EPA-450/3-74-034
May 1974
INVESTIGATION
OF OZONE AND OZONE
PRECURSOR CONCENTRATION*
AT NONURBAN LOCATION*
IN THE
EASTERN
UNITED STATES
U.S. ENVIRONMENTAL PROTECTION AGENC
Office of Air and Waste Management
Office of Air Quality Planning and Standard)
Research Triangle Park, North Carolina 2771
-------
EPA-450/3-74-034
INVESTIGATION
OF OZONE
AND OZONE
PRECURSOR CONCENTRATIONS
AT NONURBAN LOCATIONS
IN THE
EASTERN UNITED STATES
by
Research Triangle Institute
Research Triangle Park, N. C. 27709
Contract No. 68-02-1077
EPA Project Officer: E . L . Martinez
Monitoring and Data Analysis Division
Office of Air Quality Planning and Standards
and
Contract No. 68-02-1343
Program Element No. 1HA326
EPA Project Officer: Elbert C . Tabor
Quality Assurance and Environmental Monitoring Laboratory
National Environmental Research Center
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N. C. 27711
May 1974
-------
This report is issued by the Environmental Protection Agency l,o repcri
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - as supplies permit - from the
Air Pollution Technical Information Center, Environmental Protection
Agency, Research Triangle Park, North Carolina 27711, or from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22151.
This report was furnished to the Environmental Protection Agency by
Research Triangle Institute, Research Triangle Park, N. C. , in fulfillment
of Contract Nos. 68-02-1077 and 68-02-1343. The contents of this report
are reproduced herein as received from Research Triangle Institute.
The opinions, findings, and conclusions expressed are those of the
author and not necessarily those of the Environmental Protection Agency.
Mention of company or product names is not to be considered as an endorse-
ment by the Environmental Protection Agency.
Publication No. EPA-450/3-74-034
11
-------
FOREWORD
During the summer and early fall of 1973 the Research Triangle
Institute conducted three distinct yet closely related studies under
two separ ite contracts with the Environmental Protection Agency. Since
they are so closely related, the reporting for the three studies i-3
presented in this volume. The principal portion of this report, which
is subtitled Part 1, "Field Measurements," presents the results of an
ozone and ozone precursor concentration measurement program conducted from
late June through October 1973 at McHenry, Maryland; Kane, Pennsylvania;
Coshoctcn, Ohio; and Lewisburg, West Virginia. Part 2, subtitled
"Quality Assurance Program," describes the procedures employed and the
results obtained in a study designed to evaluate the interrelatability of
the ozone and ozone precursor concentration measurements. Part 3,
subtitled "Airborne Ozone Monitoring Program," describes the use of
an instrumented C-45 aircraft to obtain ozone concentration measurements
aloft. Each of the parts is complete in itself; hence, the reader
interested in only one of the parts will not find it necessary to read
all three in order to obtain the information he seeks.
-------
ACKNOWLEDGMENTS
Many organizations and individuals contributed to the successful
performance of the studies described herein. The Board of County
Commissioners of Garrett County, Maryland, authorized the location of
an air monitoring station at the Garrett County Airport; its cooperation
and that of the airport manager, Mr. John Kreuzwieser, are gratefully
acknowledged. An air monitoring station was located at the Kane,
Pennsylvania, Area High School through the courtesy of the Kane Area
School Board. Mr. Verne Johnson, principal, Mr. Ed Bryant, and Mrs.
Karen Burton of the High School staff were most generous in offering their
assistance; it is here acknowledged. Through the cooperation of the
U.S. Department of Agriculture, an air monitoring station was located
at the North Appalachian Experimental Watershed at Coshocton, Ohio.
Messrs. W. Russel Hamon and William Bentz were particularly helpful
to the study; their contribution is greatly appreciated. The Bendix
Corporation operated a cooperative air monitoring station at Lewisburg,
West Virginia, during the period of study; their contribution is also
acknowledged.
Special acknowledgments are due the National Environmental Research
Center at Las Vegas for providing an aircraft and pilot for the airborne
ozone monitoring program. The advice and support of D. Wruble, R. Evans,
and V. Andrews are greatly appreciated. Thanks also to pilot J. Knight
for his flying expertise and cooperative spirit.
Finally, it is a pleasure to acknowledge the assistance of Research
Triangle Institute staff members. Messrs. C. E. Moore, R. W. Murdoch,
and S. R. Stilley assisted in the operation of the air quality analyzers
and in data reduction. Drs. W. K. Poole and T. D. Hartwell provided
guidance Ln the statistical analysis of air quality data, Mr. B. Crissman
assisted in the preparation of graphical materials.
