Dispersive Spectrometry (30 KeV),
for the Denver Urban Plume Study
Elemental Composition
Type
Mineral
Quartz
Aluminum silicates
Clay
Potassium
Calcium
Iron
Calcium
Calcium/sulfur
Rutile
Flyash
Si-rich
Ca-rich
Metal-rich
Lead Rich
Lead bromochloride
Lead
Metals
Salts
Carbon
Particle
Auto emission
Spheres
Major
Si
Si
Si
Si
Si,(Fe)
Ca
Ca,S
Ti
Si
Ca,(Si)
(Pb),(Fe),(Si)
Pb
Pb
Various
Various
Moderate
(Fe)1
Al,(K,Ca),(Fe)
Al,K,(Fe)
Al,Ca,(Fe)
Al,Fe
Al,K,Ca,(Fe).(Ti)
Cl,Br
( ) - sometimes present
-------
-169-
For comparative purposes, total impact spots from various stages
and samples were examined using EDS.| For each examination the SEM-EDS
parameters were identical to those
tion. Stages 3 through 7 were examined for the Trout Farm (Nov.
2l) and Adams County Fairground
The impaction media for the Trout F:
while that of the Adams County Fairground was Huclepore filter.
RESULTS
I. Observation of Particle Size Di
f the detailed particulate examina-
1.6,
(Nov. Ik, 1.6, 2l) sampling locations,
rm samples was Millipore filter,
crimination by the 16 November Trout
Farm Anderson 8-Stage Impactor I
sing Millipore Substrates and Backup
Filter
The effective mean diameters,
classification for each stage and
Appendix Tables 1-27; summaries of f
Appendix Tables 28-1+3. The diamete:
for the various impactor stages and
Figures 1 and 2, the vertical lines
all the specified particle types obs
particle compaction and the lack of
was not examined for mineral or flyg
(this will be discussed later).
hemical composition and particle
radial location are presented in
article sizes are presented in
s of mineral and flyash particles
the backup filter are presented in
display the standard deviation of
erved on any one stage. Because of
distinguishable particles, stage 7
sh particles near the impact spot
The variance of diameters of
stage was examined by comparing the
mineral particles on a stage with pj
and center locations on the impact
ence between a stage's specified pa
mean diameter at the center, middle
5/o significance level: (l) flyash i
stage 6 and (2) mineral particles f:
minerals and flyash collected on each
mean diameter of all flyash or
psrticles sampled from outer, middle
edia. Only twice was there a differ-
ticle type mean diameter and the
or outer portion when tested at the
articles from the outer portion of
om the center portion of stage 2.
II. Farticulate Composition of the
Ifo November 1973
Jrban Plume at the Trout Farm,
For the backup filter and impac
relatively simple. Figure 3 illust:
particles and a lead particle; the i
torn are typical mineral type partic!
The two types of flyash in the figu
of flyash encountered: (l) the Si-]
tinguishable from mineral clay exce]
the Ca-rich type which characterist
X-ray peak with other variable elemc ntal
auto emission). Figures Ua and b a:
cles found on various stages of the
t stages 0-5 the examination was
ates two different types of flyash
ssociation of particles at the bot-
es, in this example all were clay.
e are representative of the majority
ich type which is virtually indis-
t by spherical morphology; and (2)
cally exhibited a strong calcium
peaks (note association with
examples of auto emission parti-
Anderson impactor. Figure kc is an
-------
-170-
t-t
i oaopn) -
-------
-171-
o
•H
H
-p
CJ
o>
ft
w
X
<
o
•H
-P
fl
O
to
^
I
I
-------
-172-
J
•H
o
O
-P
O
$
w
i
CO
a
o
en
-p
3
O
?H
EH
I
§
o
•rl
O
•P
3
-------
-173-
example of an EDS analysis of a particle like 4a; in many cases no
characteristic elemental X-ray peaks were observable. In the last three
stages (5-7) independent particles and particles embedded in auto
emission produced emission lines for lead, chlorine and bromine. These
particles were assumed to be lead bromochloride. Figure 5 compares
standards of lead bromide (PbBrg) and lead chloride (PbClg) with a par-
ticle of lead bromochloride found on an impactor substrate. The parti-
cle produced a fluorescence pattern that was most similar to the PbClg
pattern.
Appendix Table 44 presents the X-ray spectroscopy of the various
impact spots for stages 3~7 > Figure 6 illustrates the X-ray spectra from
which these data were derived (the spectra from stage 3 is omitted be-
cause it is very similar to that of stage 4). The Al/Br fluorescence
line can be produced by aluminum K or bromine L emission and the S/Pb
fluorescence line can be produced by sulfur Ka or lead VL emission.
When the data are reduced (Appendix Tables 45, 46, and 48a), it is ob-
vious that the relative amount of silicon contai ning material is pro-
gressively reduced from stage 3-7 while the amount of lead and/or sulfur
is increased. By observing the lead L intensity it is apparent that
lead is most prevalent in stage 5 while sulfur becomes a major component
in stages 6 and 7 (Figure 7)• It is noteworthy that while individual
particles of lead and lead bromochloride were observed, no particles
containing primarily sulfur were detected.
By comparing the fluorescence patterns of stage 6 (Appendix Table 48a)
with the particles observed (Appendix Tables 3^ and 42), it was obvious
that the particles around the compact impact spots for the later stages
were not representative of the spots themselves. Close examination of
spots on stages 6 and 7 revealed that the spots were made up of two
distinct regions; the inner area of the middle spot of Stages 6 and 7
exhibited stronger emission for the elements of S, Pb, Br and Zn (Appen-
dix Tables 45 and 46). Particles in the inner area had a mean diameter
of 0.28 - 0.15 Mm and were rounded but not spherical in shape, while the
particles in the outer area were 0.08 - 0.03 Mm in mean diameter with a
morphology similar to carbon black samples which have been examined pre-
viously by DRI's microscopy laboratory.
Using the silicon emission peak's percent of total integrated counts
for each spot (Appendix Table 48a) and comparing it with the volume-
percent of minerals for stages 3, 4 and 5 (Appendix Table 5l)> a linear
regression was observed with a correlation coefficient of 0.97- From
this regression the percent-volume mineral composition of stages 6 and
7 was estimated at 68.6 and 64.4 percent, respectively.
III. Estimate of Particulate Size Distribution
An estimate of the number of particles collected on the Anderson
-------
-174-
OJ
H
O
O
H
fi
O
CJ
^
FQ
O
J-i
PQ
LTN
U
-------
-175-
t-
-p
o
?H
EH
O
-P
O
n5
•H
cd
•P
w
i
CO
r-
GJ
•H
0)
bO
si
-p
CO
ID
tD
03
-p
cn
TJ
(U
-P
O
I
•H
f-
I
cd
•P
w
0\
O O
o
o
w
o o>
in -P
P -H
O CO
ft bQ
to fi
Ki
s
g,
£1
-------
-176-
-p
•H
a
c
§
H
cd
•p
HI
H
H
30
20
10
X - Relative intensity of PbLa emission
O - Ratios of emissions at approximately
2.30 KeV (S K and/or PbM^ emission)
and PbLa,
• - Ratios of emissions of PbLa and BrKa
-250
^-2.00
-150
•p
•H
C!
M
£
- 100
.50
Observed ratio of PbMa and
PbLa emissions in sampled
particulate PbBrCl
PbLa/BrKa
Stage
Fig. 7. Comparison "by stage of the Trout Farm Anderson impactor,, l6 Nov.
1973> for sulfur and lead X-ray emissions.
-------
-177-
impactor was made using the data collected: (l) the average mean dia-
meter for particles on each stage came frour microscopy (Appendix Tables
28-^3); it was assumed that the mineral particles on stages 6 and 7
averaged 0.28 urn and all auto emission particles were 0.08 nm; (2) the
average particle volume was estimated by simply cubing the average dia-
meter; (3) the density of minerals (and flyash) was assumed to be 2-7
and auto emission at 2.3 (2.7 is the average density of clay and 2.3
that of carbon black); (4) the percentage composition, by volume, was
previously estimated (Appendix Table 5l)j and (5) the mass loading in-
formation provided by EPA (Table 2).
Table 2
Mass Loading of Trout Farm 8-Stage Anderson Impactor,
16 November 1973
Stage
A
(M-gm)
325
216
234
366
236
111
53
65
ino
mass
do]
(16.1)
(10.7)
(11.6)
(18.2)
(11.7)
( 5-5)
( 2.6)
( 3.2)
(20.3)
p .
concentration
\ M-gni/m /
Q.k
5.6
6.0
9.4
6.1
2.9
I.h
1-7
10.6
0
1
2
3
U
5
6
7
Backup
Total 2016 (100.0) 51.9
Using the relationship
n = r^~ (PC)
Vpd
where "w" is the mass loading of an impactor stage in grams, "V^" is
average particle volume in mm^, "d" is the density in grams/mm^*and
"pc" is the volume percentage composition of mineral and auto emission
calculated to be on any one stage. Then "n" becomes the number of
particles of a selected category on a specific stage. For all stages,
there were approximately 6.00 x 10^ particles averaging > 1.0 urn in dia-
meter and 2.29 x 1010 particles < 1.0 urn; this gives an atmospheric
concentration for the sampling period of 16 November at the Trout Farm
sampling location of 1.55 x 106 particles/m3 > 1.0 urn and 5«89 x 109
particles/np < 1.0 Mia.
-------
-178-
IV. EDS Survey of Anderson 8-stage Impactor Stages 3-7 for Various
Locations and Dates
EDS X-ray data for the Trout Farm Anderson impactor using Millipore
substrates for l4, 16, 17 and 21 November are presented in Appendix
Tables hja, kQa, %)a and 50a; similar data for Adams County Fairground
using Nuclepore substrates for lk, l6 and 21 November are presented in
Appendix Tables V/b, ^8b, and 50b. Figure 8 presents the relative per-
cent intensities of silicon, potassium, sulfur/lead and lead emission
lines.
DISCUSSION
I. Anderson Evaluation
From the information derived from the 16 November Anderson 8-stage
impactor it is clear that there is little or no significant difference
in particle size distribution across the surface of the Millipore sub-
strates or backup filter; X-ray analyses suggest the same conclusion.
It is also apparent that while the impactor does separate particles by
size, large variation exists on each stage. It is also apparent that
the average size of flyash and mineral particles on the same stage is
significantly different. This might occur because the flyash is hollow;
X-ray analysis and limited transmission electron microscopy suggest that
the flyash particles are solid, from which we conclude that the differ-
ential separation of flyash and mineral particles is probably due to
morphology or agglomeration.
X-ray analyses and particle inventories show a significant separa-
tion of particle types by composition: mineral particles (or particles
with a chemical composition similar to minerals) were a major component
by mass on all stages, overwhelmingly so on stages 0-3; flyash were ob-
vious on all sta3cs hut mostly on stages 3~5j lead-containing compounds
were most concentrated on stage 5 and sulfur was concentrated in stages
6 and 7.
II. Particulate Composition of the Urban Flume at the Trout Farm,
16 November 1973
By volume and mass, the major species of particles examined are
minerals > 1.0 urn in diameter, of which the most prevalent are clay,
potassium aluminum silicates and quartz in relative abundance. Minerals
also make up the major portion of the fluorescing material of particles
-------
-179-
•d
o
•p
g
bo
-------
-180-
with some flyash and zinc particles, and vas diffusely associated with
auto emission. Sulfur predominance in the last impact stage suggests
that it is associated with the smallest particles discriminated by the
Anderson impactor.
III. Particle Size Distribution
Although it was estimated that the number of particles < 1.0 Urn was
approximately 3 orders of magnitude greater than particles > 1.0 Mm, the
majority of smaller particles were auto emission particles which were
observed in agglomerations on the various stages. Because of the nature
of the Anderson impactor it is not possible to determine the size of
these agglomerations on the various stages. Because of the nature of
the Anderson impactor it is not possible to determine the size of these
agglomerations prior to sampling and thus the Anderson sample will not
allow determination of the atmospheric particulate frequency--particular-
ly in an aged plume where agglomerates are more prevalent (Willeke 197^0 •
IV. EDS Survey of Anderson 8-Stage Impactor Stages for Various Dates,
Locations and Substrate Materials
The results of the EDS survey suggest that both the Trout Farm and
Adams County Fairground sampling locations experienced similar environ-
ments and fractioned particles into similar size groups. Lead and lead
bromochloride particles were concentrated in stages 5 and 6 in similar
amounts for the lU, 16, 17 and 21 sampling dates. Sulfur was always
concentrated in stages 6 and 7 an
-------
-181-
emlssion particulates and associated sulfur onto the last two
Anderson impact stages. However, the spots produced during impac-
tion make direct observation and identification of particulates
difficult, particularly on stages 6 and 7.
h. Morphological differences between flyash and mineral particles allow
differential deposition of particles of flyash and minerals on the
same stages. Flyash is not found in any significant amount on any
stage where its presence would not be predominated by mineral parti-
cles .
5. Because of the chemical composition of flyash and its non-preferen-
tial impaction on any stage, X-ray fluorescence analysis for flyash
is considered improbable.
6. The observation that many of the small particulates < 1.0 Hm are
those produced by automotive and combustive sources indicates the
need for fine particulate air standards to assess the cause of pro-
duction. Present methods will not allow simultaneous size and
chemical composition determination.
ACKNOWLEDGEMENTS
This research was supported by the Environmental Protection Agency
through grant R-802889j a limited portion of the results presented were
funded by the Environmental Protection Agency through contract 5-02-
4l08 6 and 7 of the
Anderson impactor. Examination of Millipore LoVol Filters exposed
successively for short periods (2 hours) indicates that auto emission
and flyash occurrence very dramatically increases during southerly air
mass movement up the Platte River basin.
-------
-182-
REFEKENCES
Davison, R. L., D. F. S. Natusch, J. R. Wallace and C. A. Evans. 197*1-•
Trace Elements in Flyash: Dependence on Particle Size. Environmental
Science and Technology 8:13, 1107-1113.
Duce, R. A., G. L. Hoffman and W. H. Zoller. 1975. Atmospheric Trace
Metals at Remote Northern and Southern Hemisphere Sites: Pollution vs.
Natural? Science 187:59-61.
Hanusiak, W. M. and E. W. White. 1975- SEM Cathode Luminescence for
Characterization of Damaged and Undamaged a-Quartz in Respirable Dusts.
In Scanning Electron Microscopy/1975; PP- 125-132 edited by Johari, 0.
IIT Research Institute, Chicago, 111.
Loh, A., A. H. Miguel, D. F. S. Natusch and J. R. Wallace. 197^.
Preferential Concentration of Toxic Species on Small Airborne Particu-
lates. Air Pollution Control Association Pub. 7^-201.
McCrone, W. C. and J. G. Delly. 1973. The Particle Atlas: Vol. Ill,
The Electron Microscopy Atlas. Ann Arbor Science Pub. Inc., Ann Arbor,
Michigan.
Natusch, D. F. S., J. R. Wallace and C. A. Evens, Jr. 197^. Toxic
Trace Elements: Preferential Concentration in Respirable Particles.
Science, 183:202-204.
O'Brien, R. J., S. R. Holmes, R. J. Reynolds, J. W. Remzy and A. H.
Bockien. 197^- Analysis of Photochemical Aerosols in the Los Angeles
Basin According to Particle Size. Air Pollution Control Association
Pub. 74-155.
Van Valin, C. C., R. F. Pueschel, P. Perungon and R. A. Proulx. 197^.
Anthropogenic Contribution to Meteorologically Important Aerosols. Air
Pollution Control Association Pub. 7^-268.
