PB-218 540
INVESTIGATION OF HIGH OZONE CONCENTRATION IN THE VICINITY
OF GARRETT COUNTY, MARYLAND AND PRESTON COUNTY, WEST
VIRGINIA
Research Triangle Institute
Durham, North Carolina
January 1973
DISTRIBUTED BY:
Natioiia! Technical information Service
U. S. DEPARTMENT OF COMMERCE
5285 Port Royal Road, Springfield Va. 22151
This deeument has been approved fcr public release and sale.
-------
Contract No. 68-02-0624
RTI Project No. 41U-764
INVESTIGATION OF' HIGH OZONE CONCENTRATION IN THE
VICINITY OF GARRET! COUNTY, MARYLAND AND
PRESTON COUNTY, WEST VIRGINIA
PHASE I
FINAL REPORT
JANUARY 1973
Environmental Studies Center
Research Triangle Institute
Research Triangle Park, North Carolina 27709
Prepared for the
Environmental Protection Agency
Research Triangle Park, North Carolina 27711
NATIONAL TECHNICAL
INFORMATION SERVICE
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27709
BIBLIOGRAPHIC DATA
SHEET
4. Title and Subtitle
'• R'fSSif°^.. ,
EPA-R4-73-019
Investigation of High Ozone Concentration in the Vicinity of
Garrett County, Maryland and Preston County, West Virginia
7. Aurhor(s)
Not Identified
9. Performing Organization Name and Address
Environmental Studies Center
Research Triangle Institute
Research Triangle Park, North Carolina 27709
12. Sponsoring Organization Nome and Address
Environmental. Protection Agency
National Environmental Research Center - RTP
Quality Assurance and Environmeiltal Monitoring Laboratory
Research Triangle Park, North Carolina 27711
IS. Supplementary Notes
3. Recipient's Acccseion No.
5. Report Bale
January 1973
6. Performing Organization Rept.
No-
41U-764
10. Project/Task/Vork Unit No.
11. Contract/Grant No.
CPA 68-02-0624
13. Type of Report & Period
Covered .
Final Report
16. Abstracts
A- field measurement program was carried out in August and September 1972 to investi
gate the source of high ozone concentrations in Garrett County, Maryland and Preston
County, West Virginia. • Approximately 11 percent of the hourly ozone concentrations
measured at the Garrett County,' Maryland airport exceeded the 0.08 ppm National Air
Quality Standard. In one episode, the Standard was exceeded.for 26 consecutive hours.
The mean hourly ozone concentration for the study period was '0.057 ppm and the maximum
hourly concentration was 0.119 ppm. High ozone concentrations persisted through the
night; the nighttime mean was 0.055 ppm. Nitrogen dioxide and nonmethane hydrocarbon
concentrations were at or near background levels throughout the study period. It was
concluded that local photochemical synthesis could not account for the observed ozone
concentrations in the study area. Analysis of meteorological data indicated that the
high ozone concentrations were associated with air masses .arriving in the study area
after passing over urban-industrial regions. ; ) '
I
17. Key Vords and Document Analysis. 17o. Descriptors
Air pollution Degradation
Ozone Diffusion
Oxidlzers Winds (meteorology)
Meteorology
Measurement
Monitoring
Sources
Photochemical reactions
Synthesis
17k. Identifiefs/Open-Ended Terms
Ozone concentrations
Precursors
17c. COSATI Field/Group
] 30,
18. Availability Statement
Unlimited
1». Security Class (This
Report)
IINCLASSIPI
20. Security Class
aSffife
UNCLASSIFIED
21. No. of Pages
22. Price
$'OO
USCOMM-DC 14B33-072
-------
ABSTRACT
A seven-week program of field measurements was carried out
between August 4 and September 25, 1972 to investigate the source
of high ozone concentration in the vicinity of Garrett County,
Maryland and Preston County, West Virginia. Approximately 11
percent of the hourly ozone concentrations measured at the
Garrett County Maryland Airport exceeded the 0.08 ppm National
Air Quality Standard. In one episode of high ozone concentration,
the Standard was exceeded for 26 consecutive hours. The mean
hourly ozone concentration for the study period was 0.057 ppm
and the maximum hourly concentration was 0.119 ppm. High ozone
concentrations persisted through the night; the nighttime mean
was 0.055 ppm. Nitrogen dioxide and nonmethane hydrocarbon
concentrations were at or near background levels throughout the
study period. It was concluded that local photochemical synthesis
could not account for the observed high ozone concentrations in
the study area. Analysis of meteorological data indicated that
the high ozone concentrations were associated with air masses
arriving in the study area after passing over urban-industrial
regions.
f-> CD
» M
rs-
36
H rt
A H>
rt n
•*§
n »
I«
Is
*< H-
Srr
•
O. H-
poqjam uotjonpojdw
^ ^odaa :*p jo ^3«4
je peonpcwdai si aSad sn^
-------
ACKNOWLEDGEMENTS
Contract No. 68-02-0624
RTI Project No. 41U-764.
INVESTIGATION OF HIGH OZONE CONCENTRATION IN THE
VICINITY OF GARRETT COUNTY, MARYLAND AND
PRESTON COUNTY, WEST VIRGINIA
PHASE I
FINAL REPORT
JANUARY 1973
Environmental Studies Center
Research Triangle Institute
Research Triangle Park, North Carolina 27709
Prepared for the
Environmental Protection Agency
Research Triangle Park, North Carolina
This project was conducted by the Research Triangle Institute,
Research Triangle Park, North Carolina, pursuant to Contract No.
\f
68-02-0624 with the Environmental Protection Agency. The statements,
findings, conclusions, and recommendations presented In this report
do not necessarily reflect the views of the Environmental Protection
Agency. • -
Many Individuals contributed to the research described in this
report. Principal among these and the areas in which they made their
contribution are:
Walter D. Bach, Jr.
Clifford E. Decker
Harry L. Hamilton, Jr.
• Linda K. Matus
Lyman A. Rlpperton
Thomas M. Royal
James J. B. Worth
Meteorology
Analytical chemistry
Meteorology
Data processing
Atmospheric chemistry
Instrumentation
27711
Meteorology, field
operations
In addition, R. W. Murdoch operated the Environmental Monitoring
Laboratory, S. R. Stilley operated the mobile unit, C. E. Moore
installed the meteorological Instrumentation, and W. K. Poole
and S. B. White provided assistance with statistical problems.
The contributions of these individuals are also acknowledged.
The Board of County Commissioners of Garrett County, Maryland
authorized the use of the Garrett County Airport as a location for
the Environmental Monitoring Laboratory; their cooperation is
appreciated. The active Interest and willing cooperation of
John Kreuzwieser, the airport manager, are gratefully acknowledged.
E. C. Tabor, Environmental Protection Agency, served as Project
Officer; his guidance is very much appreciated.
Donald R. Johnston
Project Manager
/I
111
-------
TABLE OF CONTENTS
ACKNOWLEDGEMENTS
LIST OF FIGURES
LIST OF TABLES
SECTION 1 - INTRODUCTION
SECTIOS 2 - STUDY PLAN :
SECTION 3 - DESCRIPTION OF STUDY AREA . .
SECTION 4 - PROCEDURE
A.I Fixed Ground-Level Measurements
4.1.1 Site Selection
4.1.2 Air Quality Measurements
4.1.3 Meteorological Measurements
4.1.4 Continuous Stirred Tank Reactor (CSTR)
4.1.5 Data Acquisition
4.1.6 Data Processing
4.2 Mobile Ground-Level Measurements
4.2.1 Satellite Locations
4.2.2 Power Transmission Lines
4.2.3 Description of Mobile Unit
4.3 Upper Air Measurements
SECTION 5 - SUMMARY OF OBSERVATIONS
5.1 Fixed Ground-Level Measurements
5.1.1 Air Quality Measurements
5.1.2 Meteorological Measurements
5.2 Mobile Ground-Level Measurements
5.2.1 Satellite Locations
5.2.2 Power Transmission Lines
5.3 Upper Air Measurements
5.3.1 Tethered Balloon
5.3.2 Free Balloon
Page
iii
vi
viii
1
3
5
9
9
9
10
11
11
11
11
12
12
12
12
17
19
19
19
21
30
30
30
35
35
35
TABLE OF CONTENTS (Cont'd)
Page
SECTION 6 - OCCURRENCE OF HIGH OZONE CONCENTRATIONS 39
6.1 Frequency of Occurrence 39
6.2 Episodes 39
6.3 Horizontal Extent . 42
6.4 Vertical Extent 46
SECTION 7 - RATES OF PHOTOCHEMICAL SYNTHESIS, DESTRUCTION
AND TRANSPORT OF OZONE 47
SECTION 8 - SOURCE OF HIGH OZONE CONCENTRATION 51
8.1 Contribution of Power Transmission Lines 51
8.2 Ozone Synthesis . 52
8.2.1 Local Photochemical Synthesis 52
8.2.2 Remote Area Synthesis 53
8.3 Ozone Transport 56
8.3.1 Ozone and Wind Direction 56
8.3.2 Ozone Concentration Changes in Air Masses 60
8.3.3 Air Trajectory Analysis 62
8.4 Interpretation 65
SECTION 9 - CONCLUSIONS 69
SECTION 10 - RECOMMENDATIONS FOR FURTHER RESEARCH 71
APPENDIX A - CALIBRATION PROCEDURES . 73
APPENDIX B - PERFORMANCE CHARACTERISTICS AND OPERATIONAL
SUMMARIES FOR AIR QUALITY MONITORING INSTRUMENTS 81
.APPENDIX C - THE CONTINUOUS STIRRED TANK REACTOR. 87
APPENDIX D - TETHERED BALLOON PROCEDURES 93
APPENDIX E - DISCUSSION OF SYNOPTIC WEATHER FEATURES
ACCOMPANYING EPISODES OF HIGH OZONE .
CONCENTSATION 99
REFERENCES 105
-------
LIST OF FIGURES
1 Geographical and topographical features of
study area.
2 Major pollutant sources In the vicinity of the
study area.
3 Relationship between Research Triangle Institute .
(RTI) study area and 1970 EPA study area. 8
4 Garrett County Maryland Airport site. 9
5 Satellite locations for ozone measurement and
roads travelled. 13
6 Sample page of computer printout. 20
7 Diurnal mean carbon monoxide concentration at
Garrett County Maryland-Airport for August 4-
September 25, 1972. , . 23
8 Diurnal mean nitrogen dioxide concentration at
Garrett County Maryland Airport for August 4-
September 25, 1972. 24
9 Diurnal mean ozone concentration at Garrett
County Maryland Airport for August' 4-
September 25, 1972. 25
10 Diurnal mean sulfur dioxide concentration at
Garrett County Maryland Airport for August 4-
September 25, 1972. 26
11 Diurnal mean ambient temperature at Garrett
County Maryland Airport for August 4-
September 25, 1972. 27
12 Diurnal mean solar radiation at Garrett County
Maryland Airport for August 4-September 25, 1972. 28
13 Diurnal mean thirty-foot wind speed at Garrett
County Maryland Airport for August 4-
September 25, 1972. 29
14 Frequency of occurrence of wind direction 30 ft
above ground at Garrett County Maryland Airport,
August 6 to September 24, 1972. 31
15 Temperature-altitude profiles during ascent
portion of tethered balloon flights. 36
16 Vertical profiles, temperature (T), and relative
humidity (RH) as measured by free balloon release
at 1805 EOT, September 15. 37
vi
LIST OF FIGURES (Cont'd)
Figure Page
17 Period of low ozone concentration at Garrett
County Maryland Airport. 40
18 Three episodes of high ozone concentration. 43
19 Vertical ozone and temperature profile at Point
Mugu, California. 53
20 Mean diurnal oxidant concentration at Mineral
King Valley, California. 54
21 Frequency of occurrence of wind direction with
indicated ozone (0.) concentrations. 57
22 Frequency of occurrence of wind direction with
Indicated ozone (03) concentration, normalized
by the frequency of occurrence of a wind
direction. 58
23 Ratio of occurrences of ozone concentration greater
than mean (0.055 ppm) to occurrences of ozone
concentration less than mean by wind direction. 59
24 Time sequence of 12-hour average ozone concen-
trations at Garrett County Maryland Airport. . 61
25 Trajectories of air arriving at Garrett County
Maryland Airport at the indicated time during
Case 1. 63
26 Trajectories of air arriving at Garrett County
Maryland Airport at the indicated time during
Case 2. 66
27 Trajectories of air arriving at Garrett County
Maryland Airport at the indicated time during
Case 3. 67
A-l Ozone calibration system. 76
A-2 Nitric oxide and nitrogen dioxide calibration
system. 77
A-3 Permeation tube .calibration system. 79
C-l CSTR system for rate measurements. . 90
D-l Segment of ozonesonde strip chart record showing
ozone signal, ozone calibration signals (Io> IG),
temperature, relative humidity, and reference
signals. 96
vii
-------
LIST OF TABLES
Table Page
1 CALIBRATION METHODS AND SCHEDULE 10
2 DESCRIPTION OF SATELLITE LOCATIONS 14
3 SCHEDULE FOR MOBILE GROUND-LEVEL UNIT 15
3a DETAILED SAMPLING SCHEDULE FOR SATELLITE LOCATIONS 16
4 STATISTICAL SUMMARY OF HOURLY AIR QUALITY
MEASUREMENTS AT GARSETT COUNTY MARYLAND AIRPORT . 19
5 DAYTIME AND NIGHTTIME MEANS FOR SELECTED AIR
QUALITY MEASUREMENTS AT GARRETT COUNTY MARYLAND
AIRPORT 21
6 FREQUENCY DISTRIBUTION OF ONE-HOUR AVERAGES AT
GARRETT COUNTY MARYLAND AIRPORT 22
7 OZONE AND METEOROLOGICAL MEASUREMENTS AT
SATELLITE LOCATIONS 32
8 FIFTEEN-MINUTE AVERAGE OZONE CONCENTRATIONS (PPM)
IN THE VICINITY OF 500-KV POWER TRANSMISSION
LINES 34
9 EPISODES OF HIGH OZONE CONCENTRATION (£0.08 ppm
for more than 2 hours) AT GARRETT COUNTY MARYLAND
AIRPORT 41
10 HOURLY OZONE CONCENTRATIONS AT SATELLITE LOCATIONS
AND CORRESPONDING HOURLY OZONE CONCENTRATIONS
AT GARRETT COUNTY MARYLAND AIRPORT 44
11 FREQUENCY DISTRIBUTIONS OF HOURLY OZONE CONCEN-
TRATION AT SATELLITE LOCATIONS AND CORRESPONDING
HOURLY OZONE CONCENTRATIONS AT GARRETT COUNTY
MARYLAND AIRPORT 45
12 CALCULATED OZONE DESTRUCTION RATES AT GARRETT
COUNTY MARYLAND AIRPORT FOR SEPTEMBER 2, 1972 48
13 CALCULATED OZONE TRANSPORT RATES AT GARRETT
COUNTY MARYLAND AIRPORT FOR SEPTEMBER 2, 1972 49
14 ANOVA TABLE FOR POWER TRANSMISSION LINE STUDIES 51
B-l INSTRUMENT PERFORMANCE CHARACTERISTICS 83
INVESTIGATION OF HIGH OZONE CONCENTRATION IN THE
VICINITY OF GARRETT COUNTY. MARYLAND AND
PRESTON COUNTY, WEST VIRGINIA
SECTION 1
INTRODUCTION
In the course of a study of injury to Christmas trees, Environmental
Protection Agency (EPA) investigators were surprised to find oxidant
concentrations at rural sites in western Maryland and eastern West
Virginia frequently exceeding the National Air Quality Standard (0.08 ppm)
during the period May 29-September 28, 1970 (EPA, 1971). High oxidant
concentrations were observed at three sites—the Stony River farm in Grant
County, West Virginia, and the Steyer No. 2 and Weise-McDonald farms in
Garrett County .Maryland. Measurements made with a chemiluminescent ozone
meter at the Steyer No. 2 farm showed that virtually all of the oxidant
at that site was ozone (Richter, 1970). Of particular interest was the
fact that the high ozone concentration persisted into the dark hours;
i.e., ozone concentration did not exhibit the typical diurnal pattern
in which it decreases to near zero at night. Nitrogen oxide concentra-
tions during this period were reported to be near background levels
(EPA, 1971).