v
-------
TABLE OF CONTENTS
pJ*fce
FOREWORD iij
ACKNOWLEDGEMENTS v
[,iST OF FIGURES xi
1.1 SI OF TABLES xvii
Part 1. Field Measurements
1 .0 INTRODUCTION 1-3
?.() PLAN OF STUDY 1-6
3,0 STUDY AREAS 1-9
3.1 Selection Rationale 1-9
3.2 Description 1-10
3.2.1 Garrett County, Maryland 1-10
3.2.2 McKean County, Pennsylvania 1-12
3.2.3 Coshocton County, Ohio 1-12
3.2.4 Greenbrier County, West Virginia 1-14
4.0 PROCEDURE 1-16
4.1 Description of Monitoring Stations 1-16
4.1.1 McHenry, Maryland (Garrett County Airport) 1-16
4.1.2 Kane, Pennsylvania 1-17
4.1.3 Coshocton, Ohio 1-22
4.1-4 Lewisburg, West Virginia 1-27
4,2 Instrumentation 1-30
4,3 Instrument Calibration and Maintenance 1-31
4.3.1 Logistical Considerations 1-31
4.3.2 Specific Calibration Techniques L-31
4,3.3 Maintenance of Air Pollution Monitors 1-33
4.4 Data Acquisition System 1 3'}
4.5 Data Reduction and Preliminary Processing 1-33
', ,U RESULTS 1- 37
5,1 Primary Data (June 26-September 30, 1973) 1-37
5.1.1 Summary Statistics 1-3/
5.1.2 Diurnal Patterns 1-46
5.1.3 Correlations 1-46
VII
-------
TABLE OF CONTENTS (Part 1 cont'd)
Page
5.2 Supplementary Data (October 1-November 2, 1973) 1-58
5.2.1 Summary Statistics 1-59
5.2.2 Diurnal Patterns 1-59
5.2.3 Effect of Change of Season 1-67
6.0 CONCLUSIONS 1-71
REFERENCES 1-72
APPENDICES
APPENDIX A - CALIBRATION METHODS AND PROCEDURES 1-73
REFERENCES 1-84
APPENDIX B - PERFORMANCE CHARACTERISTICS AND OPERATIONAL, 1-85
SUMMARIES FOR INSTRUMENTS
APPENDIX C - HYDROCARBON ANALYSIS OF GRAB SAMPLES 1-93
viii
-------
TABLE OF CONTENTS (cont'd)
Part 2. Quality Assurance Program
Section Page
1.0 INTRODUCTION 2-3
2.0 MOBILE LABORATORY AND EQUIPMENT 2-6
2.1 Mobile Monitoring Laboratory 2-6
2.2 Air Quality Analyzers and Calibration System 2-8
2.3 Data Acquisition System 2-11
3.0 FIELD QUALITY ASSURANCE PROGRAM 2-14
3.1 Procedure 2-14
3.2 Location and Description of Sites 2-17
3.3 Summary of Data Acquisition a Each Site 2-21
4.0 DATA COMPARISON 2-23
4.1 Lewisburg, West Virginia (August 25-31, 1973) 2-25
4.2 Kane, Pennsylvania (September 4-8, 1973) 2-29
4.3 Coshocton, Ohio (September 14-19, 1973) 2-32
4.4 Garrett County, Maryland (September 21-October 3, 1973) 2-35
4.5 Lewisburg, West Virginia (October 4-8, 1973) 2-41
5.0 STATISTICAL ANALYSIS 2-44
6.0 SUMMARY AND CONCLUSIONS 2-49
APPENDIXES
APPENDIX A: CALIBRATION SYSTEMS/PROCEDURES 2-53
APPENDIX B: DATA TABULATION 2-68
ix
-------
TABLE OF CONTENTS (cont'd)
Part 3. Airboxme Monitoring Program
Page
1.0 INTRODUCTION 3_3
2.0 MEASUREMENT SYSTEM • 3_4
2.1 Aircraft System 3_4
2.2 Ozone Analyzer 3_3
2.3 Calibration 3-12
3.0 DATA COLLECTION PROCEDURE 3_14
4.0 DATA SUMMARY 3_17
August 8, 1973 Flight 3_17
September 11, 1973 Flight 3_25
September 12, 1973 Flight 3_31
5.0 SUMMARY AND CONCLUSIONS 3_40
6.0 REFERENCES 3_44
-------
LIST OF FIGURES
Part 1. Field Measurements
Figure
1 Ozone and ozone precursor monitoring stations. 1-7
2 Photograph of a raised relief map illustrating 1-11
the topography in the vicinity of the McHenry,
Maryland (Garrett County Airport) monitoring
station.
3 Photograph of a raised relief map illustrating 1-13
the topography in the vicinity of the Kane,
Pennsylvania (Kane Area Senior High School)
monitoring station.
4 Photography of a raised relief map illustrating 1-15
the topography in the vicinity of the Lewisburg,
West Virginia (Greenbrier Valley Airport)
monitoring station.
5 Diagram of Garrett County Airport. 1-17
6 Exterior view of McHenry, Maryland (Garrett 1-18
County Airport) station.
7 Interior view of McHenry, Maryland (Garrett 1-19
County Airport) station.
8 Interior view of Kane, Pennsylvania station. 1-20
9 Diagram of Kane Area Senior High School 1-21
and grounds.
10 Arrangement of equipment in the mobile 1-22
laboratory at the Kane, Pennsylvania station.
11 Mobile laboratory at the Kane, Pennsylvania 1-23
station.
12 Plan view of interior of Coshocton, Ohio station. 1-24
13 Interior view of Coshocton, Ohio station. 1-25
14 Exterior view of Coshocton, Ohio station. 1-26
15 Greenbrier Valley Airport, Lewisburg, 1-27
West Virginia.