Willeke, K., K. T. Whitby, W. E. Clark and V. A. Marple. 1973- Size
Distributions of Denver Aerosols—A Comparison of Two Sites. Particle
Tech. Lab. Pub. No. 195. Particle Technical Laboratory, University of
Minnesota, Minneapolis, Minnesota.
-------
-183-
APEENDIX TABLES
-------
184
Vaole 1. Particles observed on the outer portion of stage 0 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Ko .
1
2
3
U
5
6
7
8
y
10
11
12
13
lU
15
16
17
18
19
20
21
22
23
21*
25
26
27
28
29
30
31
32
33
31*
35
36
37
38
39
1*0
1*1
1*2
1*3
1*1*
Particle
Effective Diameter
(pm)
5-1*
6.2
1.1*
2.8
2.6
1.9
3-3
2-5
l.i*
2.8
1.6
1.3
1.1
l.i*
1.0
0.7
0.5
0.3
0-5
1.8
1.1
0.3
0.5
8.1
5-6
3-2
2.7
6.5
7-0
12.5
lt.0
2.0
1.8
3-9
3-8
17-0
U.I
U-3
U.9
5-1*
7-3
9-2
6.5
3-9
Elements
lla ior Moderate Trace
Si,.'.l K,Ca
Si -- Mg
Si,Al K,Ca,Fe Ti
^
>
Si Al,Fe K,Ca,S/Pb,Cl
Si
Si A1,K
Si Al,Fe K,Ca,Ti
Si,Fe K,A1 MgCl
Si Al,K,Fe Ti,Mg
Si,Al K,Fe
Si Al K,Oa,Fe,Mg
S,Ca Si, 01 Mg,Al,K
Si Al,K,Ca,Fe
Al,Si,Cl
Si,Al Ca K
Si,Al -- S/Pb,K,Ca,Fe
Si Al
Si -- Fe
Si -- K,Ca,Fe
Si Fe,Al K,Ca
Si Al,Fe,Ca K,Ti
Si K,A1
Si Al,K,Ca,Fe Ti,S/Pb,Cl
--
Particle Type
Comments
clay
quartz
flyash
conglomerate
clay
quartz
aluminum silicate,
potassium
clay
aluminum silicate,
iron
aluminum silicate,
potassium
aluminum silicate,
potassium
flyash (Si rich)
clay
auto emission
aluminum silicate,
calcium
aluminum silicate, clay
aluminum silicate, clay
quartz
quartz
aluminum silicate,
iron
clay
aluminum silicate,
potassium
clay
auto emission
probably gypsum
-------
185
Appendix Table
Particles observed on the middle portion of stage 0 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 19 '3
Particle
HO.
1
It
5
6
7
a
9
10
11
12
13
lit
15
16
17
18
19
20
21
22
23
2k
25
26
27
28
29
30
31
32
33
3k
Effective Diameter
_ (mi) _
10.1
7.6
10.2
2-9
2.3
2.2
i.r
6.6
9. it
5-5
1.2
1.0
2. It
1-7
2.5
1-7
1.7
1-7
l.lt
5.0
3.0
10-9
1-lt
It. 8
3-6
3-7
Ik. 7
6.0
7.0
8.2
ll.lt
5-1
Elements
Major
Si,Al
Si
Si
Si
Si
Si
Si
Si
Si,Al
Ca
--
Si,Al
Si,Al
Si,Al
Sl,Al,Ca
Si,Al
Si,Al
Si,Al
Si,Al
Ca
Si,Al
Si
Si
Si
Si,Al
Si
Si
Si,Al
Si,Al
Si
Si
Si
Si
Moderate
K,Ca,Fe
Al,Ca
Al,Ca
__
A1,K
Al
Al
Al
K,Fe
Si
Ca,Si
—
--
Ca
Ti,Fe
Fe
__
Ca,Fe
Ca,Fe
S1,P
Fe
A1,K
Al
A1,K
—
Al
Al
—
—
--
Al
Al
—
Trace
Pb/3,Cl,K,Fe
--
__
Fe
K,Fe
Fe
K,Ca,Fe
Mg
..
Al,Mg,Fe,Pb/S
Ca,Ti,Fe
K,Ca,Fe
Fe
__
K,Ca,Ti,S
Ca,Fe,Ti,K,
Pb/S
K,Ti
Ti
Fe,Al
K,Ca
--
K
—
K,Fe
K,Ca,Fe
—
K,Ca,Fe
Fe,K,Ca,Cu
Al
K,Mg,Fe,Co
K,Ca,Fe
Al,Ca,Fe
Particle Type
flyash (Si rich)
aluminum silicate,
calcium
aluminum silicate,
calcium
quartz
aluminum silicate,
potassium
clay
clay
clay
aluminum silicate,
potassium
mineral
aluminum silicate,
calcium
flyash (Si rich)
flyash (Si rich)
aluminum silicate,
calcium
flyash (Ca rich)
flyash (si rich)
flyash (si rich)
aluminum silicate,
calcium
flyash (Ca rich)
flyash (Ca rich)
auto emission
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
aluminum silicate,
clay
clay
clay
clay
flyash (Si rich)
quartz
clay
clay
quartz
Comments
probably calcite
conglomerate
associated w/#23
associated w/#23
-------
186
Appendix _aole
3. Parzicles observed on the center portion of stage
Trout Farm location, November 13, 1973
0 of the Andersen 8-stage impactor from the
Particle
Affective Diameter
ilo. (urn)
1 7-7
2 10.7
3 5-9
l* 8.8
5 6.7
6 7-1
7 11.1
8 6.2
9 9-6
10 3-1*
11 7-8
12 10.5
13 6.2
ll* 1*.5
15 6.2
16 7-5
17 9-5
18 7-3
19 5-7
20 1*.0
21 9-8
22 2.0
23 0.6
21* 2.5
25 8.7
26 6.0
27 5-0
28 3-1*
29 1-1
30 2.2
31 2.1
32 1-5
33 2.6
31* 2.2
35 5->*
36 3-6
37 2.U
38 13-2
39 5-5
1*0 6.8
Major
Si
Si
Si,Al,K
Si
Si,Al
Si
Si
Si
Si,Al
Si
Si
Si
Si,Al
Si,Al,Ca,S
Si
Si
Si
Si,K
Si
Ca
Si,S,Ca
Si
Si
Si
Si
Si
Si
Ca,Al,Si,P
Ca,Al,Si
Si
Ca,Si
Si,Ca,Al
Si,Al,Ca
Si
Fe
Al,Ca
Fe
Si
Si
Si
Elements
Moderate
Al,K,Ca,Fe
Al
--
A1,K
K
Al,Ca,K,Fe
--
Al
K,Fe
Al,K,Ca,Fe
—
Al,S,K,Ca,Fe
K
Fe,Ti
Al,Ca,K
Al,K,Ca
Al,K,Fe
Al
Al
Si
Al
Al,K,Ca,Fe
Al,Ca,Fe
Al,K,Ca,Fe
—
Al,K,Ca,Fe
—
Ti,Fe
P,Fe
Al.Ca
Al,P,Fe
Fe
Fe,Ti
—
__
Fe
—
Al
--
Al,K,Fe
Trace
__
K,Ca,Fe
Fe
Fe
Fe
—
Al
K,Ca,Fe,Mg
Ca
Pb/S,Ti
Al,Mg,K,Fe
Cl,Ti,Mg
Fe
Mg,P,V,K
Mg.Fe
Fe
Ca,Ti
—
—
Al,Mg,K,Fe
Mg,P,K,Fe,Zn
S,T1
K,S
Pb/S
--
Mg
Al,K,Fe
--
S,Ti
S,K,Fe
Mg,S,Ti
Ti,S
Mg
Mg,Al,Ca,Fe
Si,P
Mg,Si,Ti
Si
K,Ca,Fe
Al,K,Fe
Mg,Ti
Particle Type
clay
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
aluminum silicate,
potassium
flyash (Si rich)
quartz
clay
clay
flyash (Si rich)
quartz
clay
aluminum silicate,
potassium
flyash (Ca rich)
clay
clay
clay
aluminum silicate,
potassium
clay
mineral (Ca)
mineral (Ca/S)
flyash (Si rich)
flyash (Si rich)
clay
quartz
clay
quartz
flyash (Ca rich)
flyash (Ca rich)
clay
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
quartz
iron oxide
flyash (Ca rich)
iron oxide
clay
quartz
aluminum silicate,
potassium
Comments
probably calcite
conglomerate
convoluted surface
-------
187
Appendix Taole It.
Particles observed on the outer portion of stage
Trout Farm location, November 13, 1973
of the Andersen 6-stage impactor from the
Particle
Ho.
1
2
3
It
5
6
7
10
11
12
13
lit
15
16
17
18
19
20
21
22
23
2lt
25
26
27
28
29
30
31
32
33
31*
35
36
37
38
Diameter
5.0
8.0
2.5
6.8
7-0
2-3
5-3
lt.lt
3-9
2.6
3.0
1.6
1.8
o'.B
1.0
8.0
16.0
12.5
8-5
6.0
7-5
5-5
3-0
11.5
it .8
7-3
3-7
5-1
2.8
7.0
5-5
1-7
7-9
1-3
2.7
Major
Si
Al,Si,Fe,Ca
Si
—
Si
__
Si
Si
Si,Ca
Si
Ca,Si
Si
Ca,Si
Si
Si
Si
Si
S1,A1
Si
Si
Si
Si
Ca
Si
Si
Si
Si
Si
Si.Ca
Si
Si
Si
Si
Si
—
--
--
Elements
Moderate
Al,K,Ca,Fe
S
Al,K,Ca,Fe
--
--
__
Al
Al,Ca
Al
Al,K,Cu,Fe
—
Al,K,Ca,Fe
A1,K
--
Al
Al,Fe
A1,K
K
Al,K,Ca,Fe
A1,K
Al,K,Ca,Fe
Al,K,Ca,Fe,S
Si
—
Al,S,K,Ca,Fe
Al,Fe
—
Al,K,Ca,Fe
Al,K,Fe
—
Al,Fe
Al
A1,K
A1,K
Si,Al
—
—
Trace
__
P,Ti
Ti,Pb/S,Cl
Al,Si
--
Al,Si,Pb/S.Cl
Mg,Al,Si,S/Pb
K,Ca
Fe,K,Cl,S
K,Fe,Mg,Cl
--
Al,Fe
S
Mg
Ca
S,K,Ca,Fe
Mg,S,K,Ca
Ca,S,Fe
Fe
--
Fe,Co
--
Cl
Al,Fe
Al,K,Ca,Fe,Mg
—
Cl,K,Ca,Ti
—
--
—
Al,K,Ca,Fe
K,Ca
K,Ca,Fe
Fe
Fe
P,S,C1,K,
Ca,Fe
Si,Ca,Al,Si,
S,Fe
Si,Al,Cu,S,
Mg
Particle Type
clay
aluminum silicate,
calcium and iron
clay
organic
quartz
organic
auto emission
clay
aluminum silicate,
calcium
aluminum silicate,
calcium
clay
calcium mineral
clay
calcium mineral
quartz
clay
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
clay
clay
clay
clay
calcium mineral
quartz
clay
clfly
quartz
clay
aluminum silicate,
clay
clay
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
clay
organic
organic
Comments
probably calcite
probably calcite
probably calcite
-------
188
Appendix Table 5. Particles observed on the middle portionof stage 1 of the Andersen 8-stage irapactor from the
Trout Farm location, November 13, 1973
No.
1
2
3
U
5
6
7
8
9
10
11
12
13
11*
15
16
17
18
19
20
21
22
2k
25
26
27
28
29
30
31
32
33
3U
35
36
37
38
39
Uo
1*1
1*2
Particle
Effective Diameter
(ma)
7-7
8.3
5-7
5-6
k.2
7-0
2.3
2.1*
3-1
3-1
12.8
1*.2
1-7
1.2
3-6
2.3
1.6
1-7
2.2
U.I
12.8
3-7
1-5
U.8
5-1
6.8
6-3
2.6
U.5
3-3
5-U
2.8
11.7
2.9
3-U
5-2
8.1
5-6
6.6
9-7
Major
Si
Si
Si
Ca
--
Si
Si
Si
Si
Si
Si
Si
Si
Ca
Ca,Si
Ca,Si
Si,Al
Ca,Fe
Si
Ca,Fe
Si,Ca
S,Ca
Si,Al
Si,Ca,Fe
__
Si
Si
Si,K
—
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Elements
Moderate
Al,Fe
—
Al,K,Ca,Fe
—
—
A1,K
Al
A1,K
Al
A1,K
Al
Al,K,Fe
Al,S,K,Ca,
Fe
Al,Si,S,Fe
Al,Mg,Fe,S
Al,S,Fe
Ca,Fe
Si,Al,S
Al,Ca,Fe
Al,Si,3
Al,S,Fe
_-
__
Al
Ca,S
Al,K,Ca,Fe
A1,K
Al.Fe
Si.Fe
—
—
Al,K,Ca,Fe
K,A1
—
Al,S,K,Ca,Fe
Al
—
Al,S,K,Ca,Fe
—
Al,K,Fe
K,A1
Al,K,Ca,Fe
Particle Type
Trace
K,Ca
Al,K,Fe
—
Al,Fe,Si
Al,P,Cl,Ca
Pb/S,Ca,Fe
K,Ca
Ca,Fe
K,Ca,Fe
Fe
K,Ca,Fe
Ca
P,Pb/S,Cl,
Ti
P,Ti
P,Ti
Pb/S,K
__
Pb/5,K,Cl
K,Ti
P,K
Si
Cl,K,Fe
Pb/S,Cl,K
Al,Si
Pb/S
Cl,Ca,Fe
P
Al,K,Ca
Al
Al,K,Ca,F
Pb/S
Fe,Ti
K
Pb/S
Mg
Al
Ti
Al,K,Ca,Fe
Mg
--
clay
quartz
clay
calcium mineral
organic
clay
clay
clay
city
aluminum silicate ,
clay
clay
clay
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Ca rich)
flyash (Ca rich)
salt
clay
aluminum silicate,
calcium and iron
salt
clay
aluminum silicate,
aluminum silicate,
clay
quartz
clay
clay
aluminum silicate,
quartz
clay
quartz
quartz
clay
quartz
aluminum silicate ,
aluminum silicate,
clay
potassium
potassium
potassium
potassium
potassium
potassium
Comments
probably calcite
conglomerate
-------
189
Appendix Jable 6. Particles observed on the center portion of stage 1
Trout Farm location, November 13, 1973
the Andersen 8-stage irapactor from the
Ho.
5
6
7
8
9
10
11
12
13
15
16
17
18
19
20
21
22
23
2k
25
26
27
29
30
31
32
33
34
35
36
37
40
43
44
1*5
1(6
U7
1(8
49
50
51
52
53
5k
55
56
57
Particle
Sffectiv Diameter
(wO
39-4
6.3
10.0
14.1
3-1
12.9
l(.l
14. 1
5-5
2.2
3-1
4-5
4.1*
2-3
2.9
2.7
12.7
9.4
7-8
10.0
9.7
3-1
1.4
2.7
1.3
1.8
1-7
2-7
3-3
4.8
2.9
3.4
0.6
1.6
9.6
3-6
3-9
6.4
5-7
7-9
2.2
5-5
3-3
2.7
0.9
7-2
3-4
2.1
6.4
1.6
9-1
7-5
3.0
7.6
7.8
12.3
0-9
Major
..