The high surface ozone concentration could result from local
synthesis or from vertical or horizontal transport from another location.
Ozone synthesis, in situ, from naturally occurring or manmade precursors
is a possible explanation. However, the reported low nitrogen oxide
concentrations together with high ozone concentrations persisting long
after nightfall argue against this explanation. Advective transport of
ozone from the stratosphere to the surface is unlikely, .given the extremely
high temperatures that would result from the accompanying adiabatic
compression. In contrast, horizontal transport of ozone from a remote
tropospheric source region appears possible. For example, it has been
shown that ozone rich layers occur within elevated but low-level inversion
layers (Lea, 1968). Further, Lea suggested that the high ozone concen-
trations observed aloft at Pt. Mugu, California, could be attributed to
viii
-------
precursors originating In the Los Angeles area. More recently, the trans-
port of photochemical smog from Fresno to the Mineral King Valley of
California has been suggested (Miller, et al., 1972).
From the above considerations, the following hypothesis evolved:
Ozone precursors are released into the troposphere at a location remote
from the study area. Given appropriate meteorological conditions,
the precursors are transported to the study area. During transport
and in the presence of sunlight, ozone is synthesized. The hypothesis
further assumes that by sundown the precursors (which are also
destructive agents) are consumed, leaving high residual ozone concen-
trations which are transported to the surface by mechanical turbulence.
Testing the hypothesis stated above involved accomplishing several
specific research objectives. These objectives were:
1) to verify the high concentrations of ozone previously
found in the Mt. Storm, West Virginia area;
2) to define the horizontal and vertical extent of regions
of high ozone concentration; ..
3) to determine the rates of photochemical synthesis, destruction
and transport of ozone, and
4) to attempt to determine the sources of ozone in the area.
SECTION 2
STUDY PLAN
The study plan described in this section was designed to accomplish
the research objectives stated in Section 1, above.
The study plan provided for a seven-week field program consisting
of fixed ground-level, mobile ground-level, and upper air measurements of
selected constituents of the air and meteorological parameters. The air
constituents measured at the fixed ground-level site were ozone, carbon
monoxide, methane, nitric oxide, nitrogen dioxide, sulfur dioxide, and
total hydrocarbons. Meteorological parameters measured at the fixed
ground-level site were ambient temperature, dewpolnt temperature, solar
radiation, wind speed and direction, and vertical temperature difference.
Mobile ground-level measurements included ozone, wet and dry bulb tempera-
ture, and wind speed and direction. Balloon-borne sensors were used to
obtain upper air measurements of oxidant, ambient temperature and
atmospheric pressure.
Measurement of ozone at the fixed ground-level site was undertaken
to verify the high concentrations of ozone previously found in the
Mt. Storm, West Virginia area, while mobile ground-level measurements
were obtained in order to explore the horizontal extent of regions of
high ozone concentration. Likewise, upper air measurements were made
with the Intent of Investigating the vertical extent of regions of high
ozone concentration.
A Continuous Stirred Tank Reactor (CSTR) experiment (Jeffries, 1971)
was employed to determine the rates of photochemical synthesis, destruction,
and transport of ozone.
Since corona discharge from high voltage electric power transmission
lines was considered to be a possible contributor to ambient ozone
concentration, ozone measurements were made underneath, upwind and
downwind of 500-KV power transmission lines.
A determination of the source of ozone in the area in terms of
location, i.e., local, upper atmospheric, or remote surface layer,
would result from an examination of the concentrations and diurnal
-------
characteristics of ozone and its precursors, nitrogen dioxide and
non-methane hydrocarbons; the rates of ozone synthesis, destruction,
and transport provided by the CSTR experiment; the upper air measure-
ments; and meteorological analysis.
SECTION 3
DESCRIPTION OF STUDY AREA
The study area included Garrett County, Maryland and the eastern-
most portion of Preston County, West Virginia. Bounded on the east
by the North Branch of the Potomac River, it consisted primarily of
a high plateau with an average elevation of 2500 feet above mean sea
level. The geographical and topographical features of the study
area are shown in Fig. 1.
N
Figure 1. Geographical and topographical features of study area.
(Source: EPA, 1971)
-------
Remote from urban centers, the area is decidedly rural in character.
In 1970, Garrett County, Maryland had a population of 21,476, while
Oakland, the county seat, had a population of 2,300. Preston County,
West Virginia had a 1970 population of 25,455.
Host of the study area is forested, with commercial Christmas
-tree growing a principal source of income. In addition, Deep Creek
Lake in Garrett County provides recreation for many people from the
tri-state area of Maryland, Pennsylvania, and West Virginia.
Estimated annual emissions of carbon monoxide, hydrocarbons,
oxides of nitrogen, and sulfur dioxide in Garrett County are 11, 2,
1, and 2 thousand tons, respectively. Estimated annual emissions of
carbon monoxide, hydrocarbons, oxides of nitrogen, and sulfur dioxide
in Preston County are 1, 1, 10, and 54 thousand tons, respectively.*
Figure 2 identifies some major pollutant sources within an 80-mile
radius of the study area.
The relationship between the present study area and the area of
concern in the 1970 EPA investigation Is illustrated in Fig. 3.
* Data provided by EPA.
Figure 2. Major pollutant sources In the vicinity of the study area.
-------
*TI STUDY ABEA
CM STUDY AHEA
Figure 3. Relationship between Research Triangle Institute
(RTI) study area and 1970 EPA study area (circles
with dots denote EPA oxidant measurement sites).
SECTION 4
PROCEDURE
This section describes the procedures.employed to obtain fixed
ground-level, mobile ground-level, and upper air measurements of air
quality and meteorological parameters.
4.1 Fixed Ground-Level Measurements
4.1.1 Site Selection
The site selected for fixed ground-level measurements was
the Garrett County Airport near McHenry, Maryland. The airport
elevation is approximately 2900 ft above mean sea level. Consisting
of a 2500-ft paved landing strip, an apron, hangars, and terminal
building, it is used principally by small private aircraft. The
Research Triangle Institute (RTI) Environmental Monitoring Laboratory
and a 30-ft guyed tower were located to the south of the parking area.
Figure 4 is a sketch of the airport which shows the location of the
Laboratory and tower. The site provided excellent exposure of monitoring
instruments with minimal obstruction to air movement.
0 100 200 300 tOO 500 600 ft
' 3
Figure 4. Garrett County Maryland Airport site.
-------
4.1.2 Air Quality Measurements
Ozone, nitric oxide, nitrogen dioxide, total hydrocarbon,
*
methane, carbon monoxide, and sulfur dioxide concentrations were measured
continuously by instruments housed in the RTI Environmental Monitoring
Laboratory. Non-methane hydrocarbon concentration was obtained by
subtracting the methane concentration from the total hydrocarbon
concentration.
Ozone concentration was measured with a solid-phase chemilumi-
nescent ozone meter and a Bendix Model 8300 flame photometric analyzer was
used to measure sulfur dioxide concentration. Nitric oxide and nitrogen
dioxide concentrations were measured with a Bendix Model 8101-B Chemilumi-
nescent NO-NO -NO. Analyzer. Total hydrocarbon, methane, and carbon
monoxide concentrations were measured with the Beckman Model 6800 Gas
Chromatographlc Flame lonizatlon Detector (GC-F1D). The calibration
procedures used throughout and at the termination of the measurement
program are summarized in Table 1 and detailed in Appendix A. Air for
analysis was drawn from the 17-ft level on the tower through a one-inch
I.D. teflon tube to a glass sampling manifold inside the Laboratory.
Air for each Instrument was drawn from this manifold.
Table 1. CALIBRATION METHODS AND SCHEDULE
Frequency
Parameter
Ozone
Nitric oxide/
nitrogen dioxide
Sulfur dioxide
Total hydrocarbon/
methane
Carbon monoxide
Calibration method
ultra-violet ozone generator
referenced to the neutral-
buffered potassium iodide icethod
Gas phase nitration of nitric
oxide with ozone
National Bureau of Standards
permeation tube
Certified cylinder gas8
(methane in zero air)
Certified cylinder gaaa (carbon
monoxide in zero air)
Zero and span
Every 2 days
Every 2 days
Every 2 days
Every 2 days
Every 2 days
Multipoint
Bi-weekly
Bi-weekly
Bi-weekly
Bi-weekly
Bi-weekly
a Purchased from Scott Research Laboratories
* Since sulfur dioxide Is a quantitative interference in oxidant determina-
tions, it was measured for the purpose of correcting the balloon-borne
oxidant measurements.
10
Performance characteristics and operational summaries for the air
quality monitoring instruments are presented in Appendix B.
4.1.3 Meteorological Measurements
Ambient temperature was measured'at heights of 5.5, 16, .and
28.5 ft with shielded, aspirated thermistors, allowing the determination
of temperature differences between 5.5 and 16, and 5.5 and 28.5 ft.
Wind speed and wind direction were measured at heights of 17.5 and 31 ft
with Bendix Aerovanes. Dewpoint temperature was measured with a Foxboro
Dew Point Measuring System at a height of 7 ft on the tower. Solar
radiation was measured with a Klpp and Zonen solarimeter located on
the roof of the Laboratory.
4.1.4 Continuous Stirred Tank Reactor (CSTR)
Two 72-liter pyrex glass flasks were used as CSTR vessels.
One was exposed to sunlight, the other was kept in darkness. Air flow
through the reactor vessels was 2 tpm. Air for each reactor was drawn
from a common manifold which in turn was supplied with 300 tpm of air
through a 1-incb I.D. teflon tube from a height of 17 ft on the tower.
Solid-phase chemiluminescent ozone meters were used to measure ozone
concentrations at the manifold and at the exit of each vessel. A
description of the CSTR system is included as Appendix C.
4.1.5 Data Acquisition
The signals from all fixed ground-level instruments were
digitized and recorded on magnetic tape for subsequent reduction by
computer. Strip chart recorders were used with all Instruments as a
backup to the data acquisition system, and to provide the operator with
a convenient historical record.
4.1.6 Data Processing
Data were sent weekly from the Environmental Monitoring
Laboratory to RTI at Research Triangle Park, N. C. in the form of
digital voltages recorded on magnetic tape. Computer programs were
developed which edited, applied transfer functions, corrected lag time
and drift, and computed hourly averages, frequency distributions, and
diurnal averages.
11
-------
4.2 Mobile Ground-Level Measurements
4.2.1 Satellite Locations
Satellite locations selected for mobile ground-level measure-
ments are shown in Fig. 5, and described in Table 2. The satellite
locations are arrayed along the circumference of an approximate 12.5-mile
radius circle. The Steyer No. 2 and Vteise-McDonald farms, locations
8 and 1, respectively, in Fig. 5 were selected to provide a link between
the earlier EPA investigation (See Fig. 3) and this effort. The criteria
used in selecting the remaining locations were that they be free of local
obstructions to air movement, that they be at relatively high elevation,
and in particular that they be well exposed to winds from the southwest.