16 Interior plan view of mobile laboratory at 1-28
Lewisburg, West Virginia.
17 Exterior view of Lewisburg, West Virginia station. 1-29
18 Data acquisition system. 1-34
19 Sample page of printout of pollutant concentrations. 1-36
20 Frequency distributions of hourly ozone concen- 1-44
trations for July, August, and September, 1973
at McHenry, Maryland.
XI
-------
LIST OF FIGURES (Part 1 cont'd)
21 Frequency distributions of hourly ozone concen- 1-44
trations for July, August, and September, 1973,
at Kane, Pennsylvania.
22 Frequency distributions of hourly ozone concen- 1-45
trations for July, August, and September, 1973
at Coshocton, Ohio.
23 Frequency distributions of hourly ozone concen- 1-45
tration for July, August, and September, 1973,
at Lewisburg, West Virginia.
24 Mean diurnal ozone concentrations at McHenry, 1-50
Maryland; Kane, Pennsylvania; Coshocton, Ohio;
and Lewisburg, West Virginia from June 26 to
September 30, 1973.
25 Mean diurnal nitrogen dioxide concentrations at 1-51
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia from June 26 to September 30, 1973.
26 Examples of night-time ozone concentration maxima 1-52
at McHenry, Maryland and Coshocton, Ohio during
summer, 1973.
27 Selected episodes of high ozone concentration 1-53
(<160 yg/nH) at McHenry, Maryland; Kane,
Pennsylvania; and Coshocton, Ohio during
August, 1973.
28 Mean diurnal ozone concentration at Garrett County 1-54
Maryland Airport for June 26 to September 30, 1973,
and August 4 to September 25, 1972.
29 Mean diurnal ozone concentrations at Cleveland, 1-55
and Columbus, Ohio and Pittsburgh, Pennsylvania
during August 1971.
30 Frequency distribution of hourly ozone concen- 1-62
trations October 1-November 2, 1973 at Kane,
Pennsylvania; Coshocton, Ohio; and Lewisburp ,
West Virginia.
31 Mean diurnal concentrations of ozone and nitrogen 1-66
dioxide from October 1-31 and nonmethane hydrocarbon
from October 16-31, 1973 at Kane, Pennsylvania
32 Mean diurnal ozone concentrations at Kane, 1-67
Pennsylvania for June 26 to September 30 and
October 1 to 31, 1973.
xii
-------
LIST OF FIGURES' (Par-/, ,/ cont'd)
Figure Page
33 Mean diurnal ozone concentrations at Kane, 1-68
Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia from October 1 to November 2, 1973.
34 Mean diurnal nitrogen dioxide concentrations at 1-69
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia from October 1 to November 2, 1973.
Xlll
-------
LIST OF FIGURES (cont'd)
VaTt 2. QualiLy Assurance.
Figure Page
1 Interior view of Environmental Monitoring Laboratory
showing some of the ambient air analyzers 2-7
2 Interior view of Environmental Monitoring Laboratory
with data acquisition system in foreground 2-7
3 Sample data output for five-minute scan 2-12
4 Sample data output for hourly summary 2-13
5 General procedure schedule for Air Quality Assurance
Program 2-15
6 Lewisburg, West Virginia site 2-17
7 Kane, Pennsylvania site 2-18
8 Coshocton, Ohio site 2-19
9 Garrett County Airport, McHenry, Maryland site 2-20
10 Data collection and calibration dates; mobile lab
comparison periods 2-22
11 Fixed site/mobile lab comparison
August 27-September 1, 1973
Lewisburg, W. Va. 2-26
12 Quality Assurance Program - Lewisburg, W. Va.
August 29, 30, 31, 1973 2-27
13 Fixed site/mobile lab comparison
September 4-8, 1973
Kane, Pennsylvania 2-30
14 Quality Assurance Program - Kane, Pennsylvania
September 6, 7, 8, 1973 2-31
15 Fixed site/mobile lab comparison
September 14-19, 1973
Coshocton, Ohio 2-33
16 Quality Assurance Program - Coshocton, Ohio
September 18, 19, 1973 2-34
xiv
-------
LIST OF FIGURES (Par^l !> ront'd)
Figure Pag_e_
17a Fixed site/mobile lab comparison
September 21-25, 1973
Garrett County, Maryland 2-36
17b Fixed site/mobile lab comparison
September 26 - October 3, 1973
Garrett County, Maryland 2-37
18 Quality Assurance Program - Garrett County, Maryland
September 22, 23, 24, 1973 2-38
19 Quality Assurance Program - Garrett County, Maryland
September 24, 25, 26, 1973 2-39
20 Quality Assurance Program - Garrett County, Maryland
October 1, 2, 3, 1973 2-40
21 Fixed site/mobile lab comparison
October 4-8, 1973
Lewisburg, W. Va. 2-42
22 Quality Assurance Program - Lewisburg, W. Va.
October 5, 6, 7, 1973 2-43
A-l Ozone analyzer calibration system 2-56
A-2 Gas phase titration system 2-57
A-3 Gas phase titration of NO with 0,, 2-58
xv
-------
LIST OF FIGURES (cont'd)
Part 3. Airborne Monitoring Program
Figure Page
1. C-45 Aircraft Used for Flight Program Showing Sampling 3-5
Probe
2. Instrumentation for Airborne Ozone Measurement Program 3-6
3. Bluck Diagram of Instrumentation System 3-7
4. Ozone Meter Functional Diagram 3-9
5. Typical Signal Outputs for Solid Phase Ozone Meter 3-10
6. General Flight Path Showing Elevation Above MSL of the 3-15
Fixed Sampling Stations
7. September 8, 1973 Data Acquisition Flight 3-19
8. September 8, 1973 Data Acquisition Flight 3-20
9. September 9, 1973 Data Acquisition Flight 3-21
10. Ozone Concentrations for August 8, 1973 Respective Ground 3-22
Stations.