Si
Si,Fe
--
Si
Si
Si
--
Si
Al,Si
A1,S1
Al,Si
Al,Si
Al,Si
Si
--
--
Al,Si
Si
Al,Si
Al,Si
Al,Si,S,Ca
Al,Si
Al,Si,Fe
--
--
Al,Si
Si
Si
Al,Si
Si
Al,Si,Ca
Al,Si
Al.Si
Al,Si
A1,S1
Si,Al,Ca
A1.S1
Si,Fe
Si
Si
Si
Si
Si
--
Si
Si
Si
Si
Si
Si
Si
--
Si
Si
Si
Si
Elements
Moderate
..
Al,K,Fe
Al
--
Fe
Mg
Al,K,Fe
--
--
--
K,Ca,Fe,S
K,Fe
Fe
K,Fe
Al,K,Ca,Fe
--
--
K,Ca,Fe
--
Fe
Fe,K,Ca
Fe
K,Ca,Fe
—
Al,Si,Fe
--
K
Al,K,Ca,Fe
Al
K,Ca,Fe
--
--
--
Ca,Fe
Fe
--
Fe
Fe
A1,K
Al,K,Ca,Fe
A1,K
A1,K
A1,K
Al
Si.Al
Al,Fe
--
Al,Fe
A1,K
Al
Al,K,Fe
Al,K,Ca
—
Al,K,Fe
Al,K,Fe
S,K,Ca,Fe
Al,Fe,K
Trace
Pb/S
Ca,Ti
K,Ca,Mg
Al,Si,S,Ca,Fe
K,Ca,Pb/S
Fe
Ti
Si, Pb/S
Fe
Mg,Pb/S,Cl,
K,Ca,Fe
Cl
Ca,Pb/S,Cl
K,Ca
Oa,Fb/S
—
Si, Pb/S
Si, Pb/S
Pb/S
Fe
K,Ca,Pb/S,Cl
Cl
K,Ti
Cl.Mg
K,Ca,Ti,Mg
--
Si
--
Pb/S
Fe
S,C1
--
--
K,Cu,Fe
--
K,Ca,Pb/S
K,Ca,Fe,Pb/S
K,Fb/S,Mg
K,Cu,Mg
Ti,Pb/S,Mg
Ti,Pb/S
Fe
--
—
Mg,Pb/S,Fe
K
Ca
Al,K,Fe
K,Ca
—
P,Pb/S,K,
Ca,Fe,Cl
Ca
Fe
--
—
Ca,Ti
--
Ca,Cl
Particle Type
organic
clay
aluminum silicate, iron
organic
clay
magnesium silicate
aluminum silicate,
potassium
organic
quartz
clay
clay
aluminum silicate,
potassium
clay
aluminum silicate,
potassium
clay
organic
organic
clay
quartz
aluminum silicate, iron
clay
mineral
clay
aluminum silicate, iron
clay
carbon
aluminum silicate,
potassium
clay
clay
clay
quartz
aluminum silicate,
calcium
flyash (Si rich)
flyash (Si rich)
clay
clay
aluminum silicate,
calcium
clay
aluminum silicate,
iron
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
aluminum silicate,
potassium
clay
clay
clay
quartz
clay
aluminum silicate,
potassium
clay
clay
clay
organic
aluminum silicate,
potassium
clay
clay
aluminum silicate,
potassium
Comments
probably serpentine
possibly gypsum associated
with an aluminum silicate
angular shape
-------
190
Appendix Table 6. Particles observed on the center portion of stage 1 of the Andersen 8-stage impactor from the
(Con't) Trout Farm location, November 13, 1973
Particle
Elements
Effective Diameter
No.
58
59
60
61
62
63
64
65
66
67
68
69
Appendix
(wn)
[Deleted]
6-3
1.6
3-9
0.7
2.6
2-3
3-2
0.8
7-5
4.6
4.1
Major
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Table 7. Particles observed
Trout
Particle
Farm location
Moderate
A1,K
A1,K
Al
Al,Fe
Al
Al,Fe
Al
Al,K,Fe
Al,Fe
Al,K,Fe
Trace
Fe,Mg
Al,Fe
Mg,Fe
K,Cu
Fe
K,Ca
K,Cu,Fe,Cl
Ca
K,Ca
on the outer portion of stage £
, November 13,
Elements
1973
Particle Type Comments
aluminum silicate, potassium
aluminum silicate, potassium
quartz
clay
clay
clay
clay
clay
aluminum silicate, potassium
clay
aluminum silicate, potassium
of the Andersen 8-stage impactor from the
Particle Type Comments
Effective Diameter
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
(wn)
3-3
1.8
14.0
It. 5
6.3
6.5
5-1
1-5
3-8
5-4
8.8
7-5
5-5
1.8
12.0
6.0
5-5
4.0
9-0
4.6
6.7
5-5
2.6
5.0
6.8
3-6
8.0
3-9
2.2
4.7
3.4
2-7
3.8
Major
Si,Al
Si,Al
Si
Ca,Si
Si
Si
Si,Al
Si
Si
Si
Si
Si
Si
Ca
Si
Ca.Si
Si,Al
Si
Si
Si
Si
Si
Ca
Si
Si
Si
Si
Si
Si
Si
Ti
Si
Si
Moderate
..
—
Al,Cl,K,Ca,
Fe
Al
Al,S,Ca,Fe
Trace
K,Ca,Fe
S,K,Cu,Ti
S,Ti,P,Mg
Fe,Mg,P,Fe
K,C1
Al,S,K,Ca,Fe Mg,S/Fb,Cl,
—
Al,K,Ca,Fe
K,A1
—
K,A1
Al
Al
Si
Al,K,Ca,Fe
A1,K
Fe
K,A1
Al,K,Ca,Fe
K,A1
Al,K,Ca,Fe
._
Al,P,Si,Fe
—
--
K,A1
Al,Fe
K,A1
—
K,A1
A1,K
A1,K
K,Ca,Fe
Ca
—
—
S/Pb,K,Ca,Fe
Ca
Al
Pb/S,Cl
Fe
K,Ca
Fe
Pb/S,Cl,Mg
Fe
—
Al,K,Ca,Fe
Mg,S/Pb
Mg,Al
Mg,Al,K,Ca,
Fe Fe
S/Pb,Cl,K,Ca
Fe
—
—
Fe
Fe
flyash (Si rich)
flyash (Si rich)
clay
flyash (Ca rich)
clay
clay
clay
clay
aluminum silicate,
potassium
quartz
aluminum silicate,
potassium
clay
clay
calcium mineral possibly asbestos or
•weathered calcite
clay
aluminum silicate, calcium
clay
aluminum silicate, potassium
clay
aluminum silicate, potassium
clay
quartz
flyash (Ca rich)
quartz
quartz
aluminum silicate,
potassium
clay
aluminum silicate,
potassium
quartz
aluminum silicate ,
Ti02 probably rutile
aluminum silicate,
potassium
aluminum silicate,
potassium
-------
191
Appendix Table &
Particles observed on the middle portion of stage 2 of the Andersen 8-stage impactor from the
Trout Farn location, November 13, 1973
Particle
Elements
Effective Diameter
No.
1
2
3
4
5
6
Y
8
9
10
11
12
13
14
15
16
17
18
19
(urn)
6.5
8.4
2.6
1.6
1.2
0.4
1.8
1.8
1
1
7
7
2.1
1
7
2.2
5
3
Ma ior
Si,Fe,K
Ca
Ca
Ca
Ca
Ca
Ca
Ca
Ca
Ca
Si,Ca
Si
Ca,Ti
Si
14.0
5-6
2.5
1.4
5-1
Ca
Ca,Al,Si
Ca,Al,Si
Si
Moderate
Al,Ti
Al,Si
Al.Si
Al,Si
Al,Si
Al,Si
Al,Si
Al,Si
Al,Si
Al,Si
Fe
Al,Ca,Fe
Si,Fe
Al,Fe,K
K
—
Fe
Fe
K,Al,Ca
Trace
S/Pb,Mg,Fe,
Ti
S/Fb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
S/Pb,Mg,Fe,
Ti
Al.Mg
K
Mg,Al,P,S/Pb
Mg,P,Cl,Ca
Mg,Al,Si,P,
Pb/S,Cl,Fe
Si,P,Cl
Mg,P,Pb/S,
Ti
Mg,P,Pb/S,
Ti
Particle Type
Comments
clay
flyash (Ca rich) conglomerate
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Ca rich)
clay
carbon
calcium mineral probably calcite
flyash (Ca rich)
flyash (Ca rich)
aluminum silicate morphology suggests
asbestos
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
SY
38
39
1
5
5
3-
4.
5.
5-
2
4
.2
0
0
5
3
5.0
4.
3-
l.
3-
7-
6.
3.
4.
2.
5-
4.
4.
1
9
3
7
2
0
0
0
0
0
1
3
Si
Si
--
Si
Si,Fe,Al
Si
Ca
Ca
Si
Si,Al
Si
Si
Si,Al
Si
Si
Si
Si
Si
Si
Si,Al
Al,K,Ca,Fe
Ca
—
—
Mg
—
—
Al,Ca,Fe
S
Al
—
K,Ca,Fe
Al,K,Ca,Fe
Al
Al,Ca,Fe,K
Al,K,Ca,Fe
—
Al,K,Ca,Fe
K
Ti,Mg,Pb/S,
Cl
K,A1
--
Fe,Al,K
K,Ca
Al,K,Ca,Fe
Al.Si
Al,Si
K
K,Ca,Fe,Cl
Mg,Ca,Fe
K,Ca,Fe,Al
--
—
__
—
—
—
--
clay
aluminum silicate,
calcium
carbon
quartz
aluminum silicate, iron
quartz
calcium mineral probably calcite
calcium mineral probably calcite
flyash (Si rich)
clay
clay
quartz
clay
clay
clay
flyash (Si rich)
flyash (Si rich)
quartz
clay
aluminum silicate,
potassium
-------
192
Appendix 7able
Particles observed on the center portion of stage
Trout Farm location, November 13, 1973
the Andersen 8-stage impactor from the
9
10
11
12
13
It
15
16
17
18
19
20
21
22
23
2U
25
26
27
28
29
30
31
32
33
3»*
35
36
37
38
39
1(0
1*1
1(3
kh
1(5
1(6
U7
W
1(9
50
51
52
53
51*
55
56
Particle
Effective.Diameter
Inn)
7-9
4.9
8-9
7-5
k.6
5.0
3-3
10.5
5-3
2.2
1.8
1-9
2.2
2.U
1.6
0.6
2.2
1.0
1.8
2.2
2.0
2.9
1.6
2.2
1.1+
1.2
1.8
1.2
0.7
1.2
1.8
1-7
2.1+
1.0
1.8
5-5
10.5
6.0
2.5
7.0
3-3
"(•5
6-5
5-0
6.5
5.0
5.0
8.0
3-0
5.0
5-7
6.3
8.5
8.8
Major
Si
Si
Si
__
Ca,Si
Si,Al
Si
Si
Si.Ca
Si
Si,Al
Si,Al
Si,Al
Ca
S1.A1
Fe
Ca
Ca
Si,Al
Si.Al
Ca
S1,A1
Si,Al
Si,Al
Si, Al
Si,Al
Si,Al
Ca
Si
Si
Si
81, U.
Si,Al
Si,Ca,Ti,Al
Si,Ca,Ti,Al
Fe
Si
Si
Si,Ca,Al
Si,Al
Si
—
--
Si
Ca
Si
Si
Si
Si
Si.Al
Ca
Si
Si
Si
Si
Si
Elements
Moderate
Al
Al,Ca
Al,K,Ca,Fe
—
A1,K
K
Al,Fe
Al,Ca
Fe,Al,P
—
Fe,Ca,K
Ca,Fe
Ca,Fe
Al,Si,Fe
Fe,Ca,K
—
Al,Si,Fe
Al,Si,Fe
Fe,Ca,K
Ca.Fe
Al,Si,Fe
Fe,Ca,K
Fe,Ca,K
Ca,K,Fe
Ca,K,Fe
Ca,Fe
Ca.Fe
Al,Si,Fe
Ca,Al,Fe
Ca,Fe,S
Ca,Al,Fe
Ca,Fe,S
Ca,Fe,Ti
Fe
Fe
—
—
K,A1
Fe
Ca,Fe
Al,K,Ca,Fe
--
—
K,Al,Fe
Si,Fe
K,Al,Fe
Al,Cu
A1,K
Al,Fe
K
Fe,Al,Si
Al
Al,K,Ca,Fe
K,A1
Al,Fe
—
Trace
K,Ca,Fe
—
—
--
Fe,Mg,Ti,P,
Pb/S
Fe
K,Ca
—
01 ,K
Al,K,Ca,Fe
Mg,Ti,Pb/S
Mg,K,Ti
Mg,K,Ti
Mg,P,Pb/S,Ti
Mg,Ti,Pb/S
Al,Si,Ca
Mg,P,Pb/S,Ti
Mg,P,Pb/S,Ti
Mg,Ti,Pb/S
Mg,K,Ti
K,P
Mg,Ti,Pb/S
Mg,Ti,Pb/S
Pb/S
Pb.S
Mg,K,Ti
Mg,K,Ti
Mg,P,Pb/S,Ti
K,Pb/S
K,P
K,Pb/S
K,Ti
Pb/S,Cl,K
K,Ti
K,Ti
Al,Sl,Ca
—
—
Pb/S,Ti,Mg
K,Ti
Mg
K,Al,Cl,Pb/S,
Si,P
K,Al,Cl,Fb/S,
Si,P
--
Al,Pb/S,Mg,
P --
—
Ca
K,Ca,Mg
—
Mg,P,Pb/S,Ti
Ca,Fe
—
Fe
K,Ca
Al,Pb/S,K
Particle Type Comments
clay
aluminum silicate,
calcium
clay
carbon
aluminum silicate ,
calcium
aluminum silicate,
potassium
clay
aluminum silicate, calcium
aluminum silicate, calcium
quartz
flyash (Si rich) conglomerate
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Fe rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich) y
flyash (Si rich) conglomerate
flyash (Ca rich) 1
flyash (Ca rich) 1
flyash (Fe rich)
quartz
aluminum silicate,
potassium
flyash (Ca rich)
flyash (Si rich)
clay
carbon
carbon
aluminum silicate,
potassium
mineral calcium probably calcite
aluminum silicate,
potassium
aluminum silicate,
calcium
clay
clay
aluminum silicate,
potassium
flyash (Ca rich)
clay
clay
aluminum silicate,
potassium
clay
quartz
-------
193
Appendix Table 10. Particles observed on the outer portion of stage 3 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Particle
Elements
Particle Type
Comments
No.
1
2
3
k
5
6
7
8
9
10
11
12
13
ll*
15
16
17
18
19
20
21
22
Effective Diameter
(wO
1.8
0.6
0.8
2. It
1.6
o.i*
0.8
1.1*
0-3
1.1
0.6
1.8
0.9
0.8
l*.0
0.9
2.0
1-7
0.6
2.1*
1-9
0.6
Major
Si,Al
Si
3i
Si
Si,Al,K
Fe,Al,Si
Si
Ca
Fe
Si,Al
Si
Si
Si,Al
Si
Si
Si
Ca
Si
Si
Si
Si
Si
Moderate
K,Ca
Al
Al
A1,K
Fe
--
__
Al,Si,P,Fe
—
Fe
—
Al,K,Ca
—
Al,Ca,S
A1,K
Al
Mg.Si
—
A1,K
—
A1,K
Al
Trace
Mg
—
--
—
Mg
—
Fe,Mg
Mg,K
--
—
—
Fe
--
Fe
--
K,Ca
Fe
—
—
Al.Fe
Mg,Fe
--
clay
clay
clay
aluminum silicate,
potassium
aluminum silicate,
potassium
aluminum silicate,
iron
quartz
flyash (Ca rich)
iron
aluminum silicate,
iron
quartz
clay
clay
aluminum silicate,
calcium
aluminum silicate,
potassium
clay
mineral calcium probably dolomite
quartz
aluminum silicate,
potassium
quartz
aluminum silicate,
potassium
clay
-------
194
Appendix ?able 11.