The sampling schedule followed for mobile ground-level measurements
is presented in Table 3. A detailed schedule is shown in Table 3a.
One-hour measurements were made at satellite locations.
4.2.2 Power Transmission Lines
A site north of Accident, Maryland was selected for the
purpose of evaluating the contribution of ozone from high voltage power
transmission lines. At this site, 500-kilovolt power transmission
lines run southeast to northwest across Pud Miller Road (Fig. 5); i.e.
at a right angle to a prevailing southwest wind. At this site the
transmission lines pass over the road at a height of approximately
40 feet. The sampling protocol called for fifteen-minute measurements
taken in the following sequence: underneath the lines, fifty yards
downwind, fifty yards upwind, 140 yards downwind, and 1AO-yards upwind
(northeast and southwest were considered downwind and upwind, respectively).
This procedure was repeated three times, providing a maximum of 15 fifteen-
minute ozone measurements at the power transmission lines for each day
of sampling. The sampling schedule is shown in Table 3.
4.2.3 Description of Mobile Unit
An International Harvester Scout was used as the mobile
unit. It was equipped with a solid-phase chemiluminescent ozone meter
with a strip chart recorder, and a Bendlx Aerovane with a dual-channel
strip chart recorder for wind speed and direction. Mr for analysis
was drawn through an Inverted glass funnel extending 1 ft above the
12
PENMSLVANIA
~T~ " MARYLAND
Mount Storm
Figure 5. Satellite locations for ozone measurement
and roads travelled.
13
-------
Table 2. DESCRIPTION OF SATELLITE LOCATIONS
Location
number
Description
Approximate
elevation(ft)
Zaddock Miller Rd. (called Weise-McDonald farm 2660
in EPA report), at power pole next to Miller
residence
On west side of U.S. Rt. 219, 3.2 miles south 2720
of Intersection of U.S. Rt. 219 & U.S. Rt. 40,
55 yards from road pavement
0.8 miles south of Blooming Rose Church on Frantz 2120
and Blooming Rose 'Roads
4
5
6
7
8
0.3 miles east of Mountain Dale intersection on
West Virginia County Rt. 11
Forman Field north of Terra Alta on W. Va. County
Rt. 3, at entrance lane, approximately 10 yards
off of W. Va. County Rt. 3
4 miles south of Terra Alta on W. Va. County
Rt. 53
0.5 mile south of U.S. Rt. 219 on Blue Ribbon
Road
Steyer Ho. 2 Farm (at EPA site)
2540
2546
2300
2440
2540
14
Table 3. SCHEDULE FOR MOBILE GROUND-LEVEL UNIT
Date
August
September
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 .
27
28
29
30
31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Satellite Locatimm Power Transmission Lines
X
X
X
X
X . .
X
X .
X
X
X
X
X
X
' X
. X
' X
X
X
X
x •
' X
X
X
X
X
X
X
X
X
X
X
X
15
-------
Table 3a.
DETAILED SAMPLING SCHEDULE FOR
SATELLITE LOCATIONS
Location8
Dace
Augus t 4
7 ,-
8
10
13
14
16
17
19
22
23
25
28
29
31
September 3
4
6
7
9
12
13
15
16
Starting
1000 1200
1
5
2
6
3
7
4
8
1
5
2
6
3
7
4
8
1
5
2
6
3
7
4
8
2
. 6
3
7
4
8
1
5
2 .
6
3
7
4
8
1
5
2
6
3
7
4
8
1
5
Time - EDT
1400 1600
3
7
4
. 8
1
5
' 2
6
3
7
4
8
1
5
2
6
3
7
4
8
1
5
2
6
4
8
1
5
2
6
3
7
4
8
1
5
2
6
3
7
4
8
1
5
2
6
3
7
Location numbers correspond to those of Fig. 5.
vehicle roof and connected to the ozone meter with teflon tubing.
The mobile unit ozone meter was calibrated using the procedure listed
in Table 1, above. The power supply for the instruments was four
12-volt batteries in parallel combined with a DC-AC inverter. The
instruments remained on throughout the period of a run; however,
the Scout engine was turned off while measurements were made. A
Bendix Fsychron was employed to measure wet and dry bulb temperatures.
4.3 Upper Air Measurements
Vertical soundings of atmospheric temperature and ozone were
made in the near-surface environment and through the troposphere using
a NOAA-type radiosonde with a Mast Model 730-7 ozonesonde attached as
the sensor. For the near surface measurements, an aerodynamically-
shaped balloon, tethered from the ground, was used to carry the Instru-
mentation aloft. A full description of tethered balloon procedures
is given in Appendix D. For tropospheric measurements, the same
Instrumentation was attached to a 1200-gm rubber ballon and released.
Signals transmitted by the radiosonde were received and recorded on
a 5-inch strip chart recorder operating at a chart speed of 1/2 inch
per minute.
Tethered and free balloon releases were made at the Garrett
County Maryland Airport as weather circumstances, in particular rain
and high winds, permitted. A special effort was made to obtain
both day and night soundings.
16
17
-------
SECTION 5
SUMMARY OF OBSERVATIONS
A summary of observations obtained during the seven-week field
measurement program is presented below.
5.1 Fixed Ground-Level Measurements
5.1.1 Air Quality Measurements
Five-minute average concentrations for the air quality
measurements were printed out by computer. A sample of this printout
*
is given in Fig. 6. Summary statistics for hourly air quality measure-
ments made at the Garrett County Maryland Airport are given in Table 4.
The maximum hourly concentrations of carbon monoxide, nitrogen dioxide,
ozone, and sulfur dioxide were 0.804, 0.038, 0.119, and 0.161 ppm,
respectively.
Table 4. STATISTICAL SUMMARY OF HOURLY AIR QUALITY
MEASUREMENTS AT GARRETT COUNTY MARYLAND AIRPORT
August 4-September 25, 1972
Pollutant
Carbon monoxide
Methane
Nitri^ oxide
Nitrogen dioxide
Nonnethane hydrocarbon
Oxides of nitrogen
Ozone
Sulfur dioxide
Total hydrocarbon
Mean (ppm)
.219
1.521
-.oioa
.007
-.058
-.004
.057
.005
1.452
Standard
deviation (ppm)
.149
.115
.004
.005
0.120
.007
.018
.011
.096
Case count
686
571
1,140
1,140
553
1,140
1,043
945
782
The data acquisition system acknowledges both positive and negative
voltages. An entry preceded by a negative sign should be regarded
as zero concentration.
* The complete tabulation of five-minute values is too voluminous to
present here. A copy has been delivered to the Project Officer, and
interested parties may contact RTI to arrange to see the tabulation.
19
-------
oz
sIsssssSsBsssssslsssascisss! sE'i itssIsssssasssSisisssiSsiti 55
fi HI in 'ill" ' '"
WMMMbWMW •*•»•• MM MM M*t M •> MM •• L* M MM. ob • i
£ •Wf WH •>« •«*•**-*- MMt** W*»»VI*>0«V« M
• «1 *MM •*•• *• *U*MO>* O OJV* »W* WNM M-*[ *»»•«• • *
iliVr.iiii.i'.V.oI^VlViji-k.V.;. -p|p_ Cj
! i
TfTT
i i I i 5*4, »-,«•.«»•-.»*«•
Mils! i h ! ! ! !
ll
: .
:* stsssissssisiiiijriiiissssb 3p- ' , . .
' ' : •' : ' •• • :« -
i . I . ""L0 * * •* *'* •-•-•••• ........... 5
*
,«
siissssississsssissss'isssss F: ———"•-•— p« .— ... 5-
| ' i ! i ' i. I
aiaionaond3H ION
The values for carbon monoxide, methane, non-methane hydrocarbon
and total hydrocarbon concentration shown in Table 4 should be inter-
*
preted with some caution. Of these values,those for carbon monoxide
are believed to be most nearly correct; hence they are reported in
greater detail.
Daytime and nighttime mean concentrations of carbon monoxide,
nitrogen dioxide, ozone and sulfur dioxide are shown in Table 5,
while mean diurnal concentration curves for these pollutants are
presented as Figs. 7, 8, 9, and 10, respectively.
In Table 6 frequency distributions for the several air quality
parameters are presented.
5.1.2 Meteorological Measurements
Mean diurnal curves for ambient temperature, solar radiation,
and 30-ft wind speed for the study period are shown in Figs. 11, 12,
and 13, respectively.
The mean temperature for all hours of the study period was 1S.5°C
(65.3°F). The mean nighttime temperature was 16.7°C, or 3.2°C lower
Table 5. DAYTIME AND NIGHTTIME MEANS FOR SELECTED AIR QUALITY
MEASUREMENTS AT GARRETT COUNTY MARYLAND AIRPORT
August 4-September 25, 1972
Measurement
Carbon monoxide
Nitrogen dioxide
Ozone
Sulfur dioxide
Mean
Daytime8
0.227 (382) c
0.006 (645)
0.059 (591)
0.004 (530)
(ppffl)
Nighttime1"
0.208 (304)
0.007 (495)
0.055 (452)
0.006 (415)
0600 to 1955 EOT
2000 to 0555 EOT
Number in parentheses Is the case count.
* Following completion of the field program, the instrument manufacturer
determined that water vapor permeating the neoprene diaphragm of the
hydrogen regulator could have contaminated the analytical column with a
consequent shift In retention time and error in indicated concentration.
21
-------
Table 6. FREQUENCY DISTRIBUTION OF ONE-HOUR AVERAGES AT
GARRETT COUNTY MARYLAND AIRPORT
August 4-September 25, 1972
Percent of si
Concentration
(ppm)
.0050
.0100
.0200
.0300
.0400
.0500
.0600
.0700
.0800
.0900
.1000
.1100
.1200
.1300
.1400
.1500
.2000
.3000
.4000
.6000
Case count
Ozone
100.00 X
99.81 X
98.64 %
92.43 X
81.84 X
66.80 %
40.58 X
22.23 X
11.26 X
4.56 X
1.75 I
.68 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
1043
Nitrogen
oxides
7.98 X
2.98 X
.70 X
.18 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
1140
unples > stated concentration
Nitric
oxide
0.00 X
0.00 X
0.00 X
0.00 Z
0.00 X
0.00 X
o.ob x
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
1140
• Nitrogen
dioxide
47.46 X
15.35 X
2.37 X
.35 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
0.00 X
o.oo j;
0.00 X
0.00 X
0.00 X
0.00 X
1140
Sulfur
dioxide
25.71 X
11.53 X
4.13 X
2.75 X
2.12 X
1.27 X
.74 X
.63 X
.42 X
.32 X
.21 X
.21 X
.21 X
.11 X
.11 X
.11 X
0.00 X
0.00 X
0.00 X
0.00 %
945
Carbon
monoxide
94.02 X
93.73 X
93.29 X
91.98 X
91.55 X
90.96 X
89.65 X
88.34 X
87.17 X
85.86 X
84.26 X
81.34 X
77.41 %
73.91 X
70.26 X
64.72 X
46.50 X
24.49 X
14.43 X
1.02 X
686
Concentration (ppm)
o o
£§
00 O
CO H*
rT O.
(D
•O O
It (0
fl> 3
S
3
22
-------
0.10
0.09
SO.08
S 0.07
0.06 -
0.05 -
0.04
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day .
Figure 9. Diurnal mean ozone concentration at Garrett County Maryland
Airport for August 4 to September 25, 1972.
0.010
0.009
0.008
a 0.007
c
S 0.006
OiOOS
0.0041 I 1 I I II I I I _,
0000020004000600080010001200140016001800ZOi
Time of Day
100 2200
TOO
Figure 8. Diurnal mean nitrogen dioxide concentration at Garrett County
Maryland Airport for August 4-September 25, 1972.
-------
30.00
25.00
20.00
15.00
10.00
5.00
0.00
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EDT)
Figure 11. Diurnal mean ambient temperature at Garrett County Maryland
Airport for August 4-September 25, 1972.
0.024
0.020
!• 0.016
0.012
0.008
0.004
0.00
I I I I
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day
Figure 10. Diurnal mean sulfur dioxide concentration at Garrett County
Maryland Airport for August 4-September 25, 1972.
-------
12.00
10.00 -
~ 8.00
4J
|
•g 6.00
II
a
3 4.00
2.00
0.00
i _a_
000002000400 0600 08001000 12001400 1600180020002200
Time of Day
OOOC
Figure 13. Diurnal mean thirty-foot wind speed at Garrett County
Maryland Airport for August 4-September 25, 1972.
1.50
0.00
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 00 0
Time of Day
Figure 12. Diurnal mean solar radiation at Garrett County Maryland
Airport for August 4-September 25, 1972.
-------
Chan the mean daytime temperature of 19.8°C. The difference between
the maximum (22.5°C) and the minimum (15.7°C) on the diurnal temperature
curve is also small (6.8°C). .The small range of average temperature
indicates the presence of clouds both day and night. The high nighttime
temperature suggests that strong radiation inversions occurred only
infrequently at the airport.
Mean solar radiation during the study period was 0.417 Langleys.
The average wind speed at the 30-ft level for all hours was
5.9 kt. On the average, winds reached a peak of 7.0 kt at mid-day,
with a lull at 5.2 kt near sunset. Only 23 hours of calm winds (< 1 kt)
were calculated (representing 1.8 percent of the observations) and the
hourly vector average wind speed never exceeded 17 kt. Evaluation of
wind speed observations indicates that the site was well ventilated and
that separation of winds aloft from the near surface circulation by
nocturnal inversions occurred infrequently.