11. Illustration of Rapidly Increasing Ozone Concentration 3-23
12. Illustration of Rapidly Varying Ozone Concentration 3-24
13. September 11, 1973 Data Acquisition Flight 3-26
14. September 11, 1973 Data Acquisition Flight 3-27
15. September 11, 1973 Data Acquisition Flight 3-28
16. Vertical Measurements for September 11, 1973 3-29
17. Ozone Concentrations for September 11, 1973 Respective Ground 3-30
Stations
18. Airborne Ozone Measurements 3-32
19. September 12, 1973 Data Acquisition Flight 3-33
20. September 12, 1973 Data Acquisition Flight 3-34
21. Vertical Measurement for September 12, 1973 Data Acquisition 3-35
Flight
22. Ozone Concentrations for September 12, 1973 Respective 3-38
Ground Stations
23. Illustration of Rapidly Changing Ozone Concentration 3-39
Ozone Concentratior
September 12, 1973
3
24. Ozone Concentration (yg/m ) Constant Altitute Flight 3-42
xvi
-------
LIST OF TABLES
Part 1. Field Measurermnts
Table Page
1 PARAMETERS MEASURED AT MONITORING STATIONS 1-30
2 CALIBRATION TECHNIQUES 1-32
3 STATISTICAL SUMMARY OF HOURLY OZONE CONCENTRATION j-38
MEASUREMENTS BY STATION
4 STATISTICAL SUMMARY OF HOURLY NITROGEN DIOXIDE 1-38
CONCENTRATION MEASUREMENTS BY STATION
5 CUMULATIVE FREQUENCY DISTRIBUTIONS OF HOURLY 1-39
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (June 26-
September 30, 1973)
5a CUMULATIVE FREQUENCY DISTRIBUTIONS OF HOURLY ]_40
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (June 26-30, 1973)
5b CUMULATIVE FREQUENCY DISTIRBUTIONS OF HOURLY 1-4L
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (July 1-31, 1973)
5c CUMULATIVE FREQUENCY DISTRIBUTIONS OF HOURLY 1-42
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (August 1-31, 1973)
5d CUMULATIVE FREQUENCY DISTRIBUTIONS OF HOURLY i_/43
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (September 1-30, 1973)
6 MEANS AND STANDARD DEVIATIONS OF HOURLY CONCEN- 1-47 - l-/,9
TRATIONS OF OZONE, NITROGEN DIOXIDE, AND NON-
METHANE HYDROCARBON FOR EACH HOUR OF THE DAY
7 CORRELATION COEFFICIENTS FOR SIMULTANEOUS HOURLY 1-S6
OZONE CONCENTRATION VERSUS HOURLY NITROGEN
DIOXIDE CONCENTRATION WITHIN AND BETWEEN STATIONS
8 CROSS CORRELATION COEFFICIENTS FOR SIMULTANEOUS 1-57
AND LAGGED HOURLY OZONE CONCENTRATION BETWEEN
PAIRS OF STATIONS
9 CORRELATION COEFFICIENTS AND DIFFERENCES 1-58
SIGNIFICANT AT THE .05 LEVEL
10 STATISTICAL SUMMARY OF HOURLY OZONE CONCENTRATION }_^0
MEASUREMENTS BY STATION
xvn
-------
LIST OF TABLES (cont'd)
Table Page
11 STATISTICAL SUMMARY OF HOURLY NITROGEN DIOXIDE 1-60
CONCENTRATION MEASUREMENTS BY STATION
12 CUMULATIVE FREQUENCY DISTRIBUTIONS OF HOURLY 1-61
CONCENTRATIONS OF OZONE, NITROGEN DIOXIDE, AND
NONMETHANE HYDROCARBON BY STATION (Oct. l-Nov.2, 1973)