Particles observed on the middle portion of stage
Trout Farm location, November 13, 1973
the Andersen 8-stage impactor from the
Particle
Effective Diameter
No. _ (m) _
1 1-9
2 2.9
3 2.7
It 3-0
6 2.9
8 0.9
9 2.3
10 2.9
11 1-0
12 1.0
13 0.8
Ik 1.0
15 2.7
16 2.8
17 1-5
18 3-3
19 1-9
20 2.0
21 1.2
22 1.1
23 0.7
2k 1.1*
25 0.6
26 2.2
27 1.0
28 (filter)
29 i.o
30 1.6
31 2.1*
32 0.9
33 0.5
31*- 1-5
35 1-7
36 l.l
37 2.7
38 0.8
39 1-9
1*0 1-9
1*1 1-5
1+2 0.6
1*3 0.7
1*1* 0.9
1*5 0.5
1*6 o.i*
1*7 0.50
1*8 0.5
1*9 2.6
50 0.9
51 1-7
52 2.5
53 1-5
51* 1.8
55 1-1
56 2.1*
57 5-8
58 1.6
59 2.9
Major
Ca
Si
Si
Si
Si
Si
Elements
Moderate Trace
Mg Si
Al,K,Fe Ca
Al,K,Ca,Fe
Al,K,Fe Ca
Al,K,Fe Ca
A1,K
Particle Type
mineral calcium
clay
flyash (Si rich)
clay
clay
aluminum silicate,
potassium
Si
Si
Si,Ca,Fe
Si
Ti
Pb,Cl
Si
—
Ca
Si
Si,Al
• •
• •
V te
•* •
Al,Fe
Al,K,Ca,Fe
Al
Al K,Ca,Fe
Si,Al,Fe K,Ca
Br
aluminum silicate, iron
clay
flyash (Ca rich)
flyash (Si rich)
mineral titanium
lead bromochloride
quartz
A1,S1,S,C1,
K,Ca
Si,Al,Fe
Al,K,Ca,Fe
Fe
• «• « •
• • • •
carbon
mineral calcium
clay
clay
caj
^
-bon
r
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Si
Fe
Si
Si
Si
Si
Comments
probably dolomite
probably rutile
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,K,Ca
Al,K,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,K,Ca
Al,K,Ca,Fe
Al,K,Ca,Fe
A1,K
Al,K,Ca,Fe
Al,K,Ca,Fe
Fe
Al,K,Ca,Fe
Si,Ti
Al
Al,K,Ca,Fe
Pb,S
Al,K,Ca,Fe
Al,K,Fe
Fe,Al
Si,Fe
Fe
Br,Cl
flyash (Si rich)
quartz
clay
quartz
clay
aluminum silicate,
potassium
mineral calcium
carbon
clay
clay
clay
clay
clay
clay
clay
aluminum silicate,
potassium
clay
quartz
quartz
clay
quartz
quartz
clay
mineral iron
clay
clay
auto emission
clay
aluminum silicate,
potassium
probably calcite
possibly ilmenite
-------
195
Appendix Table 13.
Particles observed on the center portion of stage
Trout Farm location, November 13, 1973
3 of the Andersen 8-stage impactor from the
Mo.
1
2
k
5
6
7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2k
25
26
27
28
29
30
31
32
33
35
36
37
38
39
1+0
in
1*2
1*1*
"*5
1*6
Particle
Effective Diameter
(Km)
i.a
0.9
1.0
0.5
i.a
2-3
2.2
i*.o
2.8
2.6
2-3
2.5
1-3
2.3
^•3
1.0
1.6
2.5
3-1*
3-0
2.2
1.6
0.7
3-0
l*.l
1-7
0.7
1-3
1.8
2.0
2.5
2.5
1-5
2.1
1.8
2.0
2-5
2.1*
1-9
2-7
1.1
3-0
l.l*
2.9
2.2
2.1*
Major
Si
Ca
Si
Si
Si
Si
Si
Si
Si
Ca
Ca
Fe
Si
Si
Si
Si
Si
Si
Si
Si
Al,Si
Si
Si
Si
Si
Ca,Si
Si,Al
Si,Al
Si,Al
Ca
Si
Si
Si
Fe,Ca,Si
Si
Si
Si
Si
Ca
Si
Si
Si
Ti
Si
Si
Al
Elements
Moderate Trace
Al,K,Ca,Fe
Al,Si,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe,S
Al,K,Ca,Fe
Al
Al,K,Ca,Fe
Al,K,Ca,Fe
A1,K
P,Si
P,Si,Fe K,A1
Si Al,K,Ca
Al,K,Ca,Fe
Al,K,Ca,Fe
A1,K
Al,K,Ca,Fe
Al,Ca,Fe
Al K,Fe
A1,K
K,Al,Ca,Fe
K,Ti,Fe
Ca K,Fe,Al
Al Ca,Fe
—
Al,K,Fe
Al K,Fe
—
K,Fe,Ca
—
Si
-.
Al,K,Ca,Fe
A1,K
Mn Zn
Al,K,Ca,Fe
Al,Fe
Al,K,Ca,Fe
Al,Fe
Al,Si,P,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,Ca,Fe Ti
Fe Al,P,Si
Al,K,Ca,Fe
Al,K,Ca,Fe
Al,K,Ca,Fe
Particle Type
flyash (Si rich)
flyash (Ca rich)
clay
flyash (Si rich)
flyash (Si rich)
clay
clay
clay
aluminum silicate,
potassium
mineral calcium
mineral calcium
mineral iron
clay
clay
flyash (Si rich)
clay
flyash (Si rich)
clay
aluminum silicate,
potassium
clay
flyash (Si rich)
clay
clay
quartz
aluminum silicate,
potassium
aluminum silicate,
calcium
clay
clay
clay
mineral calcium
quartz
clay
aluminum silicate,
potassium
mineral-Fe,Ca,Si
clay
quartz
clay
quartz
flyash (Ca rich)
clay
clay
aluminum silicate,
calcium
mineral titanium
clay
clay
aluminum (oxide)
Comments
probably apatite
possibly fractured from
#10
possibly weathered quartz
probably calcite
probably rutile
-------
196
Appendix Table 13
Particles observed on the outer portion of stage 4
Trout Farm location, November 13, 1973
of the Andersen 8-stage impactor from the
Mo.
6
7
8
9
10
11
12
13
lit
15
16
17
18
19
Particle
Effective Diameter
(W)
1.6
1-5
2.3
2.0
1-3
1.0
1.1
0.8
0.5
1.6
1.0
1.2
1.2
1.0
1-3
l.U
2.2
1-3
1-5
1.8
l.O
0.9
1.8
0.6
1.0
1.1
0.9
l.lt
0.8
0.8
1.2
0.9
1.2
1.0
0.7
1.6
0.6
1.2
1.4
0.9
0.7
l.l
0.8
0.7
1.2
0.6
1.8
0.9
1.1
1.2
l.U
0.6
1.1
0.9
l.l*
1-3
Major
Al,Si,S
Si,Fe,Ca
Si,Al
Si,Fe,Ca
Si
Si,Al
Si
S,C1
Si,Al,S,Ca
Al,Si,Ca,Fe
Si
Si,Al
Si
Si,Al
--
—
Si
Si
Si
Si
Si,Al
Fe
Si
Si
Ca,Ti
Si,Al
Si,Al
Si
Si
Si,Fe
Si
Si
Si
Si
Pb
Ca,S
Ca
Si
Sl,Fe
Si
Si
Si
Si
Si
Fb
Pb
31
Fo
Sl,Al
Ca
,-
__
Ca
Fe
Si,Al
Si,Al
Elements
Moderate Trace
..
Al
K,Fe Ca
Al
Mg,Al,Pb,S,
Cl,K,Ca,Fe
Fe Ca,K,Pb/S
Al,K,Ca,Fe
Si,Al,Ca
Cl,Ti,Fe Mg,K
—
Al,K,Ca,Fe
Fe
--
Ca.Fe
Al,Si
Al,Si
A1,K
A1,K
A1,K
Al,K,Ca
—
A1,S1
A1,K
A1,K
Al,Si,Fe
K,Ca,Fe
Ca,Fe
Al,Mg,K,Ca,Fe
—
Al
Al,K,Ca,Fe
K Al
K,Fe Al
Al,K,Ca,Fe
-_
Si
Si,Al
Al,K,Ca,Fe,Pb,Cl
Mg,Al,K,Ca
_-
Al,K,Ca,Fe
Al,K,Ca
Al,Fe K,Ca
Al K,Ca,Fe
Cl,Br/Al
Cl,Br/Al
Al,K,Ua,Fe Pb/S
Mg,Si,Cr,Mn,
Zn
Fe,Ca K
Al,Si
Cl,K,Cci
__
__
Al,Si,Ca
K,Ca,Fe
Mg,K,Ca,Fe
Particle Type Comments
mineral fluorescence similar to
that of ultramarine
aluminum silicate,
calcium and iron
clay
aluminum silicate,
calcium and iron
aluminum silicate
flyash (Si rich)
clay
unknown (salt)
flyash (Ca rich)
flyash (Ca rich)
clay
flyash (Si rich)
quartz
flyash (Si rich)
carbon
carbon
aluminum silicate,
potassium
aluminum silicate,
potassium
aluminum silicate,
potassium
clay
flyash (Si rich)
flyash (iron)
aluminum silicate, potassium
aluminum silicate, potassium
flyash (Ca rich)
flyash (Ca rich)
flyash (Ca rich)
clay
quartz
aluminum silicate, iron
flyash (Si rich)
aluminum silicate.
potassium
aluminum silicate,
potassium
clay
Lead (oxide)
mineral calcium probably calcium sulfate
mineral calcium probably calcite
clay (w/lead chloride)
aluminum silicate probably asbestos
quartz
clay
clay
clay
clay
lead bromochloride
lead bromochloride
clay
iron probably steel
flyash (Si rich)
mineral calcium probably calcite
carbon
carbon
mineral calcium probably calcite
iron, mineral
clay
flyash (3i rich)
-------
197
Appendix Table 14. Particles observed on the middle portion of stage
Trout Farm location, November 13, 1973
of the Andersen 8-stage irapactor from the
No.
1
2
3
l*
5
6
7
8
9
10
11
12
13
15
16
17
18
19
20
21
22
23
2U
25
26
27
28
29
30
31
32
33
31*
35
36
37
38
39
Particle
Effective Diameter
(Wn)
1-3
2.0
0.8
0.8
1.0
3-0
1.6
0.5
1.0
1.8
1.2
O.U
1.2
0.8
1.0
0.8
0.6
3.3
1.2
0.6
0.7
1.1
1.0
0.6
0.7
2.0
1.8
2.8
1-3
1-3
0.6
1.1
1.2
0.8
3-1*
0.8
0.5
1.8
0.6
Major
Si,Al
Si.Al
Si,Al
Ca,Al
Si,Al
Si
Si,Al
Ca
Si,Al
Si
Si
Si
Si,K
--
Si
Si
Si,Ca
S1,A1
Ca
Si
Si
Si
Si
Si,Al
Si
Si
Si
Si
Si
Si,Al,Ca
Si,Al
Si
Si
Si
Si
Si
__
Si,K
Si,Al
Elements
Moderate
K,Ca
Ca.Fe
—
Si,Fe,K
K,Ca,Fe
Al,Fe
K,Ca
Al,Si,Fe,Mg
Ca,Fe
--
Al,K,Ca,Fe
Al,K,Ca,Fe
Al
Al,Si,Fe
--
K,Ca,Al
Al
K,Ca,Fe
S
S,Ca,Al
Al.Ca
Al,K,Ca
Al,K,Ca
K,Fe
A1,K
Al
Al
Al,K,Ca,Fe
Al,K,Ca,Fe
Fe
--
Al,K,Ca,Fe
__
Al,K,Ca,Fe
—
__
S,Ca
Al
--
Trace
Mg,Ti,Cr,Fe
Mg,K,Ti
Mg,Fe,K,Ca
—
K,Ca
Fe
—
K,Mg
Mg,Fe
--
—
K,Ca,Al,Fe
—
--
—
Al,Si
—
—
--
—
—
Fe
Fe
—
—
Ti
Fe,Pb/S,Cl,K,
Ti,V
__
Al,Pb/S,K,Ca,
Fe
Al,Si,Cl
—
Fe
Particle Type
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyash (Ca rich)
flyash (Si rich)
clay
clay
flyash (Ca rich)
flyash (Si rich)
quartz
clay
clay
aluminum silicate,
potassium
clay
clay
clay
aluminum silicate,
calcium
clay
calcium-sulfur
clay
aluminum silicate,
calcium
clay
clay
flyash (Si rich)
aluminum silicate.
potassium
clay
clay
clay
flyash (Si rich)
flyash (Ca rich)
Ca flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
clay
flyash (Si rich)
calcium-sulfur
aluminum silicate ,
potassium
clay
Comments
associated
probably gypsum
probably gypsum
-------
198
Appendix 1'able 15. Particles observed on the center portion of stage
Trout Farm location, November 13, 1973
of the Andersen 8-stage impactor from the
Particle
Elements
Effective Diameter
Ho.
1
2
3
1*
5
6
7
8
9
10
11
12
13
ll*
15
16
17
18
19
20
21
22
23
2k
25
26
27
28
29
30
31
32
33
31*
35
36
37
38
39
1*0
VI
(Mm)
3-3
1.8
1.0
0.8
0.6
1.2
1-3
2.0
0.9
0.8
1.8
0.8
0.2
5-2
1.6
1.0
1.1*
1.1*
0.8
0.9
0.9
1.2
2.2
1.0
2.1
1.2
2.7
1.1
1.1+
1.0
1-7
0.5
0.9
1.0
0-9
0.7
1.2
0.7
1.6
1.1
1.8
Major
Si
Si
Si
Si
Si
Si
Si
Si
Fe
Si
—
Fe
Si
Si,Al
Si
Si
Si,Al
Si
Si
Si,Al,Ca
--
Al,Ca
Fe
Al,Pb
Ca
Si
Si
Si,Ca
Si
Si
—
Si
Fe,Sl
Si,Al
Si,Al
Ca
S1,A1
Si
Si
Si
Si
Moderate
Al,Ca
Al
Al
Al
Al,K,Ca,Fe
Al,Ca
Al,Fe,K,Ca
Al,K,S,Cl,Fe
Al,Si
Al,K,Fe
—
—
—
Fe,K,Ca
Al
Al,Ca,Fe
—
—
—
Fe,Mg
Pb,Si,Al,Ca
Fe,Ti
__
Fe.Br
__
Al,K,Fe
Al,K,Fe
—
A1,K
Al,K,Ca,Fe
Si,Ca,Fe,Al
Al,K,Ca,Fe
Al,Ca,K
Fe
—
—
A1,K
Al,K,Fe
Al,K,Ca,Fe
Mg,Al,K,Fe
Trace
--
Fe,K,Ca
Fe,K,Ca
Fe,K,Ca
—
—
—
Si,Al,Pb/S,
Cl,K,Ca
Si
Al,K,Fe
Mg,Pb/S
—
K
K,Ca,Fe
—
—
—
Fe
--
Al,Si
Cl
Si
Pb/S,Cl
Pb/S,Cl
Mg,Al,Fe
—
—
Pb/S,Cl,K
--
—
K,Ca
Ca,Fe
Al,Si,Pb/S,Fe
__
Pb/S
--
—
—
Particle Type Comments
Aluminum silicate, calcium
clay
clay
clay
clay
aluminum silicate, calcium
clay
aluminum silicate, potassium
iron, mineral
iron, mineral
carbon
iron
quartz
clay
clay
flyash (Si rich)
clay
quartz
quartz
aluminum silicate,
calcium
aluminum silicate,
calcium-vith lead
flyash (Ca rich)
iron
flyash (Pb rich)
mineral calcium probably calcite
aluminum silicate,
potassium
aluminum silicate, potassium
flyash (Ca rich)
aluminum silicate,
potassium
clay
clay
clay
aluminum silicate, iron
clay
flyash (Si rich)
mineral calcium probably calcite
clay
aluminum silicate, potassium
aluminum silicate,
potassium
clay
magnesium aluminum
silicate probably asbestos
-------
199
Appendix Table 16.