The frequency of occurrence of wind direction by 22.5°C sectors and
the average wind speed for each sector is shown in Fig. 14. Southwest
through northwest winds are clearly the predominant wind directions,
accounting for 59 percent of the 1,139 hourly wind observations. The
next most frequent directions were southeast and south-southeast,
accounting for an additional 19 percent of the observations. Northerly
to northeasterly winds rarely occurred and never exceeded a speed of
3 kts. Southeasterly winds occurred much more frequently and westerly •
winds slightly less frequently than during the previous study (EFA, 1971).
5.2 Mobile Ground-Level Measurements
5.2.1 Satellite Locations
A total of 49 one-hour ozone measurements were obtained at
the satellite locations. Those measurements, as well as the corresponding
meteorological measurements, are presented in chronological order in
Table 7.
5.2.2 Power Transmission Lines
Seventy 15-mlnute ozone measurements in the vicinity of
500-KV power transmission lines were obtained. The measurements are
shown in Table 8.
30
7.1
5.6
0
1 , ,
5
,,,!,,
io i:
, , I , , , , I
Figure 14.
SCALE
Frequency of occurrence of wind direction 30 ft above
ground at Garrett County Maryland Airport, August 6
to September 24, 1972. Average wind speed for each
direction shown at extremity of the radials. Calm
conditions occurred 1.8 percent of the time.
31
-------
Table 7. OZONE AND METEOROLOGICAL MEASUREMENTS AT SATELLITE LOCATIONS (Cont'd)
Date
September 6
Sep tember 7
September 9
September 12
September 13
September 16
Location
5
6
7
8
2
3
4
1
6
7
8
5
3
4
1
2
7
8
5
6
8
5
6
7
Measurement period
(EOT)
• Begin End
1000
1200
1400
1607
1000
1200
1400
1600
1000
1200
1409
1613
1000
1200
1412
1600
1000
1205
1415
1600
1003
1200
1400
1600
1100
1300
1300
1700
1100
1300
1455
1700
1100
1245
1555
1700
1100
1300
1510
1700
1100
1300
1500
1700
1045
1300
1500
1700
Average ozone
concentration
(ppm)
0.050
0.061
0.055
0.062
0.051
0.058
0.061
0.064.
0.020.
0.028
0.039
0.051
0.037
0.039
0.047
0.042
0.059
0.062
0.070
0.075
0.047
0.055
0.058
0.059
Wind
speed
(mph)
8.0
12.3
7.5
6.3
4.8
9.0
9.0
2.0
8.5
6.3
7.6
12.0
10.8
11.3
3.0
7.0
6.3
12.8
18.0
18.0
9.7
12.3
10.8
7.0
Wind
direction
109
115
118
145
200
258
270
180
350
261
350
350
258
268
203
294
261
275
272
285 .
290
'278
285
265
Temperature °F
Dry Wet
bulb bulb
64
69
71
70
66
74
74
77
61
66 .
69
65
65
65
67
66
72
76
74
75
68
68
72
77
55
55
57
57
56
59
60
63
59
60
60
56
65
64
65
65
68
68
68
68
60.
58
60
63
C - Calm
H - Missing data
E * Equipment malfunction
Table 7. OZONE AND METEOROLOGICAL MEASUREMENTS AT SATELLITE LOCATIONS
Date
August 23
August 25
August 28
August 29
August 31
September 3
September 4
Location
2
3
4
1
6
7
8
5
3
4
1
2
7
8
5
6
4
1
2
3
8
5
6
7
1
2
3
4
Measurement period
(EDI)
Begin End
1000
1200
1400
1610
1009
1200
1410
1610
1000
1200
1405
1600
1000
1214
1420
1600
1005
1220
1400
1600
1000
1205
1400
1600
1000
1200
1400
1600
1100
1300
1500
1700
1504
1700
1100
1300
1500
1700
1100
1300
1510
1700
1100
1305
1500
1700
1300
1500
1700
1100
1300
1500
1700
Average ozone
concentration
(PPm)
0.058
0.074
0.081
0.084
E
E
0.070
0.079
0.031
0.041
0.049
0.055
0.058
0.073
0.058
0.077
0.069
0.077
0.093
0.082
E
0.041
0.053
0.068
0.022
0.026
0.044
0.027
Wind
speed
(mph>
6.8
15.0
12.0
C
6.8
M
15.3
9.3
2.0
7.8
2.8
4.0
8.3
6.0
5.5
3.0
7.5
3.3
5.8
8.3
3.0
2.5
4.0
6.0
3.3
Wind
direction
280
274
246
C
304
292
293
258
263
288
286
334
. 321
295
157
171
154
153
245
. 280
268
M
300
310
315
Temperature *F
Dry Wet
bulb bulb
74
79
80
80
83
. 78
68
67
70
70
71
75
73
72
72
74
73
75
64
64
67
59
64
62
60
67
71
69
70
70
71
63
61
62
62
64
64
64
65
63
64
62
63
62
62
64
56
57
56
55
C - Calm
M - Missing
E - Equipment malfunction
-------
I
s
1
rt
|—
*-
•s
g
e
gg
o o
o o
' o o
go
w
r-> QD
0 0
00
Ui W
o o
o o
on u>
V)
•§
rt
O
§•
o
r»
00
5
UI
N
• K) .
sss
M QD 00
O O O
O O O
-^ * w
o o o
Isl
00 0
o o o
""*
o o o
o o o
•*J ON •"»!
W 00 f->
j£ £ £•
s 1 1
trt rt
W to
Ui
N U> U7
W H1 O
o o • o
w *• 10
r-> OD Ul
.
ooo o o o ooo
O "J H* *• OO GO OD Ul O
OOO OOO OOO
ooo ooo ooo
Ul Ul Ui Oh Ui Ui tOODQp
H* --J O Uf On to OO^OK
ooo ooo ooo
ooo ooo ooo
Ui Ui J» ^ *• ** *J *O OD
»>totO OtOtO UODO
ooo ooo .000
ooo ooo ooo
Vt ijl »• ON Ul Ul ~*4 QD OO
^ ro oo ui*.(- OOO^M
ooo ooo ooo
'ooo ooo ooo
«• ui ui a* u» ui ooa>u>
ODODf-' Q» ^J M \OUiO
a
p
rt
o
1
i
l"
&
1
at
s
1
a
g
Hi
rt
Ui
o
H
o.
a
o
1
w
rt
'•s
R
1
g.
1
i
1
1
S1
n
rt
H-
g.
H
cr
00
^-0
(H M
II
Si
Ul S
A™
si
w C^
po w
H O
S N
i!
en o
en S
M O
o H
f g
Is
en
1
M
Z
g
5.3 Upper Air Measurements
The ozonesonde used for upper air measurements was calibrated with
known ozone concentrations Immediately before and after tethered balloon
flights. For reasons not Immediately apparent, .the pre- and post-flight
calibration curves varied significantly both in scale and slope. Thus,
it is questionable whether either the absolute or relative ozone (oxldant)
concentrations are valid. Accordingly, only the temperature data are
reported.
5.3.1 Tethered Balloon
Thirteen tethered balloon flights were made between
September 11 and 22. Flights were made in early morning, at mid-day,
near sunset, and near mid-night. The maximum altitudes attained varied
from 400 to 900 ft, depending upon weather and balloon conditions.
Surface-based temperature inversions were found in some of the
early morning, early evening, and night flights. Distinct temperature
inversions aloft were not found within the layer sounded by the tethered
flights. Daytime measurements show a nearly adlabatlc condition when
averaged over the sounding depth. The temperature data obtained are
plotted as vertical profiles In Fig. 15.
5.3.2 Free Balloon
Of four free balloon releases, one provided useful data.
The temperature and relative humidity profiles from the surface to the
500-mb level (-18,000 ft) for that release are shown in Fig. 16. An
inversion based at about the 800-mb level (-6,400 ft) is clearly shown.
35
-------
9C
-
Q f*
€ --h C
-
a" »i
P D- o
n o H»
Sffjf
c o
H- H--O
H* CD O
A M
• Ut rt
Srt
-
Altlttid* (f««t)
»— W W » Wl O* *J
g 8 g 8 8 8 g
T 1 1 1 1 1 T
400
500
?
-------
SECTION 6
OCCURRENCE OF HIGH OZONE CONCENTRATIONS
This section is concerned with the frequency of occurrence of high
ozone concentrations, episodes of high concentration and the horizontal
and vertical extent of regions of high concentration.
6.1 Frequency of Occurrence
The frequency distribution of hourly ozone concentrations at the
Garrett County Maryland Airport during the period August 4-Septeober 25,
1972 was presented In Table 6, above. Examination of that distribution
reveals that the National Air Quality Standard for photochemical oxldants
of 0.08 ppm was exceeded approximately 11 percent of the time.
The mean and nm-nimnr. hourly ozone concentrations observed during
the present study were 0.057 and 0.119 ppm, respectively. These values
are remarkably close to the 0.059 ppm mean and 0.13 ppm imnrimum seen
in the special two-week ozone measurement program conducted in August
and September 1970 (Richter, 1970). It Is believed that the findings
of this study verify the high concentrations of ozone previously found
In the Mt. Storm, West Virginia area.
The only time that ozone concentration reached zero at the Garrett
County Maryland Airport was between 0530 and 0550 on September 22. This
decrease was coincident with a sharp Increase In sulfur dioxide and
nitrogen dioxide concentrations (Fig. 17). Wind direction at this
time was north-northeast. Decreases In ozone concentration associated
with increased sulfur dioxide and nitrogen dioxide concentrations were
seen on several occasions; however, only on September 22 did ozone
concentration fall to zero.
6.2 Episodes
It is of Interest to identify extended periods of tine, or episodes,
of persistent high ozone concentration. For thin purpose, episodes
were defined as periods in which ozone concentration at the Garrett
County Maryland Airport was equal to or greater than 0.08 ppm for more
than two hours. Pertinent dates and tines and episode durations are
given in Table 9. Eleven episodes occurred during the field measurement
39
-------
Septerter 22. 1972
0400 0420 0440
0620 0640 ' 0700
Figure 17. Period of low ozone concentration at
Garrett County Maryland Airport.
40
Table 9. EPISODES OF HIGH OZONE CONCENTRATION
(>0.08 ppm for more than 2 hours) 'AT GARRETT
COUNTY MARYLAND AIRPORT
August 4-September 25, 1972
Begin episode
Date Hour (EDT)
August
September
10
13
14
15
18
.31
' 1
2
11
12
1800
1300
0500
1400
0000
1200
1100
1400
0200
2200
2200
End episode
Date Hour (EDT) . Duration (hours)
August 10
14
14
15
18
31
September 1
3
12
13
2200
0000
1100
2200
'0700
1800
2000
1800
0300
0200
0800
5
12
7
9
8
7
10
5
26
5
11
41
-------
program, the longest of which persisted for 26 consecutive hours
beginning at 0200 EOT on September 2. During the August 31 episode
ozone concentration was also measured at satellite locations. At one
location the concentration approached 0.08 ppm and it exceeded that
value at two others.
The time course of three episodes is graphed in Fig. 18. Corresponding
carbon monoxide concentration data are available for two of the episodes;
they are also plotted in Fig. 18. The Interpretation of the low carbon
monoxide concentration data is difficult and the difficulty is compounded
by instrument uncertainties (see footnote, p. 21). Recall also that
the minimum detectable carbon monoxide concentration is 0.1 ppm.
6.3 Horizontal Extent
The mobile ground-level measurement program provided ozone data for
eight satellite locations. Hourly ozone concentrations at the satellite
locations and the corresponding concentrations at the Garrett County
Maryland Airport are given in Table 10. The data shown in Table 10 were
not used to test for statistically significant differences between
satellite location and airport measurements since hourly ozone measure- ,
meats within days were tested and found to have high autocorrelation
coefficients. Thus, the assumption of statistical independence which
underlies the appropriate test procedures, e.g., paired t-test, or
Wilcoxon signed-rank test, was violated.
The data of Table 10 were used to prepare the frequency distribu-
tions shown In Table 11. Note that the percent of observations equal
to or greater than 0.081 ppm was 8.0 and 8.5 for the satellite locations
and the airport, respectively.
It is concluded that the region of high ozone concentration extends
at least the 12.5 miles from the airport to the several satellite
locations. This is not surprising in light of the 1970 oxldant measure-
ments which showed high concentrations at such diverse locations as
Stony River farm in Grant County, West Virginia and the Steyer No. 2
and Welse-McDonald farms In Garrett County. Maryland (EPA, 1971).