13 MEANS AND STANDARD DEVIATIONS OF HOURLY CONCEN- 1-63 - 1-65
TRATIONS OF OZONE, NITROGEN DIOXIDE, AND NON-
METHANE HYDROCARBON BY STATION
Part 2. Quality Assurance Program
1 AIR QUALITY ANALYZERS 2-8
2 CALIBRATION TECHNIQUES 2-10
3 CALIBRATION GASES 2-9
4 PAIRED COMPARISONS ON OZONE
MOBILE VAN VERSUS FIXED SITE 2-45
Part 3. Airborne Monitoring Program
1 FLIGHT OBSERVATION FOR SEPTEMBER 12, 1973 3-36 - 3-37
2 COMPARISONS GROUND AND AIRCRAFT OZONE DATA 3-41
xviii
-------
INVESTIGATION OF OZONE AND OZONE PRECURSOR
CONCENTRATIONS AT NONURBAN LOCATIONS IN
EASTERN UNITED STATES
Part 1. Field Measurements
by
D. R. Johnston
C. E. Decker
W. C. Eaton
H. L. Hamilton, Jr.
J. H. White
D. H. Whitehorne
Research Triangle Institute
Research Triangle Park, N. C. 27709
Contract No. 68-02-1077
EPA Project Officer: E. L. Martinez
Monitoring and Data Analysis Division
Office of Air Quality Planning and Standards
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N. C. 27711
May 1974
-------
INVESTIGATION OF OZONE AND OZONE PRECURSOR CONCENTRATIONS
AT NONURBAN LOCATIONS IN EASTERN UNITED STATES
Part 1. Field Measurements
1.0 INTRODUCTION
During the summer of 1972, the Research Triangle Institute (RTI)
conducted an intensive study of atmospheric ozone concentrations in
Garrett County, Maryland and Preston County, West Virginia.~ This
2/
study was predicated on earlier reports of unexpectedly high oxidant~~
3/
and high ozone— concentrations in the study area. The 1972 study
confirmed the earlier reports of high ozone concentrations; approximately
11 percent of 1043 hourly measurements made at the Garrett County
Maryland Airport during the summer of 1972 exceeded the National
Ambient Air Quality Standard (NAAQS) for photochemical oxidants
3
(160 yg/m not to be exceeded more than once per year). Nor was
the occurrence of high ozone concentrations restricted to the
Garrett County Maryland Airport—similar findings were obtained
at satellite locations approximately 19 km distant. The mean
hourly ozone concentration at the airport from August 4 to September 25,
1972 was 112 yg/m , the daytime hourly mean was 116 yg/m , and the
nighttime hourly mean was 108 yg/m . The maximum hourly ozone
concentration observed at the airport during the study period was
233 yg/m . In an urban area these concentrations would not be con-
sidered unusual and could be readily attributed to local photochemical
synthesis from nitrogen dioxide and hydrocarbon precursors. In
Garrett County, however, nitrogen dioxide and nonmethane hydrocarbon
concentrations were at,or near,geochemical levels throughout the
study period. Also, neither natural nor manmade sources in the
study area appeared capable of producing the precursor quantities
required for synthesis of high ozone concentrations. Ozone synthesis
4/
from naturally occurring precursors has been demonstrated;" however,
it was considered unlikely that this process could account for sustained
1-3
-------
concentrations exceeding the standard. The persistence of high ozone
concentrations at night reflected the low precursor (and destructive
agent) concentrations and obviously was not the result of ongoing
photochemical processes. Thus, an evaluation of air quality measure-
ments suggested that local photochemical synthesis could not account for
the observed high concentrations of ozone.
Analysis of synoptic meteorological data, as well as an examination
of ozone wind roses for the Garrett County Maryland Airport, indicated
that the high ozone concentrations at this location developed within
particular air masses which acquired their characteristics over
broad regions of urban-industrial activity. High concentrations of
ozone did not appear to be associated with direct transport from one
or more identifiable local point or area sources of precursors.
It appeared that air, moving over urban and/or industrial areas on
its way to the study area, acquired ozone precursors which, upon
irradiation with sunlight, yielded ozone. The sharp decrease in ozone
concentration in the relatively rapidly moving air arriving behind
frontal systems supported this suggestion.
The occurrence of high ozone concentrations at nonurban locations
may signal a general deterioration of air quality. Indeed, Mineral
King Valley and other rural California locations have experienced
high oxidant concentrations.— The extent to which this occurs in
eastern United States, however, is unknown. The research described
in this report constitutes Phase I of an investigation of the areal
extent of the occurrence of high ozone concentrations at nonurban
locations in eastern United States. The objectives of the research were:
1) To provide a data base of nonurban ozone and ozone precursor
concentration measurements for future detailed analysis.
2) To provide statistical summaries of the frequency of occurrence
of concentrations of ozone, nitrogen dioxide, and nonmethane
hydrocarbons at the selected locations.
3) To determine the interrelationship between ozone concentration
and concentrations of nitrogen dioxide and nonmethane hydro-
carbons at the selected locations.
1-4
-------
Phase IT of the study will investigate the influence of synoptic-
scale meteorological conditions on the ozone and ozone precursor concen-
trations measured at the selected locations.
1-5
-------
2.0 PLAN OF STUDY
A summer program of concurrent field measurements of ozone and
ozone precursor concentrations at nonurban locations was conducted from
June through September 1973. The field measurement program employed a
four-station monitoring network with stations at/or near McHenry, Garrett
County, Maryland; Kane, McKean County, Pennsylvania; Coshocton,
Coshocton County, Ohio; and Lewisburg, Greenbrier County, West Virginia,
(Figure 1). The station at Lewisburg, West Virginia, was a cooperative
station operated by the Bendix Corporation under RTI supervision.
Ozone concentrations were measured at all stations, and nitrogen dioxide
and nonmethane hydrocarbon concentrations were measured at all stations
except McHenry, Maryland. During the study the sponsor requested that the
program be supplemented by continuing measurements through October 31 at
the Kane, Pennsylvania, station.