Particles observed on the outer portion of stage 5 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Particle
Effective Diameter
No. (urn)
1 1-3
2 0.3
3 1-2
1* 0.»*
5 0.1*
6 0.5
7 0.5
8 0.1*
9 0.3
10 0.3
11 0.1+
12 1.1*
13 0.8
lit 1.0
15 0.9
16 0.7
17 0.3
18 0.7
19 0.5
20 0.5
21 0.6
22 0.7
23 0.5
21* 0.6
2s 0.6
26 0.5
27 0.1*
28 0.9
29 1.1
30 2.0
33 3-8
32 1.6
33 0-5
31* 0.9
35 0.6
36 1.0
37 0.8
38 O.l*
39 0.5
1*0 0.1*
1*1 0.3
1*2 1.2
1*3 0.7
1*1* 1.2
1*5 0.1*
1*6 1.9
1*7 0.6
1*8 O.l*
1*9 0.6
50 0.2
51 0.6
52 0.3
53 O.l*
5U 0.6
55 0.6
56 0.7
57 0.5
58 0.3
59 0.1*
60 O.l*
Elements
Major
Pb,Br
Pb,Br
Pb,Br
Pb,Br
Si
Si
Fe,Zn
Fe
Pb,Br
Si,Fe,Mn
Si
Si
Si
Pb,Bl*
Si
--
Br,Pb
Zn,Pb
Zn.Pb
Si
Si
Pb,Br
Br,Pb,Si
Si
Si
Si
Si,Pb,Br
Pb,Br
Si
Si
Si
Si
Si,Al
Pb,Br
Ca
—
Si,Al
Si
—
Ca
Si,Al
Fb,Br
Si
Ca
Pb,Br,Cl
Si
Pb,Br
Si
Si
Si
Pb,Br
__
Si
Ca
Ca
Ca,S
Pb,Cl
Pb,Cl,Br
Pb,Cl,Br
Moderate
__
—
Al,Si,Zn
Si
Al,Ca,Fe
Al
--
—
--
—
Al
Al
Al,K,Ca,Fe
--
Al,K,Ca,Fe
—
Cl
--
—
--
--
--
K,Ca,Ti,Fe
Al,K,Ca,Fe
Br,Pb,Ca
Br,Pb,K
—
K,Ca,P,Mn,Zn,
Ti
Al/Br,Fb,K,Ca
Al,K,Ca,Fe,Mg
Al,K,Ca,Fe,Mg
--
Cl,Ca,Fe
Si
--
Al,Si,Ca,S
—
_-
Al
Al,Si
-•
--
Al
Al,Si
--
Al,Ca
--
__
—
Al.Ca
--
—
Fe,Al,Si,S
Al.Si
—
Br
—
—
Trace
..
—
--
_.
Pb/S
..
Al, Si, Pb/S,
Cl,K,Ca
Zn,Pb/S,Cl,
K,Ca
Cl,Ca,Fe
Al,Pb,Zn
Pb/S,K,Ca,Fe
Ti,Cr,V,Mn
—
—
Cr
__
Br/Al
Al,Si,C
--
--
--
--
--
--
--
—
—
,Fe
--
--
--
—
--
--
--
--
--
Si,S,Cl
Fe
Ca
Cl
K,Fe
Pb/S,Fe
Zn,Fe
P/S
Cl
Al
Al
--
Cl
Pb,Br,Cl
—
Mg,P,Ti,Zn
Fe,Cl,£n
Al,Si,Cl,K
Zn
--
—
Particle Type
lead bromochloride
lead bromochloride
lead bromochloiide
lead bromochloride
aluminum silicate,
calcium
clay
metal
iron
lead bromochloride
silicate(j)
clay
clay
flyash (Si rich)
lead bromochloride
clay
auto emission
lead bromochloride
metal
metal
quartz
quartz
lead bromochloride
clay
clay
aluminum silicate,
calcium
aluminum silicate,
potassium
quartz
lead bromochloride
clay
flyash (Si rich)
flyash (Si rich)
quartz
aluminum silicate,
calcium
lead bromochloride
mineral calcium
clay
clay
quartz
metal
aluminum silicate,
calcium
carbon
clay
lead bromochloride
cljay
mineral calcium
mineral calcium
aluminum silicate,
calcium
aluminum silicate,
calcium
quartz
quartz
aluminum silicate,
calcium
aluminum silicate,
calcium
auto emission
quartz
flyash (Ca rich)
aluminum silicate,
calcium
calcium-sulfur
lead bromochloride
lead bromochloride
lead bromochloride
Comments
associated w/auto emission
associated w/auto emission
associated w/auto emission
associated w/auto emission
associated w/auto emission
probably calcite
probably calcite
probably gypsum
-------
200
Appendix Table 16 (Con't). Particles observed on the outer portion of stage 5 of the Andersen 8-stage impactor from the
Trout Fa TO location, November 13, 1973
No.
61
62
63
64
65
66
67
68
69
70
71
72
73
7i*
75
Note:
Particle
Effective Diara
(urn)
0-7
0.6
0-3
0.6
0.1*
0.6
0.1*
0.5
0.1*
2-5
0.5
0.9
0-9
0-5
1.1*
Elements
Major
Al,Si,Ca,Pe
Ca
Pb
Si,Al,Ca,Fe
31
Si,Al
Si,Al
Si,Al
Si,Al
Si
Si,Al
Cl
Cl
Pb,Br
Si,Ca
Moderate
..
—
—
—
Al,K,Ca,Fe
-.
—
--
—
--
—
Ca
Ca
Cl
Al,K,Fe
Trace
t
__
Cl,Br
—
—
.-
—
Fe,Ca,K
Fe,Ca,K
Al
—
Pb/S,P,Al,Si
Pb/S,P,Al,Si
—
—
Particle Type
flyash (Ca rich)
mineral calcium
lead (oxide)
aluminum silicate,
calcium
clay
clay
clay
clay
clay
quartz
clay
chloride
chloride
lead bromochloride
aluminum silicate,
calcium
Comments
probably calcite
probably calcium chloride
probably calcium chloride
Particles 1-29 associated with an amorphous mass giving a trace of Pb,Br; particles from 1*8-63, 66-69 associated
with a more tenuous amorphous mass.
Appendix Vable 17- Particles observed on the middle portion of stage 5 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Particle
Elements
Particle Type
Comments
Effective Diar-eter
No.
1
2
3
it
5
6
7
8
q
y
10
11
12
13
1U
15
16
17
18
19
20
21
22
2^
*- J
2k
25
26
27
28
29
fr\
Ju
31
32
33
31*
(lim)
2.9
0.6
1-7
1.0
0.6
0.6
0.7
1-3
0.9
0.6
1.2
0.6
0.7
0.9
0.1*
1.2
0.6
0.1*
0.5
0.8
o .^
" • 0
0.8
0.6
0.6
0.7
0.3
0-7
2.0
0-5
O.U
0.6
0.1*
0.5
0.6
Ma ]Or
Si,Al
Si
Al,Si,Fe
Al,Sl,Fe
__
Al,Si,Fe
Ca
Pb,Cl
__
Zn,Cl
Si
Si
Si
__
Si,Al
Si
Pb
--
Fb
Si
Si
Si
Ca,Al,Si,K
Al,Si,Ca
Ai
..
Pb,Br
ZnPb
Si
Ca
Moderate Trace
Fe,K,Ca,Pb
—
K,Ca
K,Ca
Si
K,Ca
Al,Si
Br
__
Pb/S
--
—
A1,S K,Ca,Fe
—
Al,K,Ca,Fe
Br,Cl
Si,Al,Ca P,S
Si
Ca,Fe
K,Ca,Fe Pb/S
Pb/S
—
_-
..
Al,Si,Fe
Al,Fe
Al,Si,Fe
clay
quartz
aluminum silicate,
iron
aluminum silicate,
iron
quartz
aluminum silicate,
iron
flyash (Ca rich)
lead bromochloride
carbon
auto emission
zinc chloride
flyash (Si rich)
quartz
clay
carbon
clay
flyash (Si rich)
lead bromochloride
clay
flyash (carbon) could also be latex or
polyvlnyl acetate spherei
lead (oxide)
flyash (Si rich)
flyash (Si rich)
flyash (Si rich)
flyaeh (carbon) see particle #20
flyash (Ca rich)
flyash (Ca rich)
quartz
carbon
lead bromochloride
flyash (carbon) see particle #20
metal
flyash (Si rich)
flyash (Ca rich)
-------
201
Appendix Table 18. Particles observed on the center portion of stage 5
No.
1
2
3
k
5
6
1
8
9
10
ll
12
13
lit
15
16
17
18
19
20
21
22
23
2lt
25
26
27
28
29
30
31
32
33
Trout
Particle
Effective Diameter
(wn)
0.7
1.2
0.6
0.5
0.4
1-3
0.4
0.8
0.6
0.8
0.7
0.7
0.9
0.9
0.5
1.2
0.8
0-7
0.5
0.7
0.1*
0.4
0.8
1.2
0.6
0.9
0.7
0.4
1.4
1.2
0.8
0.6
1.1
Farm location, November 13, 1973
Elements
Major Moderate Trace
Si,Al
Si
Fe
Si.Al Fe
Si
Fe
Pb
Pb,Br
Si,Al
Si
Si,Al — Fe
Fb,Br
Si,Al,S
—
Si
Si.Fe Al
Si Al,Fe
Ca Al,Si,P,S,Fe
Fe
..
Fb,Br Cl
Si,Al
Si Al,Fe,K,S
Ca
Si -- Mg,Ca,Fe
Si
Fe,Zn Pb,K,Al,Si
Pb Br
Si
Al,Si,Pb/S,Cl
Pb,Br
Si Al,K,Ca,Fe
Si
of the Andersen 8-st
Particle Type
clay
quartz
iron (oxide)
clay
quartz
iron (oxide)
lead (oxide)
lead bromochloride
clay
quartz
flyash (Si rich)
lead bromochloride
clay (w/sulfur)
carbon
quartz
aluminum silicate ,
iron
clay
flyash (Ca rich)
iron (oxide)
flyash (carbon)
lead bromochloride
clay
clay
mineral calcium
flyash (Si rich)
quartz
flyash (metal rich)
flyash (lead rich)
quartz
auto emission
lead bromochloride
flyash (Si rich)
quartz
'rom the
Comments
[see Appendix Table 17]
calcite
-------
202
Appendix Table 19. Particles observed on the outer portion of stage 6 Of the Andersen 8-stags impactor from the
Trout Farm location, November 13, 1973
No.
1
2
3
1*
5
6
7
8
9
10
11
12
13
li*
15
16
17
18
19
20
21
22
23
2k
25
26
27
28
29
30
31
Particle
Effective Diameter
(nm)
2.8
1.0
1.1
0.3
0.5
1.2
0.9
1.1
0.6
0.2
0.2
o.i*
1*.0
1.1
1.1
0.9
1.1
o.k
1-5
0.8
0.1*
0.1*
1-5
0.5
0.3
0.8
0.8
0.6
2.0
1.2
0.3
Major
..
—
Si
Fe
Si,Al
—
Si
Si
Mn,Fe
Si
--
--
--
—
S,Ca,Cl
—
Pb/Br
__
—
--
—
__
--
Pb.Br
P
Fb,Br
Si
—
—
—
Appendix Table 20. Particles observed on
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
Ik
15
16
17
18
19
20
21
22
23
Trout
Particle
Effective Diameter
((im)
1.2
0.3
0.1*
0.3
0.7
0.1*
0.6
0.5
0.6
0.9
0.3
1-3
0.3
0.2
0.3
0-3
2.2
2.9
2.6
3-5
1.1
1-3
3-8
Farm location,
Major
—
Pb,Br,Cl
Al,Si,S,Fe
Pb,Br,Cl
Mn
Pb,K
Pb,Br
Si
Pb,Si
—
Pb,Fe,Zn
—
Zn,S
Pb,Br
—
„
Ca
Si
--
Elements
Moderate
S,K,Zn
Pb,Br
__
—
—
—
Al,K,Ca,Fe
A1,K
—
Al,K,Ca,Fe
Si
—
--
—
--
—
—
__
—
S,Zn
Al,Si,Pb,
S,Ca
—
--
—
—
—
—
—
—
Si
Trace
Al,Si,Cl,Ca
Cr,Mn,Fe
—
Al,Pb/S
Si
Pb/S,Fe
—
—
—
—
—
--
Pb/S,Al/Br
Pb/S,Al/Br,Si
—
-_
Si,S/Pb
—
__
Pb/S,Si,
Al/Br,Fe
zinc/sulfur s
—
__
Pb/S,K
Fe,Zn
S
—
—
Pb/S
—
~~
the middle portion of stage
November 13,
Elements
Moderate
__
Al,Si,Ca
Zn
—
—
—
—
—
Al,K,Ca,Fe
—
—
Si
—
—
—
—
—
--
„_
—
Al,K,Ca,Fe
—
1973
Trace
__
—
__
Pb,Cl,Br
K,Ca
Si
Al, Si, Pb/S
Cl,Fe,Zn,P
—
—
—
Pb/S, Si,
Al/Br,Fe
__
Mn
—
Fe,Cl
Pb/S
Pb/S,Zn,
Al/Br
Pb/S,Al/Br,
Cl.Zn
Pb/S,Zn
A1,P
--
--
Particle Type
auto emission
auto emission
quartz
iron (oxide)
clay
carbon
clay
aluminum silicate,
potassium
flyash (metal rich)
flyash (Si rich)
quartz
auto emission
auto emission
auto emission
calcium salt
carbon (soot)
lead bromochloride
auto emission
auto emission
alt
aluminum silicate,
calcium
auto emission
carbon( soot)
lead bromochloride
phosphate
lead bromochloride
quarts
auto emission
auto emission
quartz
6 of the Andersen 8-stage
Particle Type
auto emission
aluminum silicate,
calcium
lead bromoehloride
metal
clay
lead bromochloride
metal
unknown
lead bromochloride
flyash (Si rich)
flyash (raetel rich)
auto emission
metal
carbon w/metal
zinc -sulfur salt
lead "bromochloride
auto emission
auto emission
auto emission
flyash (Ca rich)
flyash (Si rich)
auto emission
Comments
no lead Ma peak (l)
associated
si-
no areas of elemental
concentration
embedded in #29
embedded in #29
embedded in #30
embedded in #32
impactor from the
Comments
possible flyash
associated with #23
associated vith #23
embedded in #23
embedded in #23
embedded in #23
embedded in #23
embedded in #23
embedded in amorphous mass
-------
203
Appendix Table 21. Particles observed on the center portion of stage 6 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Particle
Elements
Particle Type
Effective Diameter
No.