Ozone concentration (ppm)
o o
ro ._
O
O
Carbon monoxide
concentration (ppm)
o o o o
. •. .
o H- ro w *-
o
t-f>
sr
^*
•5-
(B QQ
42
-------
Table 10. HOURLY OZONE CONCENTRATIONS AT SATELLITE LOCATIONS
AND CORRESPONDING HOURLY OZONE CONCENTRATIONS
AT GARRETT COUNTY MARYLAND AIRPORT
Satellite
locations
0.058
0.074
0.081
0.084
0.070
0.079
0.031
0.041
0.049
0.055
0.058
0.073
0.058
0.077
0.069
C.077
0.093
0.082
0.041
0.053
0.068
0.022
0.026
0.044
. 0.027
0.050
0.061
Ozone concentration (ppm)
Garrett County Satellite
Maryland Airport locatiqns
0.050
0.060
0.065
0.072
0.053
0.049
0.041
0.046
0.053
0.053
0.054
0.072
0.075
0.081
0.073
0.091
0.091
0.087
0.064
0.074
0.077
0.033
0.045
0.061
0.054
0.056
0.057
0.055
0.062
0.051
0.058
0.061
0.064
0.020
0.028
0.039
0.051
0.037
0.039
0.047
0.042
0.059
0.062
0.070
0.075
0.047
0.055
0.058
0.059
Garrett County
Maryland Airport
0.065
0.068
M
0.057
0.061
0.065
M
0.040
0.047
0.055
0.059
0.059
0.068
0.063
M
0.074
0.073
0.070
0.056
0.062
0.076
0.073
M - Missing
44
Table 11. FREQUENCY DISTRIBUTIONS OF HOURLY OZONE CONCENTRATION
AT SATELLITE LOCATIONS AND CORRESPONDING HOURLY OZONE
CONCENTRATIONS AT GARRETT COUNTY MARYLAND AIRPORT
Percent of observations
> stated concentration
Ozone concentration
(ppm)
Satellite locations
Garrett County
Maryland Airport
0.081
0.075
0.070
0.064
0.059
0.055
0.050
0.041
8.0
16.0
24.0
30.0
42.0
58.0
66.0
80.0
8.5
14.8
31.9
44.6
59.5
70.2
82.9
93.6
45
-------
6.4 Vertical Extent
While it was not possible to define the vertical extent of the
region of high ozone concentration, the following can be inferred
from meteorological analysis: The lack of well defined nocturnal
surface-based temperature inversions, the .small diurnal temperature
range, and the relatively constant wind speed from day to night
suggests that the air masses over the base station were relatively
well mixed in the lower levels. As a consequence, the assumption
that the high concentrations of ozone are not restricted to a shallow
surface layer appears warranted, to fact, higher concentrations of
ozone above the surface seem probable since vegetation, soil, and
all structures and obstacles on the surface form an ozone sink.
SECTION 7
RATES OF PHOTOCHEMICAL SYNTHESIS, DESTRUCTION
AND TRANSPORT OF OZONE
Due to difficulties not yet resolved, the rate of ozone (0.)
synthesis (S) could not be determined from Continuous Stirred Tank
Reactor data. Examination of ambient 0, concentrations, however, indi-
cates that synthesis and sunlight have little to do with maintaining
high concentrations such as occurred on September 2. It was possible
to determine the rate of destruction (RD); the rate ranged from approxi-
mately zero to an isolated high of -97 x 10 ppm min , with an average
of approximately -19 x 10 ~ ppm min (Table 12). With an ozone concen-
tration of 0.9 ppm, the EK D value (average value of rate constant times
-3
concentration of O.-reactive substance in ppm) is approximately 2 x 10
_1 •*
min . The absolute rate of destruction is approximately twice that
observed in rural piedmont North Carolina (Jeffries, 1971), but the 0^
concentration was approximately five times as high. The rate of
destruction normalized for ozone concentration was -2.1 x 10 ppm min
per ppm (-2.1 x 10 min ). The normalized rate determined in rural
piedmont North Carolina for a period in January 1971 was -5.7 x 10 min
(Jeffries, 1971). Thus, the normalized rate of destruction observed in
this study area was well within the range expected for naturally occurring
destructive agents of local origin.
4 -1 -1 -5
With an illustrative rate constant of 10 I mole sec (-3 x 10
ppm min ), the KD values listed in Table 11 represent concentrations of
0,-reactive material of 0.01 to 0.08 ppm. The normalized RD values are
characteristic of natural rates and do not suggest an "active" pollution
system (Jeffries, 1971). The 0,-destructive material could be of natural
origin and recently mixed with O.-rich air.
The corresponding rates of change in 0, concentration due to transport
(RT) (for daylight hours, transport plus synthesis) are shown in Table 13.
A negative daylight-hour RT value would require the mixing of 0,-poor air
with air previously sampled. For an Initial 0, concentration of 0.08 ppm
-------
Table 13. CALCULATED OZONE TRANSPORT RATES AT GARRETT COUNTY
MARYLAND AIRPORT FOR SEPTEMBER 2, 1972*
Time
1305
1320
1350
1730
1750
1800
0305
0310
0355
2000
2140
2150
°3
(ppm)
0.085
0.086
0.093
0.102
0.107
0.109
0.090
0.080
0.089
0.122
0.106
0.109
R(T + S)
(ppm mln x 10 )
-13.3
+ 3.8
-18.1
+26.4
-36.3
+14.0
-92.5
-100.8
-11.0
-56.8
+19 . 1
\+15.2
X
(Langleys)
Daytime
.50
.53
1.00
.46
.30
.29
Nighttime
Time
1310
1315
1345
1725
1745
1755
0315
0340
0345
2010
2015
2030
°3
(ppm)
0.087
0.095
0.099
0.116
0.104
0.116
0.079
0.087
0.088
0.115
0.123
0.127
R(T + S)
(ppm mln x 10 )
+80.9
+37.3
+42.1
+44.8
+48.8
+63.7
+47.6
+237.5
+29.7
+48.3
+11.0
+26.6
X
(Langleys)
.50
.33
.86
.44
.32
.32
Table 12. CALCULATED OZONE DESTRUCTION RATES AT GARRETT COUNTY
MARYLAND AIRPORT FOR SEPTEMBER 2, 1972a
Time
1303
1320
1350
1730
17SO
1800
0303
0310
0355
2000
2 HO
21SO
°3
(ppm)
0.085
O.OB6
0.093
0.102
0.107
0.109
0.090
0.080
0.089
0.122
0.106
0.109
Negative dOj/dt
d°3/dt M>
(ppm/mln) (ppm/mln)
-43.2
-23.3
-45.1
- 3.B
-39.9
-17.2
-98.3
-120.3
-19.8
-65.7
-15.8-
- 2.1
-30.0
-27.0
-27.0
-30.3
- 3.6
-31.2
- 5.6
-19.5
- 8.7
- B.9
-34.9
-17.3
SD/Oj
-ZKD
-3.5
-3.2
-2.9
-3.0
-0.3
-2.9
-0.6
-2.4
-1.0
-0.7
-3.2
-1.6
X
(Langleys)
Daytime
.50
.53
1.00
.46
..30
.29
Nighttime
Time
1310
1315
1345
1725
1745
1755
0315
0340
0345
2010
2015
2030
°3
(ppm)
0.087
0,095
0.099
0.116
0.104
0.116
0.079
0;087<
0.088
0.115
0.123
0.127
Positive do /dt
dOj/dt BD
(ppm/mln) (ppm/min)
+62.8
+21.5
+30.8'
+19.4
+ 6.2
+48.1
+38.4
+212.7
+11.5
+ 6.1
+ 4.5
+ 4.8
-18.0
-15.7
-11.3
-25.4
-42.6
-15.7
- 9.2
-24.8
-18.3
-42.2
- 6.5
-19.8
BD/Oj
-£KD
-2.1
-1.7
-1.1
-2.3
-4.1
-1.4
-1.2
-2.8.
-2.1
-3.7
-0.5
-1.6
X
(Ungleyg)
.SO
.33
, .86
.44
. .32
.32 '
" Multiply d03/dt, BD, and BD/Oj by 10"
-------
or greater, any mixing with natural air would result In a negative value
of RT. A positive RT value at night Is Interpreted to mean mixing the
previously sampled air with air relatively rich or more concentrated In
0 . Starting with an 0 concentration of 0.08 ppm or greater, a
positive nighttime RT suggests the presence of an O.-rlch (> 0.08 ppm)
reservoir of air.
Although the daylight synthesis term was not determined, the rate
of synthesis at night is obviously zero; thus nighttime transport can be
calculated. The average RT value for the above mentioned study in
piedmont North Carolina was 10.1 x 10~ ppm oln . The maximum RT value
determined in this study was 237.5 x 10 ppm min - more than an order
of magnitude greater than corresponding values for-North Carolina.
It is believed that the results of the CSTR experiment provide an
Independent line of evidence in support of a transport mechanism to
account for the observed high concentrations of ozone in the study area.
50
SECTION 8
SOURCE OF HIGH OZONE CONCENTRATIONS
This section considers the contribution of power transmission lines
to atmospheric ozone levels, and examines the evidence for local photo-
chemical synthesis of ozone and for atmospheric transport of ozone
into the study area.
As a preface to the discussion that follows, it will be helpful
to recall the research hypothesis set forth In Section 1: Ozone
precursors are released into the troposphere at a location remote
from the study area. Given appropriate meteorological conditions,
the precursors are transported to the study area. During transport
and in the presence of sunlight, ozone is synthesized. . The hypothesis
further assumes that by sundown the precursors (which are also destructive
agents) are consumed, leaving high residual ozone concentrations which
are transported to the surface by mechanical turbulence.
8.1 Contribution of Power Transmission Lines
If corona discharge from 500-KV power transmission lines contributes
significantly to ambient ozone levels, the ozone concentrations at locations
downwind of the power transmission lines should be greater than those
upwind or underneath the lines. To evaluate this possibility, a two-way
analysis of variance (ANOVA) procedure (Dixon and Massey, 1957) was used
to.test the null hypothesis; i.e., ozone concentration means for each
of the five locations indicated in Table 8 above, are equal. The analysis
of variance table is shown in Table 14 where it can be seen that both
days and location are significant sources of variation. Accordingly,
the hypothesis of equal means is rejected. The mear ozone concentrations
were 0.064, 0.064, 0.059, 0.062, and 0.064 ppm for locations 140 yards
upwind of, 50 yards upwind of, underneath, 50 yards downwind of, and
140 yards downwind of the power lines, respectively. The fact that the
quadratic effect was highly significant while the linear effect was of
negligible significance suggests that the variation due to location is
primarily attributable to the observations made underneath the power lines.
However, no physical explanation for the lower ozone concentrations
51
-------
Table 14. ANOVA TABLE FOR POWER TRANSMISSION LINE STUDIES
Source of
variation
Days
Location
Linear
Quadratic
Remainder
Residual
Total •
Sum of
squares
0.013592
0.000266
0.000130
0.000136
0.001004
0.014862
df
4
4
1
1
2
61
69
Mean
square
0.003398
0.000067
0.000130
0.000068
0.000016
F ratio
212. 4a
4.19a
8.133
4.25b
.Significant at 0.01 level.
b Significant at 0.05 level.
occurring underneath the power lines Is immediately apparent. In
light of this anomalous finding, no conclusions were reached regarding
the contribution of power transmission lines to the high ozone concen-
trations observed at ground-level in the study area.
8.2 Ozone Synthesis
8.2.1 Local Photochemical Synthesis
The most startling fact that emerges from this study is that
ozone concentrations at the Garrett County Maryland Airport exceed the
0.08 ppm National Air Quality Standard approximately 11 percent of
the time. In an urban area this would not be remarkable and would be
readily attributed to local photochemical synthesis from nitrogen dloxldi
and hydrocarbon precursors. At Garrett County, however, nitrogen dioxidi
and non-methane hydrocarbon concentrations were at or near background
levels throughout the study period. Also, neither natural nor manmade
sources in the study area appear capable of producing the precursor
quantities required fcr synthesis of high ozone concentrations. Ozone
synthesis from naturally occurring precursors has been demonstrated
(Ripperton, et al., 1971); however, it is considered unlikely that it
could account for concentrations exceeding the standard. The persistence
of high ozone concentrations at night reflects the low precursor (and
destructive agent) concentrations and obviously cannot be the result
of ongoing photochemical processes. Thus, an evaluation of air quality
measurements suggests that local photochemical synthesis cannot account
for the observed high concentrations of ozone.
52
8.2.2 Remote Area Synthesis
Another possibility Is that ozone vas generated from
precursors of urban origin and transported into the study area aloft—
aloft, because contact with soil and vegetation probably would destroy
most of the ozone. This would be the case particularly at night when
there is no possibility of replacement through ozone synthesis. That
layers of high ozone concentration can be transported aloft from urban
areas is shown by various ozonograms, particularly those for Bedford,
Massachusetts (near Boston), and at Point Mugu, California (north of
Los Angeles on the coast) (Bering and Borden, 1967). A Ft. Mugu sounding
prepared by Lea (Lea, 1968) and shown in Fig. 19 shows a layer of ozone
about 0.5 kilometer aloft with a partial pressure of 300 umb (-0.30 ppm).
It is known from chamber studies that once ozone precursors have
been fully utilized, i.e. are no longer detectable, measurable ozone
often remains in the chamber air. Since no nitric oxide, nitrogen
dioxide, or olefins remain to react with ozone,- it tends to persist,
reacting slowly with chamber walls and other reaction products. This
situation is approximated In the atmosphere when air leaves an urban
area; ozone-generating reactions tend to go to completion and there are
no additional large Injections of nitric oxide or olefins which would
destroy the ozone.
0 00 200
Ozone partial pressure (umb)
300 » X 3D
Temperature ( C)
Figure 19. Vertical ozone and temperature profile at Point
Mugu, California. (Source: D.A. Lea, 1968)
53
-------
If the concept of a "high" concentration of ozone and low nitrogen
oxide and hydrocarbon concentrations in a "spent" air pollution system
drifting to another area is correct, the phenomenon should be observable
in a number of areas. This is the case, although the'characteristics
of such systems need not be, and indeed are not, identical.
Two such cases are Indio and the Mineral King Valley, both in
California. Ozone concentrations at Indio, a city of about 15,000
residents located east of the Los Angeles Basin and the intermediate
valleys of southern California, have exceeded the California Air Quality
Standard (0.10 ppm for 1 hour) more frequently than any other sampling. .
station in California (Pitts, 1972). Carbon monoxide values at Indio
were described by Pitta as "low; 1 or 2 ppm."