In the analysis of the data collected during the summer of 1972—
it was found that local surface wind speed and temperature values showed
no significant relationship to the ozone concentration values. However,
analysis of meteorological data on a synoptic scale suggested relationships
that warranted further investigation. Accordingly, synoptic scale
meteorological data collected routinely by the National Weather Service
were to serve as the primary base for the Phase II analysis. Wind speed and
direction were recorded at the McHenry, Maryland, and Lewisburg, West
Virginia, stations for subsequent analysis as appropriate.
Ozone concentration data were collected at the four stations
to provide further insight into the areal extent of the occurrence
of high ozone concentrations. The frequency of occurrence of ozone,
nitrogen dioxide, and nonmethane hydrocarbon concentrations was deter-
mined from an examination of the frequency distribution of hourly mean
concentrations of each pollutant at each station. The statistical relation-
ship between ozone concentrations and nitrogen dioxide concentrations was
determined using correlation techniques. Lag correlation coefficients
were computed to examine the association between ozone concentrations at
1-6
-------
;.®. Cinclnnaii
*• ^ ~ •" } Charleston
KENTUCKY j
.^--^
SCALE
Station No.
1
2
3
4
Town
Me Henry
Kane
Coshocton
Lewlsburg
7 8
(km X 100)
County
Garrett
McKean
Coshocton
Greenbrier
Figure 1. Ozone and ozone precursor monitoring stations
(Stippled areas indicate the identified counties)
1-7
-------
pairs of stations. Data were compared with data obtained in 1972 at
McHenry, Maryland, and with the appropriate NAAQS. Data for Kane,
Pennsylvania, were examined to determine what influence, if any, the
change of season has on ozone concentrations.
1-8
-------
3.0 STUDY AREAS
A discussion of the rationale for the selection of study areas
and descriptions of the areas selected are presented in this section.
3.1 Selection Rationale
The rationale for selecting study areas is based on previous
observations of high ozone concentrations at McHenry, Maryland.
The horizontal extent of the; region included in the study area
was determined by considering several factors. First, station
distances from McHenry must be such that there would be small risk
of the stations falling outside of the suspected region of high ozone
concentrations. Second, the separation between stations must be
great enough to offer a reasonable expectation of having a well-
defined air mass front lying between one or more pairs of stations.
Consideration of these factors led to the selection of outlying
stations within approximately 250 km of the McHenry station. An
additional criterion, inherent in the nature of the study, was
that all stations be remote from metropolitan centers to avoid
direct contamination by ozone precursors from urban-industrial
activity.
McHenry, Maryland (Garrett County Airport) was the pivotal
station of the sampling array. The ozone and ozone precursor data
base established there in 1972 provided historical guidance and a
benchmark. Without this station, there would be no way of knowing
if additional measurements were characteristic or anomalous.
Kane, Pennsylvania is remote from any major population center
or industrial area. The nearest principal city is Erie, Pennsylvania,
121 km to the west-northwest; Buffalo, New York is 129 km north; and
Pittsburgh, Pennsylvania and Youngstown, Ohio are about 160 km to
the southwest. The nearest known source of ozone precursor material
is an oil refinery at Bradford, Pennsylvania, 32 km to the north;
however, the general area often lies downwind of the industrial areas
of Lake Michigan and Lake Erie.
1-9
-------
Coshocton, a small town in rural east-central Ohio, is located
closer to urban-industrial areas than any of the other sites selected.
Akron, Canton, Youngstown, Wheeling, Columbus, and Cleveland are
within a 130-km radius of Coshocton. There are several potential
sources of nitrogen oxides and hydrocarbons from nearby industry.—
Lewisburg, West Virginia, lies in the Greenbrier River valley
in the southern portion of the state. Ozone concentrations in
excess of the NAAQS had been reported for this area.— The Lewisburg
station gave a southern extension to the sampling array and provided
an opportunity to measure ozone concentrations in mountainous
terrain at a site other than McHenry, Maryland.
3.2 Description
Brief descriptions of the four counties in which air monitoring
stations were located are given below.
3.2.1 Garrett County, Maryland
The westernmost county in Maryland, Garrett County is
bounded on the east by the North Branch of the Potomac River, on
the west by West Virginia, and on the north by Pennsylvania. It
lies in a physiographic transition zone with both the folded
Appalachian and Appalachian plateau land forms present (Figure 2).
Elevation above mean sea level (MSL) varies from approximately
760 to 914m. Backbone Mountain, the highest point in Maryland,
rises to 1024m MSL.
At a station located at Deep Creek the January average temperature
for an 11-year period was 271.7K, while the July average was 293.IK.
The maximum temperature observed at this station was 308.7K, the
minimum 242.6K. The first killing frost in the fall usually occurs
about October 2, ending a 130-day growing season. Annual average
3
precipitation is about 1.14 x 10 mm.
The 1970 population of Garrett County was 21,476 and the
2
population density was 12.6 per million m .