1
2
3
4
5
6
7
8
9
10
ll
12
13
14
15
16
17
18
19
20
21
22
23
24
(wa)
1.1
3-6
0.6
0.7
o-3
0.3
0.3
1.6
0.6
0.7
1-3
0.4
0.6
0.4
0.9
3-5
1-3
0-3
1.4
1-5
3-10
0.9
0.8
0.6
Major
Si,Al
--
Si
Mg
—
—
—
—
—
Pb,Br,Cl
--
—
Zn
—
Si,Al
—
—
—
—
—
—
Fe
Pb
—
Moderate
__
Pb
Al,K,Ca,Fe
S,C1
Si,S
Cl
Ca,P,Al,Si
__
--
-~
—
S,C1
Pb,Br,Cl
Fe
—
—
—
—
—
—
Si.Al
—
—
Trace
Fe
—
Si,Al,Cr
—
Pb/S
__
__
__
—
Pb,Br,Zn
Oa
—
—
—
Fb,Cl,Zn
Pb,Cl,Zn
—
—
Zn
Pb
—
Fe
—
flyash (Si rich)
auto emission
flyash (Si rich)
unknown mineral
flyash (Si rich)
salt
unknown mineral
auto emission
auto emission
lead bromoehlorf.de
auto emission
calcium mineral
zinc salt
auto emission
flyash (Si rich)
auto emission
auto emission
carbon
auto emission
auto emission
auto emission
flyash (Fe rich)
lead (oxide)
carbon
Comments
possibly sodium chloride
embedded in #16
embedded in #16
embedded in #16
embedded in #19
Appendix Table 22.
Particles observed on the outer portion of stage
Trout Farm location, November 13, 1973
7 of the Andersen 8-stage impactor from the
Particle
Effective Diameter
No. (uml
1 2.3
2 1.2
3 1-9
4 0.4
5 0.6
6 0.6
7 1-5
8 0.4
9 0.2
10 0.3
11 0.3
12 0.4
13 0.2
14 1.0
15 0.1
16 0.1
17 0.4
18 0:9
19 0.4
20 0.1
21 0.2
22 1.4
23 4.6
24 0.7
25 0.4
26 0.3
27 0.9
28 0.5
Elements
Major Moderate
..
—
--
—
S
—
S
—
—
—
—
-_
—
-_
—
—
—
Pb,K
—
—
—
--
Fb,Br
Si,Al,K
Si,Pb,Ca
--
Sl,Pb,S,Ca,
Fe
Trace
Pb
Cr,S,Zn
Pb/S,Br/Al
S
Zn
Pb/S
Pb/S,Zn
S,Zn
S
S
S
s
Pb/S,Cl,Al/Br
S
Si,S
S
S
Al/Br
Pb/S ,Al/Br
Si, Pb/S
--
—
--
Pb,Zn,Fe
Fe,K,Cl,Al
Fe
Cr
Zn,Al,K
Particle Type
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
auto emission
unknown
auto emission
auto emission
auto emission
auto emission
auto emission
aluminum silicate,
potassium
auto emission
auto emission
auto emission
flyash (metal rich)
Comments
embedded in
embedded in #22
numerous spots with
stronger fluorescence
-------
204
Appendix Table £3. Particles observed on the middle portion of stage 7 of the Andersen 8-stage impactor from the
Trout Farm location, November 13, 1973
Particle
Elements
Particle Type
Comments
Effective Diameter
No.
1
2
3
k
5
6
7
8
9
10
11
12
13
ll*
15
16
17
18
19
20
(Mm)
»t.5
0.2
0.2
0.3
0.5
0.2
0.2
0-3
0.5
0.15
0.3
__
0.2
0.1*
0.1*
0.2
0-3
0.8
0-5
Major Moderate
—
—
—
—
—
—
Pb
S
S
—
S
Pb,Br P,Cl,Si
Pb
„
S
S
Pb
—
—
Trace
Pb/S
Al,Pb/S,K,Ca,Fe
—
—
—
Pb/S
Pb/S,Al/Br,Si
Al/Br,Cl
Al,Mg
—
Pb/S
Al/Br,Si
__
Al/Br,Si,Cl,Ca,Fe,Zn
Pb/S,Al/Br,Si
A1,K
A1,K
Al/Br,Si,Cl
—
--
auto emission
flyash (?)
flyash (?)
flyash (?)
flyash (?)
flyash (?)
flyash (?)
flyash (?)
auto emission
flyash (?)
auto emission
—
__
auto emission
auto emission
soot
soot
auto emission
auto emission
auto emission
selected area of #1
selected area of #1
Note: Particles 2-15 associated w/#l - Flyash (?), distinct spherical morphology
Appendix Table 2k. Particles observed on the middle portion of stage 7 of the Andersen 8-stage impactoi from the
Trout Farm location, November 13, 19 f 3
Particle
Elements
No.
Effective Diameter
(urn)
Major
Moderate
Trace
Area with relatively high Pb/S peak, selected areas and embedded particles
1
1.8
2 1.5
3 1-1
k
5
Si S Al,K,Ca,Fe,Zn,Pb
Ca,Si,S Fe,Br Pb
Pb,Br,Ca Si,Fe,Zn
Pb,Si,Br/Al Fe,Zn,K,Ca Cr,Mn,Mg
Pb,3i,Al/Br Cl,K,Ca,Fe,Z,n Ti,Mn,Mg,P
Particle Type
quartz
flyash (Ca rich)
auto emission
auto emission
auto emission
CommBnts
particle
particle
particle
area
area
Note: Matrix area made up of small particles vhich appear spherical.
Area with relatively low Pb/S near center of spot
Si,S K,Ca,Fe,Pb,Zn,Al Ti,Mn,Cl,Br
0.4
0.1*
0.2
0.6
0.7
0.5
0.6
0.1*
0.6
0.1*
0-3
area fluorescence
-------
205
Appendix Table 25.
Particle
Particles observed on the outer portion of
the Trout Fa-m location, November 13, 19Y3
Elements
the backup stage of the Andersen 8-stage impactor from
Particle Type
Comments
Effective Diameter
No.
1
2
3
it
5
6
7
8
9
10
11
12
13
lit
15
16
17
18
19
20
21
22
23
2U
25
(mn)
15-0
n/a
8.6
it.l
5-1
2.1
u.s
5-1
3A
8.5
11.2
18.6
»t-5
2-3
10.2
8.5
7-6
11.0
2.8
4.2
6.8
5-9
lt.lt
5-0
6.8
Major
S
—
--
Si
Si
Si
S
—
Si,Al
S
S
S
Si
Si
S,Pb
—
—
S
Ca,Al
S
Si
—
Si,Al
--
--
Moderate
Si
—
S,Pb
Al,K,Ca,Fe
Al,K,Ca,Fe
--
Al,Fe
S,Pb
K
Pb
Pb
K,Ca
—
—
S
S,Pb
Sl,K,Ca
Ti,Fe
Al
S,Pb
K
S
S,Pb
Trace
„
—
Ca,K,Cr/Nd,Fe,Cl
—
—
—
Ca
Si,K,Cl
Fe
Al,Si,K,Ca,Mg
Mg,Si,Cl,K,Ca
Mg,Si
Fe,Al
K,Fe
Cl,Ca
Cl,K,Si,Mg
Cl,Si,Mg
Mg,Fe
Mg,Si
Pb,Si,Mg,Fe
K,Ca,Fe,Mg
Br,Cl
Fe
Al,Cl,K,Mg
K,Mn,Fe,Si,Al,Cl,Mg,K,Ca
auto emission
—
auto emission
clay
dlay
quartz
clay
auto emission
aluminum silicate, potassium
auto emission
auto emission
auto emission
quartz
quartz
auto emission
auto emission
auto emission
auto emission
flyash (Ca rich)
auto emission
clay
auto emission
aluminum silicate, potassium
auto emission
auto emission
Appendix Table 26.
Particles observed on the middle poition of the backup stage of the Andersen 8-stage impactot from
the Trout Farm location, November 13, 19Y3
Mo.
1
2
3
It
5
6
7
8
9
10
11
12
13
lit
15
16
17
18
19
20
21
22
23
2>t
25
26
27
28
29
30
31
32
33
34
35
36
Particle
Effective Diameter
(UP)
13.2
18.7
13-4
2.2
3-1
5-9
7-4
4.4
5-9
k'.O
5-9
15-3
2.7
6^4
20.6
6.8
1.4
5~4
1-9
3-9
3-9
5-3
4.4
4.8
3-8
6.6
5.4
5-9
1.8
2.0
Elements
Major
__
--
--
Si
Si
Si
—
--
Ca
—
—
Si,Al
--
__
Si
Si
__
Al
Si
—
—
--
--
—
Si
—
Si
Si
--
--
--
—
--
Si
Si
Moderate
S
S
S
Al,K,Ca,Fe
Al
K,A1
S
S
Si,Fe
S
Si,S
K,Ca,Fe
S
S,Si
Al,Ti,Fe
Al,Fe
S
__
Al,Ca
Si,S
S
Si,S
Si,S
Si,S
Al,K,Ca,Fe
S,Si
—
Al,K,Ca,Fe
S
—
--
S
S
Al,K,Ca,Fe
Al,K,Ca,Fe
Trace
Pb,Cl,Ca,Al,Si,Fe.Zn
Si,Al,K,Ca,Cl
Pb,Si,Fe,Mg,Al
S
Ca
--
Pb,Cl,Si,Al
Pb,Si,K,Fe
S,C1,K,A1
A1,C1,K
Cl,Fe,Zn
S
Ca,Si
_.
K
K,Ca
Pb,Si,Cl,K
Zn
Pb/S
Al,Fe,Cl
Pb,Si,K,Ca,Fe
Cl,Ca
Cl,Ca,Fe,Zn
Pb,Zn,Fe,Mn,Co
—
Pb,K,Cl,Hg,Pb,Mn,Fe
--
—
Si,Pb,Al
Pb/S, Si
Pb/S, Si
Si
Si
—
--
Particle Type
auto emission
auto emission
auto emission
clay
clay
aluminum silicate, potassium
auto emission
auto emission
mineral calcium
auto emission
auto emission
clay
auto emission
auto emission
clay
clay
auto emission
contaminant (not used)
aluminum silicate, calcium
auto emission
auto emission
auto emission
auto emission
auto emission
clay
auto emission
quartz
clay
auto emission
auto emission
auto emission
auto emission
auto emission
clay
flyash (Si rich)
Comments
-------
206
Appendix Table 27.
Particles observed on the center portion of the backup stage of the Andersen 6-stage impactor from
the Trout Farm location, November 13, 1973
Particle
Effective Diameter
Ho. (urn)
1 4.7
£ 2.8
3 7-1
1* 2.3
5 5-7
6 5-5
7 9-3
8 10.1*
9 3-0
10 5.0
11 3-0
12 6.0
13 9-6
Ik 6.2
15 7-0
16 3.8
17 9-3
18 6.0
19 19-0
20 9-3
21 6.7
22 3.2
23 3-7
2k 5.9
25 3-5
26 2.9
27 5-0
28 7-7
29 2.8
30 9-0
31 9.8
32 2.8
33 3-8
Elements
Ma.lor
Si
Si
—
Si
—
Si
—
—
Ag
—
P,Fe
Si
Si
Si,Ca
Si,Al,Ca
—
S
Ca,K,Cl
S
—
Si,Al,Ca
P
Si
__
Al,Si
—
Ca,Al,Si
_-
S
Pb
Si
Moderate
K,A1
Al,K,Ca,Fe
S,Si
—
—
Al,K,Fe
S,Si
S
—
S
S/Pb
Al,K,Ca,Fe
A1,K
Al,Fe
Fe
S
K,Ca
S,Si
Si
S
Fe
Si
—
S,Si
Ca
S
Fe
S
Si
Al/Br,Cl
Al,Ca
Trace
—
I
—
Si,S
—
K
Si
Si
Si
__
Si
—
Fe
—
--
Si
—
—
--
Si
--
—
--
K
Fe
Si
—
Si
—
--
Fe
Particle Type
aluminum silicate, potassium
flyash (Si rich)
auto emission
quartz
auto emission
aluminum silicate, potassium
auto emission
auto emission
(Contaminant)
auto emission
mineral phosphate
auto emission
clay
aluminum silicate, potassium
aluminum silicate, calcium
aluminum silicate, calcium
auto emission
sulfur mineral
unknovn
auto emission
auto emission
flyash (Ca rich)
mineral phosphate
quartz
auto emission
auto emission
Al,Si,Ca
auto emission
Al,Si,Ca
auto emission
auto emission
PbBrCl
clay
Comments
possibly tire particle
-------
207
^
U)
1
UJ
e
3
^
o-l
f-
ON
iH
|
VO
J
O
-s.
OJ
?5l
5f
3 2
w 3
s
C M
£
4)
W
•§
«
•H
O
10
g
V
a
1
•H
O
O 0
£" a
d vH
1 0
to in
Bl
co 5
H -*-4-moo t^HvoH
00 CM -4-
VOC^ ON VOJ- VO ITVCM IT
VO -* CO CM IA_* O l/N t— OO OO t--
l/N H l/N HO*
OOOD O -=f VO OO t^ rH O'
iTvO-^i-l t— CM OO CM VO OOVO O
CM CM 3
. Jf> jo) ja
*• 5 &
Q we £3
o
t<
1
It
5
i a
3 S
+> -d
S! 45 8
•H > C,
*->«)«)
o -8 a
» O
Z 1 °
4) .«) L,
•9 a>
§ 1 i
U 43 9
B 10 a
II K II
d" J» c
0
t.
s
i
«J
«
J!
•H
+J
O
«
Vi
V
*-,
o
§
«
n
*
-------
208
3)
,1
T
u)
E
CO
r™
1
m
f-
H
g
rH
s
3
o
rH
&
s a
us 3
w
1 1
I
01
_r>
O
tl
H
O
0
H
E
1 „
O 0
3 o
CO W
0
N -a
•H £3
CO <5
?
H
Qj
X
•i~i
•T1
|
r-
c
H
c.
i".
o
•H
+J
£
jf«j
rH rH
O rfl
•H
(J
•3
t)
1 **
\
romg ^fonos^
^S ^rH "^
• 1 •
OJ * rH CM
CO03 ^O
CM H CM l/\
^t O\ ro ^O vo
OOrH OOO LT\ I H IT*
rH
O O t-^ t^- rH ( H
irvirHiTv oniHOOi onirHon
rHVO -*
t~-rn^tco
CM
\O vo ^t f--^ on
~=t O oo oo on o CM t*-
co O CM oovo c^ lf^ t\J on
t~VO .-t rH IA H \O r-l - i C\J flit)
no no rH i/>
H a\ CM ro vo m t— CM vo
VO CO CO O ^O OO t*~ LTN »^~ PO CM O
r^l rH H
-4-O -=t C—t^ H LfNCTxCO
^ CM CM CM
fcy to> to
4) fc.