A situation in the Mineral King Valley of California (Miller, 1972)
is analogous to, but not identical with the situation in the present
study area. Figure 20 shows a slight diurnal trend with an oxidant peak
approaching 0.10 ppm between 1700 and 1800 FST and nighttime concentrations
of approximately 0.06 ppm which are unusually high. Oxides of nitrogen
were reported to be "very low". Wind and other meteorological data
12
10
8
00 02 04 06 08 10 12 14 16 18 20 22 24
Hours(PST) '
Figure 20. Mean diurnal oxidant concentration at Mineral King
Valley, California. (Source: P.R. Miller, et al.,
1972)
were interpreted as showing that some of the air sampled at the Mineral
King Valley had been advected from Fresno.
From these considerations a picture emerges of a "spent" photochemical
system with depleted ozone precursors, low carbon monoxide and sulfur
dioxide concentrations, and ozone concentrations lower than those found
In urban areas, but higher than could be accounted for by synthesis
from naturally occurring precursors or by transport from the stratosphere.
In order to maintain high ozone levels, especially at night, the
pollution system would have to travel aloft out of contact with the
ground and large pollution sources. Alternatively, if the pollution
systems travels in contact with the ground—where ozone destruction
occurs—the system must have sufficient vertical extent to provide a
reservoir of ozone-rich air to be mixed with the depleted air near the
ground. When local turbulence occurs, bringing ozone to the ground and
mixing additional precursors in the air, destruction can begin anew (day
and night), as can synthesis (day). In.a truly rural area, however,
these processes would be limited by the quantities of precursors being
emitted to the air.
To reiterate, the picture presented by the measurements is one of
ozone values In what was thought to be an unpolluted area which exceed
the National Air Quality Standard. The rates of change of ozone concen-
tration due to synthesis and destruction during the study period are,
consistent with a "spent" photochemical system model. Calculated
destruction.rates approximate those of naturally occurring precursors
(destructive agents). Transport (here meaning all air movement)was
the dominant process producing change in ozone concentration. The
picture is consistent with transport of ozone-rich air from a source
of precursors remote from the study area. A question naturally arises
as to where the pollution system originates and how long it has been
aloft. As to the latter, the air quality measurements, the lack of
real diurnal differences, and the comparison with the California experience
suggest that the system has been traveling a minimum of two days.
Transport mechanisms are examined in the following paragraphs.
55
-------
8.3 Ozone Transport
The preceding discussion indicated that atmospheric transport
would account for the high ozone concentrations. A consideration of
atmospheric diffusion processes (Paetzold, 1961) and perusal of a
thousand or more ozonograms (Hering and Borden, 1967), furnish strong
evidence that 0 concentrations of 0.08 ppm and greater have not been
advected from the stratosphere. Accordingly, the discussion that
follows concerns itself with tropospheric transport.
8.3.1 Ozone and Wind Direction
If the ozone source is confined to a particular area, or
areas, within 100 mi. of the Garrett County Maryland Airport, some
dependency of ozone concentration upon the wind direction should be
evident, especially high ozone concentrations. To examine this possi-
bility, the frequencies of occurrence of the average hourly wind
direction associated with each of six categories of average hourly
ozone concentration were computed. The limits selected for the ozone
concentration categories were the upper 98, 85, 50, 13, and 2 percentile
of observed concentrations. The 13 percentile corresponds to the
0.08 ppm standard. These limits were used to construct the ozone
wlndrose in Fig. 21. Preferred directions for high ozone concentra-
tions appear to be the west-northwest and northwest; however, the
latter is also a preferred direction for low ozone concentrations.
Ozone concentrations above 0.08 ppm occur most frequently with west
and west-southwest winds which also are frequently observed. Other
directions, such as north through east are Infrequently observed and
thus a preference for either high or low ozone concentrations is
indistinguishable In the figure.
The data of Fig. 21 were normalized by the frequency of occurrence
of each wind direction to eliminate the bias of prevailing or infrequent
wind direction. The ozone wlndrose with bias removed Is shown in Fig. 22.
The normalized wlndrose Indicates that the categories of high ozone
concentration (>0.08 ppm) are reasonably evenly distributed in all
directions, whereas the two low concentration categories occur most
frequently when winds are from the northeastern semicircle. A comparison
56
N
i ,
03 > 0.096 ppm
0.096 >0s > 0.080 ppm
0.080 >0a 2 0.055 ppm
0.055 > 03 > 0.032 ppm
0 5 10%
I i i i i I i i l i I
SCALE
0.032>0s > O.O2I ppm
0.021 ppm >'0s
.Figure 21. Frequency of occurrence of wind direction with
indicated ozone (0,) concentrations.
57
-------
03 >0.096 ppm
0.096>03 > 0.080 ppm
0.080 > 0j > 0.055 ppm
O.055»03 > 0.032 ppm
0 50 100%
1 i i i i I i i i i I
SCALE
0.032>03> 0.021 ppm
0.021 ppm > 03
Figure 22. Frequency of occurrence of wind direction with
indicated ozone (0,) concentration, normalized by
the frequency of occurrence of a wind direction.
of the occurrence of greater than mean concentration to the lower than
mean concentration is shown in Fig. 23. This figure shows a preference
of ozone concentrations above the mean to occur with winds from south-
southeast through southwest to west-northwest and for concentrations
below the mean with wind from the remaining directions, although those
directions occur less frequently.
The fact that this analysis did not show a definite source-receptor
relationship combined with the results of the mobile sampling program
suggests non-local sources and a larger time and space scale for ozone
transport.
o
a
a
o
z
4,00
3.00
2.00
1.00
0.50
0.40
0.30
0.25
16
Figure 23. Ratio of occurrences of ozone concentration greater
than mean (0.055 ppm) to occurrences of ozone concen-
tration less than mean by wind direction. August 22-
September 25, 1972.(Total number of occurrences given
at extremity of bar.)
58
59
-------
8.J.2 Ozone Concentration Changes in Air Masses
The next larger time and space scale meteorological data
that are readily available are the twelve-hourly synoptic upper air
observations. To correspond with this data, average concentrations
of ozone over the 12-hour period 0700 to 1855 EDT (daytime) and over
the 12-hour period 1900 to 0655 EDT (nighttime) were computed. These
are shown in Fig. 24 for the entire period of observations.
A striking feature of the data, presented in Fig. 24 is the
repeated occurrence of a gradual increase of ozone concentration
to a peak value, followed by a sharp decrease in concentration to
approximately half the peak value. Particularly apparent is the
sequence of ozone concentration increase over the period from the1
daytime of 10 August through the night of 14-15 August and the
sharp decrease for the daytime period of 15 August. Similarly, the
increase of ozone concentration from the daytime of 29 August through
the night of 2-3 September and the decrease which continued through
the daytime, period of 3 September and the night of 3-4 September
is quite narked. .From the minimum reached on the night of 3-4 September
ozone concentration.again Increased to a peak value during the daytime
period of 8 September. This peak, which was only 0.067 ppo, is followed
by a sharp decrease in concentration for the next 12-hour period and
a continued decrease until a minimum is reached during the night of
10-11 September. A rapid increase in ozone concentration followed
this iHiHimmi value and a maximum value was reached during the 12-13
September nighttime period. A gradual decrease in concentration
occurred over the next 24 hours followed by a sharp decrease to a
new minimum concentration value during the night of 14-15 September.
The rates and magnitudes of increase and decrease of ozone
concentration during the periods discussed vary considerably. However,
there is a remarkable similarity in the patterns Indicated by the
concentration changes. More significant though, is the fact that
the passage of an air mass front occurred either Immediately preceding,
or during, the 12-hour period characterized by the low ozone concentra-
tion following the sudden decrease. The approximate times of passage
60
T9
12-Br Avttrag* Oson* Concentration
* M. M. r
3 B •
oo 8
O C
O
C
(T
CD cr*o n
*1 0» (D
H CL t-i (n
Q (D H- <
3 0. O Q
D- *
« (U CD (D
P 3 00
H O. O* (D
P» (D
(0 H OQ O
a. p H- N
3 P 0
EH o
§°
P- O M
-------
of these fronts have been Indicated on Fig. 24. Since the air masses
on eitner side of a front have different origins and histories, trajectory
analysis using the 12-hourly synoptic meteorological data was used to
Investigate the marked changes In ozone concentrations accompanying
the frontal passage.
8.3.3 Air Trajectory Analysis
Air trajectories at the 850-mb pressure level (approximately
5000 ft MSL) were prepared for periods preceding and following the frontal
passages Indicated In Fig. 24. The arrival time at the Garrett County Maryland
Airport for air following a particular trajectory is 0800 EDT (approximately
the beginning of the averaging period for the daytime ozone concentration) and
2000 EDT (approximately the beginning of the nighttime averaging period). These
times coincide to the times at which the meteorological measurements are made.
The air trajectories prepared are only approximations. Even the
care taken in their preparation cannot eliminate a decrease in accuracy
as time and distance from the terminus increase. Further, it must be
recognized that air parcels normally follow isentroplc rather than Iso-
baric surfaces. Thus these trajectories represent estimates of the flow
along the 850 mb constant pressure surface and inferences concerning sources
should not be attempted because a particular trajectory passes over a specific
geographic point.
Selected trajectories associated with period of increase followed
by rapid decreases In ozone concentrations are presented below.
Case 1 - Frontal Passage of 0500 EDT 15 August
Trajectories for four days preceding this frontal passage
and two and a half days following are shown in Pig. 25. It should be
noted that trajectories A through F, prior to the frontal passage,
and thus during the period of increasing ozone concentrations, tend to
show decreasing speed for air parcel movement, especially during the
24-hour period immediately prior to arrival at Garrett County.
Trajectories G through K for air parcels arriving after the frontal
passage Indicate quite rapid air movement.
oG ? &* S
§A- S H3. £.
8i gg S
S" &Sl S
s 9 a ? a
SB • 8S
* These frontal passages were determined independently of the ozone data
as part of the documentation of the meteorological conditions occurring
during the period of observation. A discussion of the meteorological
conditions associated with ozone concentrations greater than 0.08 ppm is
given in Appendix E.
62
(D rt rt rt
H- O ro O
rt M o. n
H-
O -O
M» O D-
3 C CD
rt n P-
tr o H- n
n H. a
•-! oo CD
CD n n
o"
B.8S1
SIS
Ul Ul
S P
00 h*
OO 00 O 00
bo > 8?
2§ " °c
O O OB O O>
b o
ss
O O
O O
O O h- P H>
utui Ci S Ci
-------
a o
n n
a o
a, s
2 "
*
ss s
&n s
S? ^2
§•«»
n r-
5"1
o1?
S3
'S s
§S
is
u> u>
O O OB
O O t"
mo* o*
e"
Is
Case 2 - Frontal Passage of 0400 EDT 3 September 1972
The period of buildup of ozone concentration from 29 August
through 2 September Is characterized by air arriving at Garrett County
after passing over eastern coastal areas as shown by trajectories
A through E in Fig. 26 . A definite change in the path of air and the
speed of movement is Indicated by trajectories F through H, which are
post frontal and Include the nighttime period of minimum ozone concen-
tration on 3-4 September.
Case 3 - Frontal Passage of 0300 EDT 9 September 1972
In spite of a rather uniform rate of increase in ozone concen-
trations during the period from 5 September through 8 September 1900 EDT
there are significant differences in the trajectories for the same period
(Fig. 27). It should be noted, however, that the movement for all of
these Is relatively uniform and rapid. The ozone concentration during this
period increased to a maximum of only 0.067 ppm (12-hour average), which
is less by about 0.03 ppm than the maximum reached in cases 1 -and
2.
The marked change in trajectories occurs after the frontal passage
as shown by trajectories H and 1. -
8.4 Interpre tat ion ".
The discussion of a possible chemical model (Section 8.2) and the
transport of air parcels to a receptor area (Section 8.3) suggests that
the following conditions may exist over a large (multi-state) urban-Industrial
area:
1) Ozone precursors and destructive agents, i.e. nitrogen oxides
. and reactive hydrocarbons, are emitted by the numerous recognized
stationary and mobile sources.
2) In the presence of solar radiation,reaction of the precursors
produce,ozone throughout the surface-based mixing layer.
3) Concurrently, destruction of ozone takes place, but over a
, sufficiently long period (perhaps 48 hours) the destructive
agents are depleted as a result of their entering Into the
ozone production reaction.
65
-------
. 9/9 - 0300 EDT
Trajectory
Arrival Date
Arrival Tina
Data
12-hour Mean
Concentratli
(EDT)
Ozone
m
0700-1855
1900-0655
9/5
2000
9/5
0.048
0.053
9/6
0800
9/6
0.058
0.059
9/6
2000
9/7
0800
9/7
0.061
0.063
9/7
2000
9/8
0800
9/8
0.067.
0.0*7
G
9/8 '
2000
H
9/9
0800
9/9
0.045
0.037
I
9/9
2000
Figure 27. Trajectories of air arriving at Garrett County Maryland Airport at the indicated
time during Case 3. Dashed portions indicate an estimated trajectory. Open
circles (O) and crosses (X) show the 0800 and 2000 EDT positions of the air.
FROPA
9/3 - 0400 EDT
Trajectory
Arrival
Arrival
Date
Date
TiM
(KDT)
12-hour Mean Oione
Concentration
0700-1855
1900-0655
A
8/31
2000
8/31
0.079
0.068
B
9/1
0800
9/1
0.077
0.075
C
9/1
2000
D
9/2
0800
9/2
0.094
0.095
8 .
9/2
2000
F
9/3
0800
9/3
0.069
0.035
G
9/3
2000.