Estimated annual emissions of nitrogen oxides and hydrocarbons
are 4.34 x 10 and 6.95 x 10 kg, respectively. Annual carbon
1-10
-------
Air Monitoring Station No. 1
Figure 2. Photograph of a raised relief map illustrating the
topography in the vicinity of the McHenry, Maryland
(Garrett County Airport) monitoring station.
1-11
-------
monoxide, particulate and sulfur dioxide emissions are estimated
at 32.7 x 105, 1.48 x 105, and 3.14 x 105 kg, respectively.*
3.2.2 McKean County, Pennsylvania
Bordering New York state NNW of Pittsburgh, McKean
County is typical of the Allegheny Plateau of western
Pennsylvania (Figure 3). Throughout much of the county,
elevations are about 610m MSL.
At Bradford the average January temperature for 14 years of
record was 268.8K; the July average was 293.6K. The maximum
temperature observed during the same period was 310.9K; the
minimum, 242K. Growing seasons approximate 120 days, ending with
the first killing frost about September 23. Annual precipitation
3
averages about 1.07 x 10 mm.
The 1970 population of McKean County was 51,915; the population
2
density was 20.2 per million m . In 1970 the Borough of Kane had
a population of 5,001.
Estimated annual emissions of nitrogen oxides and hydrocarbons
are 47.3 x 10 and 49.5 x 10 kg, respectively. Annual carbon
monoxide, particulate, and sulfur dioxide emissions are estimated
at 220.1 x 105, 85.5 x 10 , and 54.8 x io5 kg, respectively.*
3.2.3 Coshocton County, Ohio
East-northeast of Columbus and SSW of Canton, Coshocton
County is situated near the western terminus of the unglaciated
Allegheny Plateau. The elevation of the county varies from about
229 to 335m MSL.
At the town of Coshocton, January temperatures average 271.6K
and those for July average 296.9K. The maximum temperature, at
this station for a 23-year period was 314.2K, the minimum, 2.40.9K.
The growing season is approximately 162 days, ending with the first
killing frost in the fall about October 12. Average annual precipi-
3
tation is about 1.04 x 10 mm.
The 1970 population of Coshocton County was 33,486, while the
town of Coshocton had a population of 13,747. Population density
2
for the county was 23 per million m .
*Emission estimates in tons per year were provided by the Office of Air Quality
Planning and Standards, U.S. Environmental Protection Agency; those estimates
were converted to kilograms per year.
1-12
-------
Figure 3.
Photograph of a
;='=
1-13
-------
Estimated annual emissions of nitrogen oxides and hydrocarbons
are 269.9 x 10 and 33 x 10 kg, respectively. Annual carbon
monoxide, particulate, and sulfur dioxide emissions are estimated
at 208.7 x IQ5, 223.2 x 1Q5, and 1065.9 x 1Q5 kg, respectively.*
3.2.4 Greenbrier County, West Virginia
East-southeast of Charleston, West Virginia, Greenbrier
County is situated in the folded, or new Appalachians (Figure 4).
Elevations in the county range from about 518m MSL to 1333m MSL
at Grassy Knob.
At Lewisburg the average January temperature was 273.3K based
on 37 years of record and the July average for the same period
was 294.8K. The maximum temperature observed at this station was
312K, the minimum, 234.8K. The growing seasons is approximately
150 days, ending with killing frost about October 5. Annual
3
precipitation averages about 1.02 x 10 mm.
In 1970 the population of Greenbrier County was 32,090, while
the population of Lewisburg was 2,407. County population density
2
was 12.1 per million m in 1970.
Estimated annual emissions of nitrogen oxides and hydrocarbons
are 13.4 x 10 and 19.3 x 10 kg, respectively. Annual carbon
monoxide, particulate, and sulfur dioxide emissions are estimated
at 75.6 x 105, 3.4 x 1Q5, and 9.3 x 105 kg, respectively. *
*Emission estimates in tons per year were provided by the Office of Air
Quality Planning and Standards, U.S. Environmental Protection Agency; those
estimates were converted to kilograms per year.
1-14
-------
Air Monitoring Station No. 4
Figure 4. Photograph of a raised relief map illustrating the
topography in the vicinity of the Lewisburg, West
Virginia (Greenbrier Valley Airport) monitoring
station.
1-15
-------
4.0 PROCEDURE
This section describes the procedures employed to obtain
measurements of air quality and meteorological parameters. The
principal criteria for the selection of a monitoring station
location were:
1) that the location be free of natural and manmade
obstructions to air movement;
2) that it be at a higher elevation than the surrounding
terrain, and
3) that it be removed from local sources of ozone and ozone
precursors.
4.1 Description of Monitoring Stations
4.1.1 McHenry, Maryland (Garrett County Airport)
The McHenry, Maryland station was located at the Garrett
County Airport, approximately 884m MSL. The airport complex consists
of a 762-m paved landing strip, an apron, hangars, and a terminal
building. The airport is used mainly by small, private aircraft.
Figure 5 is a diagram of the airport which shows the location
of the air monitoring station and the meteorological tower. While
the site provided excellent exposure for the air monitoring
instruments, exposure of the meteorological instruments was less
than optimal. An exterior view of the station is shown in Figure 6.
The aircraft shown in Figure 6 was used to transport equipment and
personnel from station to station.