*" T1 B
cf ^ 8
H rH rH
OOCM H
IT\ OO-* OS
CM LTv
HVD
• I
OJ * rH
CO CO VO
C\J H CM WN
O C7\ O\
^t H -=t l£N
ONO CO
cnHm,j
rH\O -*
t— on-* co
ru
HSO t-
^OL^O
rOCM -*
ITN on O1) rH
IT>LTN rH
ITS OO t*~ VO
VO VO
on
co CM en
J- CM vovo
rH C-
JU
ly to Q
M co c W
3
o
H
1 £
.. 9
8 £
3 c 'u
•rj O
+3 T3 ^
p al a) *ri
> T* > -P
^ t* M O
•*-* 4J 4) 4)
O T3 W 4-i
0) ^ V-
c^ -o o o*
t c: cf
W W U
-------
209
11
I
V)
1
CJ
1
V
b
H
en
r-
1
vo
H
O
OJ
f)
5 to
10 H
§ H
CT
"2 §
£
n
(A
O
CO
s
1)
0)
ji
1.
o o
& 1
1 c
3 o
CO (A
0> 0)
•H g
CO
CD
1
§
(7
•is
w
t~-co in in in _j- _f H r-
incMOoo ^HOQ vooj^tc^
<*"t t*- c\i ON CM in
tH H
t— ' t-
• 1 •
CM H CM
CO CO CO CO O\ O O
r-iir-IH inocoin inirHin
H
•1 , 1
in i H in
O 1 O
. i .
O O ^t -* CMCO OJ
voiHvo iniHin voojinH
CO
CO f~~ VO CO CO VO ^* VO
J-HCTVCO .=»• 1 H-* VOCMVOCJN
ion in COCM in om vo
VO CO CM CM -4- CM O\ J- r— rH O\ CM
H CO CO VO
CM CM O\ COCM CM H in CO
vocovovo -^H-^t^ inocoin
CO H rH
Q^cQ ff^c^ (S^ad"
Q; ^
J-t rH Qj
Q) 'y "^
•5 "H a
3 --5 4)
o E o
CM CM CM
«C CO VO
u> CM st f--
-J
t- t--
• 1 •
CM 1 H CM
invO t—
^t H ifNCM
CM
H 1 H
in rH in
O 1 O
• 1 •
-* H J-
H CO VO
VO CM t— CM
in CM in
in CM vo t—
HCO
HVO VO
VO CM O CO
coco
i-H
CO CO J-
in CM CO O\
>t-t
> -rl f +J
•H i> l< 0
4^ Q> CD d>
O T) 01 <(H
S 'S ° "
" •§ a; "3
§ § •§ a
E
-------
210
„
1
n
i
'•XJ
e
a
+j
"3
E
on
ON
H
0
VO
H
V
Si
•M
O
CO
&
Id
co H
§ I
o *
&
V
CO
o
in
H
O
O
14
- CO H O
HOOt— HH-*CO
CM CM -* t—
CO ' CO
H H H
CO ' CO
• I •
O HO
O O O\ Lf\ t~
CM l H CM H O -* t—
CO LT\ Lf\ OO OO
O O CM H CM H CM
co co t^n H vo
O I H O H O CM CM
r-t CM ir\ CM O O\
CMHl/AO H rl UMTv
H
H O t— t— H H on_=t-
CM .*
-* CO LTN CO CT\ -*
HO-*ir\ HOC— CM
H
Jl) 1
5 3
•p t3
£
r^t rn
CMCO t-
ro
VO O H
H H UNCO
O CO CM
CM H J CO
H CM
H H H ro
UN ON
O\CO CO
H O UNC—
H CM
^JD
H
-t*
r^,
1 £
t-* y
S |
1 c ^
•H O
•f^ *O P
(p 03 (D -H
^ T* r **
*-t > U o
•*-* 4) 4) 0*
O T3 0) $-<
41 ^3 Vi
•3 a* o
§ § 1 a
S in C co
II II II II
hj
Q* co ^" w
-------
211
So
$
t
CO
a
•p
b
E-i
co
t--
ON
o
H
O
-*
CO
tO 3
o 9
_, H
t
m
O
w
^j
O
o
S!
a
H
g
0)
fl
O
O 0
a If
1 a
'J O
to OJ
I*
N •?>
••H 5
CO fl. tt) jo) Q wi i-
O d OJ H
CO l/N CO
rH O VO t—
VO CO O
H O l/NCO
ON H t—
O O OJ rH
ro I-H ON
H rH VO t—
-=t VO VO
H O t— o-l
rH OJ
-4- ON CO
rH O -4- ON
.4- LTv
O LT\ rH
H O t— C—
Mt
W) 10) Q
(3 CO C tj
rH
??
1
"*
t*
a
i
9
^
S5
•H
•p
'0>
t*
«-.
01
§
9
It
d"
c
o
•H
1
a>
ts
•s
gl
1
M
II
i)
10
1
0)
a)
o
l_J
JS
B
II
c
-------
212
1
1
CO
e
a!
•p
3
S
ro
f-
ON
O
S5
H
S
O
UN
3
C Q
I ^
OJ
1
ul
rH
O
§
8
1
-» g
3 o
to w
JH
o> o>
M -3
•H C
tn <2
ro
ro
rH
f
X
1
I
r—
S
c
£
JC
5
u
T-
(E
•H a
O ja£ #jfa$
M rH Jj
W T) -P
3 "2 H
3 ~H 4)
-, £
Q> aj
- 9
S ^
3 c3 *3
•H O
*O -H flj
-P T5 £
S 3 S 3
•rt > U O
o *tJ to *-«
4> J3 Vi
*-* T3 O OJ
Vn I-»
(U d t-l %H
S w c to
11 it it u
Ju
fa) i) G Qi
-------
213
£
en
i
CO
e
o!
+j
3
S
6-1
m
r-
i
z.
H
•(->
"H
O
SO
d)
CO
$ !3
CO 4
§H
2
!
4)
O
en
•3
w
o
i
H
fi
01
£
1 „
O 0
>•» P«
g -H
i a
3 o
CO 10
fj U. Li. Q^ L. L.. J*^
o w ti M droTatj eforc«
S, " V
-P T3 fl
Ǥ s 0
\o so vo
so c^n on
o o on t—
H
u"s on o
O O OJ H
O 1 H O
-* H f-
O O OJ O
SO 1 so
• 1 •
O HO
f s
€
3. m
'*--• ^i
O ro O ^ +^
H O ro no ^j 8
H *H
3 O "u
-* H SO ^5 f)
OOJ-H tU a) CU -H
5 ? c t;
i« " * 8 «
tti ti rt^ W *^ ^
QWT c9 t; "E ° **
G C AJ
S 5 § §
S co c en
H II II II n
5 h)
E^ w' o) 9
-------
214
•p
3
O
I*
a
§•
Jrf
o
c
o
%
«
«
3*.
§5
Li O
1}
1
4) O
a a
•d l«
B 41
i« "a
O •<
I? ft
fi "!
§ +-1
U CO
41 S
N C
•H Cd
CO fc
oo
41
H
I
X
•H
C
d
£
LI
1
oooo ON cvjvo H OO ml
_* H OsMD OO rH CVJ CO l/s CVI-4- O
OO H OO H t~-
-^ ITN
ooo
2|
HI • *
3l U) 1
03 [
*&
0 b Q to ll
Cftn c W CTco C w Q CQ
0)
1
i
M
n w. «
t-
CV1 VO
o
HVO
VO
H
__,.
OO f-
H
J.
CVI OO
H
CVJ
CVJ CO
CVI ON 00
-J- H ITN ITN
OO^t
H
-4- ITN t~-
OO O CVJ VO
• 1 •
• 1 •
-=f 1 H -*
OH O
-3- CVI ir\O
H CVJ VO
UTi H ^~O O
oo
O 00 h-
_* CVJ ^t ITN
H U~N
t— ir\ os
OOH vo H
jU j,,
fl8 ^^««
1
-------
215
S c
4 £
^ K
t
0 L
O
a
C
o
TJ
g
0>
to
n
03
8
V
0)
35 -
*O gl 3
S| §
H 8
3 §
Sco »rt
vi a
u
H 0)
• «?
H «
A 1
o) CO
J^ 6
a «1
j?^
*-• *< 0)
OH !>
i, (v, g
S o^
™ s
H
•H^O
CO H
vj
m
'w
C'
X
1
1
0
•H
W) i
£
-------
216
0 10
•H Q)
C -rt
CTj O
& 0)
L. 6,
o en
1 3
03
n
H
o
to
O H
Li
o -P »d CQ
"° S. ^ ^
|3 c o C
as -s
H 0)
d H
2? °
to CO
if 1 "3 "8
Oj fe -P -H
S^ a * *
8U1
Oj X!
OH S 0
r-( -H
g ON aJ
g
w > J3
O 0
1) K -H
03 r-l -H
03
tn
IU
H
*M
EH
x d
3 B
c <
K i
5* §
w
rn ro
CJ 1 H CM
CO CO
• 1 •
r-l 1 rH H
ro H ir\ H f- c\l
Illl CMrHCOCO HOCMCM
Illl H
Illl Illl Illl
1 1 f I Illl Illl
CM O f>
till • • • Illl
r-l
^OJ- CM H t— CM
Illl CMHCMin HOCMCM
> > J
D Oj C (^ u ob H us O Oj H ts
0> Li
4> Tj -P
v n c
1 ^r ^^
H H H
CO CO
• 1 •
H 1 H . 1
H H t-
CM H O O
H CM
Illl
Illl
CM O ro
CM HMD ro -— •
H J
^
O\ CO -3- «£
H H-=f l^- U
9
^
> fcj -r-t
C^ Wj fl tg o
to
£H
3 8
r° "
a
o
Ll
V
«
§
C 5
3 « s
? N
4) ft) ft)
TJ 00 *-t
^1 S-.
*o o u
«S t* V*
*O 4> O
w c n
n n n
i" cd-
-------
217
O
" *
CO H
CO
CO
•§
X
ft
^
O V)|
"3 -rt
ffy
O COl
a
0
3 «
M
to
H
o
•H
•P
On
•p
1-4
co
3
n
co
rH
A
£
£
0
E-l
I
M
S
A
-H
^
1
CD
U
•s
TJ
I
•H
03
a
i
l
ca
VO OJ OJ t— C— OJ
ONM3OJON rooroH
H r-l
HH OJ -4-OJ \O OH rn
H H-=f OJ LTN
t— H J"
iiii iiii HOOiro
V001 H (Y10N^|. m^ ^
H OJ H ro
VDH H rHOJ OJ COVO J-
OJHOJLT> OOHJ-OJ HOoorn
H H. OJ
U ht Cl* U 1 *-\ l« I Q
U ^
v< H «>
flj *O +^
•P *^3 C
1 s? v
OJ OJ
OJ
H LPv -3-
VO -^ LA O
C^l
OJ OJ H
OJ H O OJ
LA H
H
C-H J
H O OJ 00
onvo o
WH8?S J £
fc *
OJ O (J\O M 3
H-=f g _ 33
CO C tJ
TH O
T3 -H O
•P -O ?
ft) 0) ft> *H
k. -H V C U
u, h, > t5 ^ S fi
(a of OH D p it> aJr
6 " &
-------
218
IV
5 o
O £{
o
CO
§
a>
a
o
H &
-I
V +5
H U
d i
nco
a a!
C 3 JS
, P 2
OEH ?
&P
J2f
to >
b
8^
CO rH
^
ON
CO
3
.a
13
EH
X
1
ft
**
0 Oil
bO «jl
O tSI
C
Q «H
£ n
2 m
Cu
to
o
•H
H
a.
+>
•3
CO
3
i
CO
"H
£
d
o
33
^ ^
s
I*
«
CJ
-H
o
i!
•H
CO
15 S
1) tf V v4
1,. -H > »- Cl
' i,, ... J" +> 01 «> «>
AWPc- a ° "S «> *H
S3 tO C H j, _£3 V,
£? "2 ° "
01 Ol ^C *H
•a «) o
1 « 1»
£ B in C to
o it it n it
i* * °^
-------
219
OJ
5 1
v< ^
o a
o
.4
to
a
o
I
1
n
to
g
S
1
S5 3
T) «a 1
it |
H c o
•3 o
as -s
01
H 41
~ S
H f
ID CO
JS 5
S fe
1 "
OH 1
j>» rn E
1 ^
5
M >
o
OK
rHVO
CO H
1
Appendix
CI t
to g
O v
c
0-S
•P w
3 «
J
&
u
4
o
f-
V
£
d
CO
1
CO
"P5
^
1
£
d
fi
*3 •"
5 -H
£ tl
•3
t
i
o
|
"v
CO
Stage -Area
I
1
OJ-4 t- CO CO
... . 1 .
H O -4 -4 H > H H
ON CO
i O 1 iH O
ON ON
O 1 H O
ON-4- CO
O O OJ H
T , T
O 1 H O
O
H f"> CO O-4 H HrH on
H 1-1 nin
ON ON O O
O 1 H O H 1 i-l H
O UN H ON -4- ^O OJ iH on
HOI/NU-N ooonoj HOOJOJ
HHCO O-4- l^O O
r-IOt-r- HOOJOJ HlHH
ID . . J«) . te
0) >H
u y -p
•g t) C3
^ ? j
Jt -*
OJ J- t~-
H O -4 .-!•
CO CO
O I H 0
H ¥ C 0
^«9 « « s«
11 f>
-------
220
3 fl
J> o
XI
CM *•<
10
a
o
•d
4)
o
•g
to
e
3
*i 1
tJ tf 9
55 o
al |
82 3
•>4i JS
0>
"3
(1 "
ID CO
- g
•g 3
(Q CO
i.
r? +*
SH 3 ,J3
g to
O t-t 5
fl
p£2 "*
1 £
10 >
cu 55
tO H
•
^
1
X
1
•S4
O (0
3-
oj o
bo ft)
f< &
o to
a
•P n
o
•H
£4
PM
"3
M
•g
*
CO
1
£
|g
g
I
3 a
Z
4j "
s
Tj
V
tS
o
Tj
M
1
•H
to
to
1
i
*>
CO
t- t-
• 1 •
O I H O
-* rH t—
O O OJ O
O\ CO
O O OJ H
IPvrH ON
O O VO OJ
VO CO OJ
O O t— O
rH i — 1
ICO CO
O 1 H O
O t— rH
H O -4- -4
t— rH CO
O O OJ H
-4- CO CO
H O OJ OJ
U) Jjj Cj
H 01C W
|
$
i/N
[^
•
O
VO
o
VO
0
OJ
H
.4-
O
t—
o
o
VO
o
VO
o
VO
o
A
i
3j!
lf\
rH H
O COOJ
VO
1 •
1 H O
CO CO
O CO rH
OJ
1 •
1 H rH
.4-
1 H 0
tr\ H
O COOJ
1 rH O
rH O
O O VO
OJ CO
O-* OJ
l-l f-
O ^O co
Id)
• t •
O 1 rH O
OJ OJ
» 1 •
rH 1 H H
ON Os
O 1 H O
CO-4- .4-
O O CO OJ
r- cvl f-
O O ^t OJ
• 1 •
0 1 rH 0
VO rH t-
O O VO CO
VO OJ H
O O OJ H
f- t~
O I H O
VO rH CT
o o n H
I
am a w
L.
1
ITS
t— H CO
0 C3 OJ
CO CO UN
o o noj
VO CO .4-
O O VO CO
O OJ O
H O CO CO
VO co C--
O O O UN
H
-
VO-4- O
O O -4- UN
OJ H
-4 rH H
• • •
O O CO rH
t~- co CO
O O O CO
OJ H
VO OJ rH
O O OJ rH
VO H CO
O O t---d-
CO-4 ^t
O O rHCO
• j-jrcdr
H
0
fr^
1 S
^ 9
+S a
§ a 3
+J TJ S
Si 43 S 3
5 t! C1J
4J 41 4> 4>
CJ -3 CO V.