H
9/4
0800
9/4
0.045
Figure 26. Trajectories of air arriving at Garrett County Maryland Airport at the indicated
time during Case 2. Dashed portions Indicate an estimated trajectory. Open
circles (O) and crosses (X) show, the 0800 and 2000 EDT positions of the air.
-------
4) A receptor, appropriately located at some presently unspecified
distance and travel time from sources of ozone destructive
agents, will experience an ozone-rich atmosphere containing
Insignificant concentrations of ozone precursors or destructive
agents.
5) Ozone destruction will take place at the air-ground Interface,
but turbulent mixing will maintain high ozone concentration
throughout the mixing layer.
6) High ozone concentrations will be the "normal" condition in
air masses originating or modified over extensive urban-industrial
areas.
7) Low ozone concentrations will occur over extensive urban-industrial
areas only during the period immediately following the arrival
of a relatively rapidly moving air mass having its origin in
an extensive non-urban-Industrial area.
68
SECTION 9
CONCLUSIONS
The principal conclusions arising from this research are enumerated
below:
1) The research reported herein verifies the high concentrations
of ozone previously found in the Mt. Storm, West Virginia area.
The maximum hourly concentration for the study period was
0.119 ppm, while, the average hourly concentration for the
study period was 0.057 ppm. Approximately 11 percent of
the one-hour samples exceeded the 0.08 ppm National Air
Quality Standard.
2) The occurrence of high ozone concentrations Is not restricted
to a single location. Rather it appears to be the case for
most of Garrett County, Maryland and that portion of Preston
County, West Virginia in which ozone measurements were made.
3) The lack of well-defined nocturnal surface-based temperature
inversions, the small diurnal temperature range, and the
relatively constant wind speed from day to night suggests
that the air masses over the base station were relatively
well mixed in the lower levels. As a consequence, the
assumption that the high concentrations of ozone are not
restricted to a shallow surface layer appears warranted.
In fact, higher concentrations of ozone above the surface
seem probable since vegetation, soil, and all structures
and obstacles on the surface form an ozone sink.
4) High ozone concentrations at the Garrett County Maryland
Airport result from ozone transport and not from in situ
photochemical synthesis.
5) The observed high ozone concentrations cannot be explained
In terms of air flow from a specific point source or a
single urban-industrial area source. However, episodes of
69
-------
high concentration do appear to be associated with air
masses arriving in the study area after passing over urban-
industrial regions. Ho clear pattern of air mass trajectory,
speed, or time of day of precursor Injection is apparent
from the available data. This meteorological interpretation
is compatible with current concepts of atmospheric chemistry.
6) The results of this research provide substantive support for
a hypothesis which Is as follows: Ozone precursors are
released Into the troposphere at a location remote from
the study area. Given appropriate meteorological conditions,
the air mass containing the precursors is transported to
the study area. During transport and In the presence of
sunlight, ozone is synthesized. The hypothesis further
assumes that the precursors (which are also destructive
agents) are consumed, leaving high residual ozone concen-
trations which are transported to the surface by mechanical
turbulence.
70
SECTION 10
RECOMMENDATIONS FOR FURTHER RESEARCH
During the course of this study a number of research needs were
Identified. The most pressing among them are enumerated below:
1) The frequency of occurrence of high ozone concentrations
(>0.08 ppm) in the study area should be determined over an
extended period of time, Ideally one year.
2) A study to further delineate the horizontal extent of the
region of high ozone concentration is indicated. Such a
study should involve continuous ozone monitoring at.the
Garrett County Maryland Airport as well as at several sites
approximately 100 miles distant.
3) As part of 1) and 2), in-depth synoptic-scale meteoro-
logical analysis should be performed. The analysis would
identify an air mass according to source region, characterize
its thermal structure, and plot its, trajectory as a basis
for Interpreting variations in ozone concentration at
monitoring sites.
A) The vertical distribution of ozone, ozone precursors,
ambient temperature, and dewpolnt temperature in the
study area should be determined. An instrumented aircraft
would be required for these determinations.
5) A reaction chamber simulation to model the chemical
processes suggested In this report should be undertaken.
. 6) A long-term study of the contribution of corona discharge
from high voltage power transmission lines to atmospheric
ozone levels is needed. Such a study should examine power
lines of various voltages In a number of locations with
different terrain features and under a variety of meteoro-
logical conditions.
71
-------
APPENDIX A
CALIBRATION PROCEDURES
APPENDIX A
CALIBRATION PROCEDURES
Precediflg page blank
73
Dynamic calibration procedures were used to. calibrate all instrumen-
tation during the field measurement period. The procedures used and the
frequency of use by pollutants were specified In Table 1, Section 4. The
validity of the procedures was established in earlier studies (Decker,
et al., 1972). Those procedures are described In the following paragraphs.
A-l Ozone
A dynamic calibration system using an ultraviolet ozone generator
(Hodgeson, et al. 1970) was used to calibrate the solid-phase chemilumlnescent
ozone meter. Briefly, the generator consists of an 8-in ultraviolet mercury
vapor lamp which irradiates a 5/8-in quartz tube through which clean,
compressed air flows at 5 Ifm. Ozone concentrations from 0 to approxi-
mately 1.0 ppm can be generated using this technique. Although the
ultraviolet ozone generator has been shown to be quite stable and repro-
ducible, the neutral-buffered potassium iodide technique was used as a
reference method (Public Health Service, 1965). A permanent calibration
assembly consisting of a zero air source, calibrated rotameter, ozone
generator, and glass manifold with sampling ports was set up in the RTI
Environmental Monitoring Laboratory for calibration of all ozone meters.
A schematic diagram of the ozone calibration systems Is shown in Fig. A-l.
A-2 Nitric Oxide/Nitrogen Dioxide
Calibration of the Bendix Model 8101-B Chemilumlnescent NO-NOx~N02
Analyzer was accomplished using a nitric oxide-ozone conversion unit
developed by EPA. Nitric oxide In nitrogen (-50 ppm) was purchased
from Scott Research Laboratories with a certificate of analysis. The
desired concentration range for calibration (0-1.0 ppm) was achieved
by dilution with compressed air. Nitrogen dioxide concentrations
were produced by mixing ozone with nitric oxide In a conversion chamber.
An ozone to nitric oxide ratio of 1.2 was required for 100 percent
conversion of nitric oxide to nitrogen dioxide. The nitrogen dioxide
produced is diluted to the desired concentration with diluent air.
A drawing of this calibration system Is shown in Fig. A-2.
75
-------
Compressed
Air
Figure A-l. Ozone calibration system.
To Calibration
Manifold
To
Regulated
AC Voltage
76
Ozone
Generator
Capillary
Restrlctor
Rotameter
r-l-1 Rotameter
Figure A-2. Nitric oxide and nitrogen dioxide
calibration system.
77
-------
A-3 Sulfur Dioxide
A dynamic calibration system using a gravimetrically calibrated
sulfur dioxide permeation tube (O'Keeffe and Ortman, 1966; Scaringelli,
1966) as a primary standard and zero air as a diluent was used to provide
known concentrations of sulfur dioxide to calibrate the Bendix flame
photometric analyzer. A schematic diagram of the system is shown in
Fig. A-3.
A-4 Total Hydrocarbon, Methane, Carbon Monoxide
Calibration of the Beckman Model 6800 GC-FID was accomplished
using standard calibration gases certified as to component concentrations.
These gases were purchased from'Scott Research Laboratories. Zero air
certified to contain less than 0.1 ppm total hydrocarbon and carbon
monoxide was used to zero the instrument. Span concentrations of methane
and carbon monoxide in zero air up to 5 ppm were used for calibration
purposes.
PERMEATION TUBE
•NT-TEMPERATURE
IATER BATH
THERMISTOR TEMPERATURE MONITOR
Figure A-3. Permeation tube calibration system.
78
79
-------
APPENDIX B
PERFORMANCE CHARACTERISTICS AND OPERATIONAL
SUMMARIES FOR AIR QUALITY MONITORING INSTRUMENTS
Preceding page blank
APPENDIX B
PERFORMANCE CHARACTERISTICS AND OPERATIONAL
SUMMARIES FOR AIR QUALITY MONITORING INSTRUMENTS
B-l Instrument Performance Characteristics
Minimum detectable concentration, range and precision of the air
quality monitoring instruments used in the Mt. Storm study are
summarized in Table B-l.
B-2 Operational Summaries
A summary for each air quality monitoring Instrument used In
the Mt. Storm study is presented below.
Table B-l. INSTRUMENT PERFORMANCE CHARACTERISTICS
81
Instrument Parameter
Solid-phase 0,
chemiluminescent
ozone meter
Bendlz model NO
8101-B chemilum-
inescent NO-NO - x
NOj analyzer * NO,
Bendlx Model 8300 SO.
flame photometric
analyzer
Beckman model THC
6800 gas
chromatographic 4
flame lonlza- „-
tlon detector
NMHC
Preceding page blank
M-in-iimim detectable
concentration (ppm) Range (ppm) Precision (Z)
0.002 0.2 + 1.0
0.005 0.5 +0.5
0.005 0.5 + 0.5
0.010 0.5 + 1.0
0.005 0.2 + 1.0
0.100 10.0, 5.0 + 1.0
0.100 10.0, 5.0 + 1.0
0.100 10.0, 5.0 + 1.0
0.150 + 2.0
83
-------
B-2.1 Chemiluminescent Ozone Meter
During Che period August 4 to September 25, .1972, two solid-
phase Chemiluminescent ozone meters vere used for ambient monitoring at the
Garrett County Maryland Airport. Intermittent problems (I.e. shutter
failure, timer failure, electronic problems) necessitated replacement
of a meter borrowed from the University of North Carolina which had been
online from August 4 to August 22, 1972 with a solid-phase Chemiluminescent
*
ozone meter obtained from the Division of Atmospheric Surveillance, EPA.
The meter obtained from EPA operated from August 22 to September 5, 1972
virtually, trouble free. Less than 18 hours of data were invalidated due
to failures which were mechanical rather than electronic. Minimal cali-
bration drift (both zero and span) were encountered during this period
of time.
B-2.2 Bend IK Model 8101-B Chemiluminescent NO-NO^-l^ Analyzer
During the period August 4 to September 24, 1972, the instrument
operated with no failure periods. The analyzer exhibited good zero and
span stability throughout the measurement program. Negative concentra-
tion were frequently encountered for NO and NO values. It Is believed
that zeroing the instrument with dry compressed air and then monitoring
ambient air with humidities greater than 60-70 percent depressed instrument
zero to some point below that previously defined as zero. Since the
instrument output Is linear and both NO and NO were similarly affected,
it is reasonable to assume that the relative difference between NO and
NO, i.e. N02 is valid.
B-2.3 Bendix Model 8300 Flame Photometric Analyzer
During the period August 4 to September 24, 1972, this
analyzer experienced three failures. Two of the three failures were
electronic and required readjustment of the gain on the photomultiplier
tube. The other failure was a flow regulation problem that resulted in
the invalidation of approximately 4 days of data. Excellent zero and
span stability was exhibited by the unit while it was operating.
B-2.4 Beckman Model 6800 Gas Chromatographlc Flame lonlzation
Detector
Several failures and malfunctions occurred which severely
affected the performance of this -Instrument. Some problems resulted
from electronic failure of integrated circuits. Diffusion of moisture
through the hydrogen regulator diaphragm contaminated the analytical
column which In turn caused a shift in the elution time of methane and '
carbon monoxide from the column. Intermittent switching on and off
of the calibrate solenoid valve resulted in a loss of sample and thus
invalidated a considerable amount of data. Negative non-methane
hydrocarbon values were obtained when methane concentrations were
greater than indicated total hydrocarbon concentrations.
During the period August 4 to August 11, 1972 the Instrument
performed satisfactorily. During the period August 11 to September 24,
1972 the problems cited above were encountered. The carbon monoxide
data are considered to be the most valid of three measurements.,
Methane and total hydrocarbon data are less reliable.
* Obtained from EPA warehouse on August 18, 1972.
84
85
-------
APPENDIX C
THE CONTINUOUS STIRRED TANK REACTOR
APPENDIX C
THE CONTINUOUS STIRRED TANK REACTOR
In order to determine the synthesis, destruction and.transport
of.ozone near the surface of the earth (Jeffries, 1971) used an'
experimental technique consisting of two non-steady-state continuous
stirred tank reactors (CSTR) which use ambient air as the reacting
medium. Figure C-l shows this experimental set-up.
Behavior of a reactive substance in a CSTR is represented by
dO
-T-^ (outlet) - f • 0, (inlet) - f • 0, (outlet) - vr (1)
at • . J J
where
f - flow rate through the reactor,
0, - appropriate 0, concentration,
v - volume of reactor, and
r - rate of reaction affecting 0^ concentration per unit
volume and time.
Equation (1) Is a mass balance equation for the reactor. Letting
Q = f/v and rearranging (1) gives
dO
r = Q 03 (outlet) - Q 03 (inlet) + -^ (outlet) . (2)
Only 0, concentrations as a function of time at the inlet and
outlet to the CSTR are needed. dO3 (outlet)/dt can then be computed
numerically. The rate, r, depends on how the reactor is operated.
If light is excluded from the reactor, only 03 destruction can occur
and r = r,. If light is allowed in the reactor, both synthesis and
destruction of 0, may occur and r - rd + rg. Knowing r^, ra and
87
Preeeuteg pge Hank
89
-------
AMB1!