1) X> Vi
Vi Tj O l>
V< ^
4, _g V,
-------
221
o a
I*
g*8
0*
o 1 £|
VO ° if
Bj |g
« o
g S
•o aS
g
t)
•s d
m
! i
,
* o "8 S
•2 * 1 ^
9 & 2 V
• S B S
8 9
8 a t]
*« «
-1 a
NO **** f"> u*\ o LT\ co OJ on
r-l H OOVO OJ H t- H H O 03
OJ n H H H
t— ^t CO CM OJ IfNCM ON
OO-4-OJ OIO OOOJO
t^" OO rH -^ c-i CO UN CO ON
O O fOOJ O O O O O OJ O
ITv ITN ^t CM OJ
O 1 H 0 0 C3 H
U>PO m-3-r-lC-- t— t—
OOOOH OOH OIHO
® , °2
OIHO
CO r-f O ON-4" CO CO CO
OO-4-H OO-^t oou^nn
rn rH co f^ ON O^
O O POO O i H OIHO
H. ,
H I H
OJ OJ H rO t-- J- O\
OIHC3 HOOI OO-*OJ
oof « 8 i" ti^u Q ffoT c Q
« y +>
•g -d O
vo vo \o
ONOJ H
HHOCV,
NO rn ON
O O t — (*^
IfNOJ CO
O O C— OO
C3 0^1- H
ifNi-1 ON
O O co on
co CO
• 1 •
OIHO
o o ooco
-* m o
O O ITNOJ
H H
• i •
H * H H ^^
-*^
co J- on .?
O o t-u> §
3
j hi Jj
Huco''>fl 9 ij
i !
^ «"
£
1 ]
*j -o j.
V Co
* If
a> an
III
ki a Ji
S
-------
222
I
g
•0
I
!
!
p
en
O O\
oo t--* t-
i
3** I
r* «
el
ON OO f-
W I H W
-------
223
' I I I I O VD O\ O C\J rj
r~\ r~\ r~\ »H
§
E
LTNIAIJ^CO-^
1
e
OH
«a
O ir\
O
H
IOJ O LT\ LfN (AJ -if Q> O
i i i.iF.3c!Nf-t-- HCOH
i i irocvlfnmr^SSai
1 : \
o
M5
icococoo\t-oj-Ha\
i CVJVOmoJ(M^tir\ir\^t
ra &
p i*
o a
43 >
id
o
73
K 00
X ON
SI
3'
-O R
0)
-P
C
o<
wl
5
3
-------
224
— --^ j-
CVJCOO
ON-
CO
I
d
w
I
•p
IS
rtfi^
t— CVJ ro
Ol LA IT\ LTN .4-
C\J
i ir» ir\ OJ cvj
P • ON
OJ CVJ •
H 01
s ®^^
I H
^L—• O
fl o
O\>-
ON
t— ITM—<
-* CVJ rH
• oo
VO
P
H
rH "
coo-—• roc
•—• O
I
Ml
-^
"
8 -d
c7 S
o o
o o
I
CM
O
-------
225
00
H
ON
CO
VJD
CM
O
a
en
en t—
H r-i
o **
U J-i
a o
0) -P
O O
fl) fi(
I fl
3l §
fl C o5
S W
*S
^ **
° 0
a &
>
*H ,£j
O -P
(0 ^H
O O
^OJ t-
CO ON
ro
OO t>-
l/N LA
OJ rn
(M
^OJ en
t— OJ
I — ) UN
OJ
-4- 1
HI I
I I
-4" CO **Q ON CO
H OJ H UN en
H eAJ J- \£1 J-
H
ON H -4- t- -4-
ON i-l en -4- -4-
H OJ I/N -4" en
en
co cvi en ON c\j
en UN vo j- OJ
l/N H OJ r-\ -4"
VO UN 1 1 CO
• •II-
H •* •*
eAj UN
en i i i ON
• ill*
OJ ^O
en
en
OJ
rH
t—
o
UN
o
<£
o
t—
H
ITS
iH
H
^
O
r-l
t—
O
t^
O
H
H
H
^
ON
O
8
^
o
H
3
H
UN
ON
0
!--
H
CO
rn
CO
vo
€
f_
Sfi
ON
H
UN
en
i-l
I
Ll
o
w o o
*~$
PP H
^-^ CP
7
-------
226
ooo co i
g
al
g
a
&
t-
oo
M
5
c
CJ
o
CM VO
w
41
3
TH
CJ\
S^
§ i
in S
g^j
!*"
ll
*£
0)
p- 0)
•H §J
Cfl +^
r-i CO
£
X
I
'
I H CM ^t
r
SVO
£i
C— t~- CvJ
O CO
fi
til
a
^
t»
o
o
8
-------
227
1
3
PX
&
§
0
CJ
i
I
oo
to
8,
co
0
to oo
Q) t—
•,-f
§ g
H -*
H
t)
O •*
C (H
85
ra o
K |
§fl
H
P^ fl)
r9^ Oj
CO -H
I-t IQ
1 1
X CO
S 0
•H 03
±> ^
id 0)
H -g
CM
-------
228
a
o
to
0)
1
fi
Bj
fc.
•p
2
EH
£
h
5
1
CO
01
cn
.
o
•>H
s
C CO
ON
V H
0
O O
1)
§ ^
fe I*
o
>>•**
oj o
« a)
*!*
£ &
•rH 60
^
r-f CO
05 CO
.
1
1
fi
Ji,
<*
^
tt
C
C
nc
Qj
1
t
H
^"
?J
4"
B
I
«
45
C
*~
J
r
.J!
c
i
^.
c*
^ i i
tO i I
LTv OJ CO PO O VQ O\ CM C\J
-, OJ_^-» OCO- O\ t*- •
g t^- CM i-l H*— '^D O\ OO t—
x_sf_l c^^->. . H-^->' •
ot t— co -4- J- OJ m o co -*
c5 °D Qi ' H t— • rou^ •
u vornoj HCVJt^- H^ON
— -"— i * ^< — • ^^i — i
s-^.r—i y—N. . x— sr— »
VDLTvfn ir\LT\t— OHH
" CO f*O CM CO rn LP\ t^" i — 1 U"\
^,1—1 ^^i— . ^^'— •
i H'OJCO H'-NCQ
__i 1 O\ t~- • CO f*"l •
Q OJ^'H rOH CM
"*J'
^^j
3 : : ^
•^ H o] H
«l ' — '~J
J3I ' —i LrN-4" O- — 'CO
Zjl 1 VO- — • ITNCO •
^1 O CM CM H
g,| •— i - — *— •
i
£ 1 rox — . CM CM O ON O H ir\
v t~— -^J~ O
^ t^--^ H ^- H H O -3 H
ON-— •* r— ON Ox^O t — H LP»
-^0?
^| ^^ r-, r-.
Hi -^"HO OOCM VOVDCM
«j;| HLP\H OJf'Of-i H-3-H
gj
3l ro -^ ^
^-^o1
CO J- •
CO CMjH
VD'H'^!
ir\ O '
UAHM3
,_%, ,
O CO *
^B v_x vo
*— j
r_ ,
-^•'-Na\
1 — •
tl'o °>
H LT\ H
^^ "*
^s3
1—1
rH O ON
CO ro •
— • O
1—1
,_,
Sv'occ2
ONLT\ H
H- — CM
ir\ ON ro
on — ^-4-
, ,
nS^
vo
m^-^'cM
i-l CO •
rH^r-j
-*"?"
L/N [ — •
-4-- 3-
^^, ,
a?H J.
* — -.< 1
, ,
CM -^ •
LT\ _4" ITS
^-"— '
Oj'3'rCM
rt^''?
1—1
no H O
C3NCU •
^-* H
c_j
^ %) ,
H CO VO
ITv CO •
CM LTN ro
oo^^' on
O CM •
^co co
ro c\J m
*«I!) ON C^
-* c6 •
co- — co
t*_
+-J
0
a
to
a>
G
0)
O
TJ
-^
-CJ
S
^
^
•^
"i
0
3
L,
_o
>!
-H
ca
u tn
B C
•H d
O
§ °
§T3
(U
!H CD
O H
(I)
0 G
•H -H
-p
OJ ^
•a o
o
t K
C O
5
ca
J^ -
— J^
^ s
. — '
-------
229
•3
§
1
s
5"
3
o
o
t)
OJ
+5
0
$ !§
a
H ^D
O -
*— •
a
c
I
•^
1
+•
c
H
**"
£
«H
S
•j
c
•j
3
M
C
N
•j
i i r- • i
1 1 UJ O I
««J
H H t~- 1~- t\j OJ on c\j
^X)« ON* on* -3-»
t—co ON irv o u> LTMT\
• — i i — t ,-1' — i , — i
8
CO H OJ uc\ UN O on ir\
Co c — on t — -5 ON co co
i — I r-(i — 1 H1 — • ' — '
X
O ^O IAVQ CO O t~— OJ
aj • oj • OJ • -5 •
t" — c — on t — oo UN vo NO
H
O
O O **O O ON CO i-H
J- • O • UA • H •
ON i-H HO HO OJ OJ
' — ' ' — ' ' — ' ' — '
rtl ^-,
(21 1 1 OH I
Ol i i oj • i
pql •— •
^
a")
t- •
UN ON
•— •
•3^
IT\ VO
1 — '
rH ON
P- •
$*
H H
C-OJ
H •
. j
fi
ON
H •
CvJOJ
a u>
i i
x co
v a
2 e
^^
I
ir\ •
C\J OJ
on co
o •
o\
rn
t^^-i
on UN
-
on
CO L
$5
H on
oj co
t—\O
8"?
VOCO
O on
CU OJ
o
o
8
-------
230
$
"P
to
co
a
o
B
-a
bo fi
0 ^
*
r?^
X
g£
•H -P
•P 0
r-1 P*
&a
1
ID
f>
g
C I I
N] l i
a> IA • m •
fr* U"\ -J" LP\^"
tf I/NVO f- O
OH- CT\ •
co MD ON o\
CM t^~ LT\CO
« H • m •
t*- LP\ VD UN
O H C\J ON
H _=f • O -
CJ rH rH rH O
TJ
d ! !
_,
£1
•
,K| Q \D CO c7
^ CMH cojvj
^ c\[^ oxTn
co ro LT\ t— O
l_l 1— _l
^1 ,_, ^
PQI Q f— | p"> |— |
"^O fT) • f^- •
HI vo U"N vo vo
X
1
S, gj
_g< 31 ro _=t
U>
i H •
I -d- O
t~~*
at a\ o c^-
t— • ro •
^O LT* ~=t UA
UN u'^ ro o
t- • O •
O ON m.-t
ir\ ON -4; LT\
co • vt> •
t^-v£> OJ <^o
•^O f~ rH H
O\ • ro •
H H CAJ ro
( ,
S°
L/N • CvJ •
O H H
(_i
,— ! \Q f\J
O ^N VO •
\5 • oj o
O\co OJ ro
§"? £f*
O O O
VD ro^f
i — > t — i
, — i t—
§•
^, ^
IA VO
CpVD
H H
l_4
8"?
UN£_
IVJ CO
CM •
O C —
ro ir\
rovo
H CM
, ,
^°
CM ro
i_i
o\co
rH •
CM ro
R?
O O
OJ ro
*~^
^
O H
IAOO
t-
rn
•p
u
•d
1
•H
•d
1
o
o
E
0
•«£.
II
*— »
fr~4
M
Ml
-------
231
I !? "ON "ON
g
to
1
1
pt.
.p
3
O
1
^ *>"^
oS I
CO u
1' *
0- §
1 £
-p ^.
•H jj
a ?
£ i
C "1 5
"a 3
« • 8
ID > 3
o o 3
W (3 "
§^ 1
^ B 1
H O)
*!
£
(Jj CM -* -4- NO J- ITN ro J- H H
-. O NO UN OO CM OO O -d" oo
rTH' -4- • OD- UN- CO-
^tt^- UNt-— NDCO CMOO rOO
CMON t— CM CO-^ OO QON
*. NO • OO» LTN" Cv]* O»
" CM -=f UN t— OO_* CM 00 i-l O
t_l L_J l__l k_J t—J
O O oo E — CO O -^t C —
i \^ • f— • -d"" O* -^*
>9| i i ONO? "H "H
*5l 1 I H • CO • NO •
IU| O O O
ml i— i i_« i— «
.
.91 l i C— o" O 'UN ooS
^jj 1 1 NO • M . CT\ •
jpl (H CM r-1 rH O
Rl U~J ' ' ' '
,-, f-. r-.
rtl ' — ' i—"1 NO NO C\j CM O —t
Cl CM ON ON NO t— • -4- • NO •
•5TI O» H« CMUN -^"O t^-CM
OJI H H -* ^£} 1-1 .H "^2 O.0^
OO ON ^4" CO ON ON CM O ' — '
~jt-CO t^oo -4-00 NOOJ t^ •
"' OONO -4-NO -4-UN HCVJ CMCM
• i i i , ii i i__i
03) CM H O O O C~- UNCO ON oo
-Cj ON* Q* oo- O- -4*
fJt ft ON NO CO NO r— CM CM rH rH
1
P.
||
31 00 ^t UN NO t-
"
O
•o
fi
M
+•*
1
O
0
fi
o
II
1—1
J--
rH
-------
232
!?
li,
^
SH
CM
OJ
O •*
R l<
0) O
o +=
tn o
t) id
ti CO
I I
>
-------
233
1
ITN
t^- ON H ON t^ t^ OS
IfN t/N tf\ ON -4"
flj
w
g
H
o
ra
0)
*c1
W
S
3
3
ca
CO l/N 1 J- MD OJ ir\
H 1 on >*Q
W
1
1 1 i I HO on J- l
l 1 l l -^ • —
H
•g
CQ i i ir\ • H • cvi • '
n*> i i o O mi
H ^~> *~s
a
H -N ^
pn
g
•s
-* t—
1 1 1 1 H • LTN •
1 1 1 1 O t— OJ
^ ^ 01
H
1 1 1 H O H H '
111 ~ A A
m ON co oi \o • H oi c— o ON
>J CO ON on * O--?» ON
VJ3 -^ LTN ^O O OJ LTN CO LTN
•s "\ »i ON OO ^" •*•
CO CO ro Cx Ci -O OJ
S H H
i
ip
g
•H
ID
m
1
a
o
1
I
S
-------
-234-
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing!
1. REPORT NO.
EPA-600/9-76-007a
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
DENVER AIR POLLUTION STUDY - 1973
Proceedings of a Symposium. Volume I
5. REPORT DATE
June 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Philip A. Russell (Ed.)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Denver Research Institute
University of Denver
Denver, CO 80210
10. PROGRAM ELEMENT NO.
1AA008
11. CONTRACT/GRANT NO.
R-803590
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research & Development
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final 1/74-6/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
EPA, university, and private researchers conducted a study of Denver's urban plume
during the month of November 1973. The objective of the study was to characterize the
pollutants that cause the appearance of the visible colored haze, the so called
"Brown Cloud", which frequently occurs over Denver during the fall and winter months.
Gaseous and aerosol pollutants, and meteorological parameters were measured
periodically under selected conditions.
In March 1975, a symposium was held to present and discuss the results of this
study. This report, volume I, contains important research papers given at the
symposium. The papers cover local windflow patterns; Lidar observations; Aitken,
cloud, and ice nuclei concentrations; and hydrocarbon analyses.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. cos AT I Field/Group
*Air Pollution
Field tests
*Aerosols
*Particles
*Meteorological data
*Transport properties
*Hydrocarbons
*Optical radar
Denver, Colorado
13B
14B
07D
04B
07C
17H
13. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report}
UNCLASSIFIED
21. NO. OF PAGES
238
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
------- |