AIR
3001pm
DARK FLASK
STIRRER
MOTOR
•ALUMINUM FOIL
72-LITER PYREX
GLASS FLASK
72-LITER
PYREX
GLASS
FLASK
the concentration of Oj in the ambient air, the rate of transport
of 03> r£, necessary to maintain the ambient 0, concentration can
be computed.
CLEAR FLASK
Figure C-l. CSTR system for rate measurements.
90
91
-------
APPENDIX D
TETHERED BALLOON PROCEDURES
APPENDIX D
TETHERED BALLOON PROCEDURES
page blank
Preflight checkout of Che system consisted of the standard radiosonde
baseline measurements. The ozone sensor was calibrated with zero air
and air containing 0.05, 0.10, and 0.15 ppm of ozone before each flight.
The strip chart ordlnate corresponding to each ozone concentration was
recorded and subsequently a calibration curve was prepared. A final
check of reference signals, temperature, and ozone concentration was
made immediately prior to launching. All launchlngs were made with
prior approval of the airport manager.
Initially, the balloon was allowed to rise until 200 ft of tether
line was out. A five- to eight-minute sample of ozone and* temperature
or humidity was taken at this height. This procedure was carried out
at Increments of 200 ft of tether line, until 1000 ft of line were out.
Two-hundred ft Increments were used during the balloon descent as well.
Measurements of the elevation angle between the tether line and the
horizontal at the tether point were made and recorded for each significant
change of angle during a flight. Comments on the flight characteristics
of the balloon and/or meteorological conditions were recorded on the
flight log as necessary. The duration of a flight was approximately
90 minutes.
A post-flight calibration of the instrument package was performed
for each flight in which the radiosonde transmitter battery was still
functional after approximately two-and-one-half hours of operation.
The procedure was identical to the preflight checkout except that zero
air and 0.10 ppm ozone were used.
The ozonesonde transmissions were received, demodulated and recorded
on a strip chart. Figure D-l Is a segment of strip chart record. For
each sample, the ozone signal was averaged over two-minute intervals,
and the drift of the 1 and I signals was carefully noted. The ozone
partial pressure was determined from the preflight calibration of the
Precetifif pige Wank
95
93
-------
Figure D-l. Segment of ozonesonde strip chart record showing
ozone signal, ozone calibration signals (I , I ),
temperature, relative humidity, and refereSce .
signals.
96
sensor, adjusted as necessary for any Ic and IQ drift, and converted to
concentration units using the estimated pressure at altitude. An average
value of temperature or relative humidity was obtained by applying standard
radiosonde procedures to each strip chart sample period.
The approximate altitude of the Instrument package was estimated
using two independent techniques. The first of these used the measured
station pressure, the baroswitch calibration chart and tne switching
sequence of the baroswitch, to estimate the pressure at the sampling
altitude. • The hypsometric equation, with measured temperature and
humidity, gave the altitude. At temperatures and pressures encountered
at the study site, a one millibar change of pressure corresponded to
an approximate 30 ft altitude change.
The second technique assumed that the tether line formed a catenary.
This being the case,the altitude of the balloon could be determined when
1) the free plus aerodynamic lift of the balloon, 2) the weight per unit
length of the tether line, and 3) the angle that l£ne makes with the
horizontal at the tether point are known. The latter two requirements
were measured. The free lift of the balloon is equal to the bouyancy
of the helium within the balloon less the weight of the balloon and
the Instrument package. It varies from flight to flight due to leakage In
flight and during storage of the balloon. The roof of the hangar used for
balloon storage was not high enough to permit measurements of the actual free
lift of the balloon In static air; however, outside the hangar In nearly
calm air, a free lift of 1.5 Ib was measured. The aerodynamic lift, a
function of the wind speed and angle of attack, was not known. Observation
of the balloon flight characteristics Indicate that this additional lift
significantly affects the flight altitude.
Sets of curves of altitude versus elevation angle for different arc
lengths were prepared for several values of lift. These curves were used
to estimate the altitude of the balloon from the measurements taken.
Subtracting 25 ft from that altitude gives the altitude of the instrument
package. The estimated error In altitude estimates la + 40 ft. Final
altitude estimates were made by Insuring that the two methods gave comparable
. results.
97
-------
APPENDIX E
DISCUSSION OF SYNOPTIC WEATHER FEATURES ACCOMPANYING
EPISODES OF HIGH OZONE CONCENTRATION
Preceding page blank
99
APPENDIX E
DISCUSSION OF SYNOPTIC WEATHER FEATURES ACCOMPANYING
EPISODES OF HIGH OZONE CONCENTRATION
The following discussion is based on Che 0800 EDT weather maps as
they appeared In the Dally Weather Map Series produced by the National
Weather Service.
August 10 - 1800 EDT to 2200 EDT
An occluded front passed through the central and northeastern states
on August 9. A large high pressure system behind the front moved out of
central Canada over the Great Plains, then moved eastward. On the morning
of August 10, the center of the system was located over the Indiana-Ohlo-
Mlchlgan border. The system had moved to the New Jersey coast by the next
morning, with the center moving over the Garrett County Maryland Airport
during the episode.
August 13 - 1300 EDI to August 15 - 0700 EDT
The surface high which moved across the northeastern united States
In the previous episode had migrated slowly eastward. At 0800 12 August,
the.high (1027 mb) was centered over the New Jersey coast and continued to
move eastward at about 20 kt while maintaining a ridge extending over the
southeast United States. As the high receded a weak cold front moved
southeastward, but little contrast of temperature or dewpoint temperature.
The front passed the Garrett County Maryland Airport at approximately
0400 EDT August 13. No substantial change of air mass seems to have
occurred. A trough through the upper Graet Lakes developed pinching the
ridge between Canadian air and stagnant maritime air. By the 14th at 0800
a weak high pressure center (- 1022 mb) established Itself over West
Virginia, and a cooler, strongei high (- 1025 mb) was developing west of
James Bay, with a stationary front (the "pinching" trough) separating them.
The West Virginia high was the surface evidence of a strong ridge aloft
over the central USA. The Canadian high Intensified and began to push
southward, transforming the stationary front Into a cold front, passing
the field site near 0500 EDT, August 15. From the 15th, 0800 EDT map,
temperatures and moisture ahead of the front were similar to those about
101
-------
100 miles behind, Indicating mixing across the frontal zone. By the 16th,
the cold front had pushed southward and westward Charleston, S. C. and the
ridge line lay about 120 miles inland of the Atlantic seaboard with the
high center (1028 mb) near Albany, New York.
August 18 - 1200 EDT to 1800 EDT
By August 17, 0800 EDT the Canadian high of the previous series had
moved offshore, .with an extension over.,the southeast United State's and with
ridging into the Great Plains. Maritime air, originating in the Gulf of
Mexico pushed across as far eastward as Ohio and Michigan behind a warm
front across Lake Huron to very near the Garrett County Airport. By the
next morning the airport remained in the maritime air, which showed little
movement. A cool front moved across New England, leaving a stationary
front along an arc from Flint, Michigan to Buffalo to Washington, D. C. by
the next morning. The airport remained in the original maritime air. Aloft
an east-west ridge over the southeastern United States persisted throughout
the period.
29 August - 0200 to 3 September - 1200 EDT
In the few days previous to this episode, a maritime polar air mass
moving across the Rocky Mountains and central plains gradually was trans-
formed into a maritime tropical air mass from the Rocky Mountains to the
east coast. At 0800 on the 28th a cool front lay along a line from Boston
through Richmond to Tallahassee and was moving out to sea. A second cool
front moving out of Canada lay along the Minnesota-Canada border. On the
29th the situation waa changed little except that the Canadian front had
penetrated eastward to a central New York-Michigan line and showed some upper
air support. By the morning of the 30th, Canadian air pushed south-eastward
behind a front on an arc from Chicago to Cincinnati and through south central
Virginia. A cool high pressure center was located over New York-Pennsylvania
border. The surface ridge line extended from San Antonio to Evansvllle,
Pittsburgh and north northeastward. Aloft, ridging over Illinois and
Indiana was increasing.
Over the next 24 hours, the front dissipated but the ridge line
remained on the southwest-northeast axis to a high over southwest New England.
Morning temperatures were cooler over mountain areas. The ridge aloft
formed into a high over the Philadelphia-Washington, D. C. area. The coolest
and driest air on the morning of- the 31st was found through New England,
102
central Maryland, and eest of the mountains to Indiana but not Illinois.
Flow from the ocean modified the coastal air. For the next 24 hours the
ridge line followed the Appalachians, keeping them relatively dry.
Tropical storm Carrie began approaching the mid-Atlantic coast, from the
southeast and cold front was crossing the mldwestern states. By
September 2 at 0800, Carrie was advancing northward, 200 miles east of
Wallops Island, Virginia. Moisture remained east of the mountains. The
progress of the frontal system had slowly moved to Buffalo to Evansvllle line.
Post-frontal precipitation covered an extensive area. The ridge line,
squeezed between Carrie and the front along the Appalachians, although
weakened, persisted and remained dry.
By September 3, Carrie,approaching Cape Cod from the south, lay
east of New York City. The front progressed slowly eastward with its
rain as the ridge line dissipated. The front passed the airport 0400 EDT.
11 September - 2200 EOT to 13 September - 0800 EDI
On September 9, a continental polar air mass pushed south southeast
across the united States from central Canada. On the 10th, the high
pressure center (- 1030 mb) was located over southwest Ontario bringing
cool air and clear skies to the eastern united States. Aloft, high
pressure centers were scattered across the southeastern states. Tropical
Storm Dawn was moving east and southeast away from the North Carolina
coast. By the morning of the llth, the surface high drifted to the North
Carolina-Virginia border, with the continental polar air from New York City
to Atlanta. West of the Appalachians, air mass modification to maritime
tropical air was underway. A system of Pacific waa moving eastward across
Northern Plains. Within 24 .hours the air over the eastern United States
was fully modified to maritime tropical air, and in a southwest flow. The
Pacific front was stationary from Kansas to the St. Lawrence River valley.
Another Canadian high pressure center (" 1022 mb) north of stationary front
was moving east northeast, but not pushing southeastward. By the morning
of the 13th, warm, moist air extended up to the Great Lakes, with rain and
showers for ISO miles south of a warm front through Michigan to Buffalo
to New York City.
103
-------
REFERENCES
Decker, C. E., T. M. Royal, and J. B. Tommerdahl, Field Evaluation of
New Air Pollution Monitoring System, Research Triangle Park,
N. C.: Research Triangle Institute, May 1972.
Dixon, W. J. and F. J. Massey, Jr. Introduction to Statistical Analysis,
New York, N. V.: McGraw-Hill Book Company, Inc., 1957.
Environmental Protection Agency. Mount Storm. West Virginia-Gorman.
Maryland, and Luke. Marvland-Keyser. West Virginia. Air Pollution
Abatement Activity. Air Pollution Control Office Publ. No.
APTD-0656, Research Triangle Park, N. C., April 1971.
Bering, W. S. and T.-R. Borden, Jr. Ozonesonde Observations Over North
America, Vol. 4, Bedford, Massachusetts: Air Force Cambridge
Research Laboratories, 1967.
Hodgeson, J. A., B. E. Martin, and R. E. Baumgardner. "Laboratory
Evaluation of Alternate Chemlluminescent Approaches for the
Detection of Atmospheric Ozone," Presented at the American
Chemical Society Meeting, September 1970.
Jeffries, H. E. An Experimental Method for Measuring the Rate of
Synthesis. Destruction, and Transport of Ozone in the Lower
Atmosphere. ESE Publication No. 285, Chapel Hill, North Carolina:
University of North Carolina, 1971.
Lea, D. A. "Vertical Ozone Distribution in the Lower Troposphere Near
an Urban Pollution Center," J. Appl. Meteor. T_, 252-267, 1968.
Miller, P. R., M. H. McCutchan, and H. P. Mllligan. "Oxidant Air
Pollution in the Central Valley, Sierra Nevada Foothills, and
Mineral King Valley of California," Atmos. Environ. 6_, 623-633,
1972.
O'Keeffe, A. E. and G. C. Ortman. "Primary Standards for Trace Gas
Analysis," Anal. Chem. 38, 760-763, 1966.
Paetzold, H. K. "The Photochemistry of the Atmospheric Ozone Layer,"
In Chemical Reactions in the Lower and Upper Atmosphere, New
York, N. Y.: Interscience, 1961.
Pitts, J. N., Jr. Personal communication to L. A. Ripperton, 1972.
Public Health Service. Selected Methods for the Measurement of Air
Pollutants, Cincinnati, Ohio: R. A. Taft Sanitary Engineering
Center, 1965.
Rlcheer, H. G. "Special Ozone and Oxidant Measurements in Vicinity of
Mount Storm, West Virginia,"Research Triangle Park, N. C.:
Research Triangle Institute, October 1970.
Preceding page blank
105
-------
REFERENCES (Cont'd)
Rlpperton, L. A., H. Jeffries, and J. J. B. Worth. "Natural Synthesis
of Ozone in the Troposphere," Environ. Sci. and Tech. 5^ 246-248,
1971.
Scaringelli, F. P., et al. "Evaluation of Teflon Permeation Tubes for
Use with Suliur Dioxide," Presented in part at the American
Industrial Hygiene Association Annual Meeting, May 1966.
106
THE FOLLOWING PAGES ARE DUPLICATES OF
ILLUSTRATIONS APPEARING ELSEWHERE IN THIS
REPORT. THEY HAVE BEEN REPRODUCED HERE BY
A DIFFERENT METHOD TO PROVIDE BETTER DETAIL.
-------
'-A
-i "•
Research Triangle Institute Environmental Monitoring
Laboratory at Garrett County Maryland Airport
"S ~
If-
------- |