The only pollutant concentration measured at this station was
ozone. The ozone monitor, stripchart recorder, and data acquisition
system was located in a small (2.4m x 3m x 3m) workroom at the east
end of the hangar. An interior view of the station is shown in
Figure 7. An air conditioner was mounted in the wall to maintain
temperatures near 298K. Ambient air was sampled through Teflon lines
from an intake mounted 1.5m above the hangar roof. The hangar roof
is about 3.7m above grade.
1-16
-------
HANGAR
WORKROOM
(LOCATION OF INSTRUMENT)
APRON
762m RUNWAY
PARKING
AREA
GAS
PUMPS
70 105 140 175 210
t i i i i
TOWER
TERMINAL
BUILDING
Meters
Figure 5. Diagram of Garrett County Airport.
Wind speed and wind direction sensors were mounted on a 9.2m
tower which was braced to the terminal building. A dual channel
recorder for wind speed and direction was located in the airport
manager's office.
4.1.2 Kane, Pennsylvania
Surrounded on three sides by the Allegheny National
Forest, the Borough of Kane is located approximately 160km north
of Garrett County, Maryland. The Kane, Pennsylvania monitoring
station was located in the industrial arts room of the Kane Area
Senior High School. An interior view of the station is pictured
in Figure 8. Situated at the highest point in the Borough (630m MSL),
the school provides excellent instrument exposure from all directions.
A diagram of the school grounds is shown in Figure 9.
Ozone, nitrogen dioxide, and nonmethane hydrocarbon concentra-
tions were measured and recorded at this station. Instruments were
placed on tables inside the industrial arts room. The air inlet
was at a point 1.8m above the roofline. Teflon tubing connected
the inlet to a glass manifold. The manifold was brought into the
room through a plywood panel which replaced a window.
1-17
-------
..
Figure 6. Exterior view of McHenry, Maryland
(Carrett County Airport) station.
1-18
-------
Figure 7. Interior view of McHenry, Maryland
(Garrett County Airport) station.
1-19
-------
Figure 8. Interior view of Kane, Pennsylvania station.
1-20
-------
RAILROAD
KEY:
(T) High School
(T) Industrial Arts Room and Bench With Analyzers
(T) Mobile Laboratory
Ut) Additional Building Constructed During Study
Figure 9. Diagram of Kane Area Senior High School and grounds.
1-71
-------
On September 8, 1973, when classes resumed, the instruments
were transferred to a mobile laboratory located immediately outside
the industrial arts room. The ambient air intake was not moved.
Equipment was arranged in the mobile laboratory as shown in Figure 10.
The mobile laboratory is pictured in Figure 11.
4.1.3 Coshocton, Ohio
The North Appalachian Experimental Watershed 16km
northeast of Coshocton was the location of the Coshocton, Ohio
station. Operated by the United States Department of Agriculture,
this complex of buildings is 354m MSL. The acreage surrounding
the station is used mainly for farming. There are no apparent
obstructions to air movement.
MANIFOLD ENTRANCE
AND MANIFOLD
DATA
ACQUISITION
SYSTEM
Figure 10. Arrangement of equipment in the mobile
laboratory at the Kane, Pennsylvania
station.
1-22
-------
Figure 11. Mobile laboratory at the Kane, Pennsylvania station.
1-23
-------
Nitrogen dioxide, ozone, and nonmethane hydrocarbon analyzers
were housed in a room on the second floor of the Engineering Building.
The room was modified by adding temporary panelling (to make a smaller
room), an air conditioner, and another electrical circuit. A plan view
of the station interior is shown in Figure 12 and a photograph of
the equipment arrangement is shown in Figure 13.
An exterior view of the station, which shows the air intake
and Teflon intake tubing is shown in Figure 14.
STAIRS
DOOR
KEY:
ooo
>**S.~ "1
-C2l J
Ozone
NMHC
N02
Data Acquisition System
Air Conditioner
Ambient Air Manifold
Table for Calibration Gear
Figure 12. Plan view of interior of
Coshocton, Ohio station.
1-24
-------
Figure 13. Interior view of Coshocton, Ohio station.
1-25
-------
Figure 14. Exterior view of Coshocton, Ohio station.
1-26
-------
4.1.4 Lewisburg, West Virginia
The Lewisburg, West Virginia station was located at the
Greenbrier Valley Airport, approximately 160km southwest of Garrett
County, Maryland. The airport elevation is approximately 705m MSL.
The airport serves both private and commercial aviation. The valley
surrounding the airport is mainly gently rolling pastureland with
some wooded areas.
Analyzers and associated equipment were housed in an air-conditioned
2.5 x 9.2m mobile laboratory owned by the Bendix Corporation, Ronceverte,
West Virginia. A 9.2-m, guyed tower was located 15m from the laboratory.
Wind speed and wind direction transducers were mounted on it.
The general features of the airport are shown in Figure 15.
TOWER
e
APRON AND AIRPLANE STORAGE
RUNWAY
Figure 15. Greenbrier Valley Airport, Lewisburg, West Virginia.
1-27
-------
Figure 16, a diagram of the interior of the mobile laboratory,
shows the location of air monitors, recorders, and the data acquisition
system. An exterior view is given in Figure 17.
Ozone, nitrogen dioxide, and nonmethane hydrocarbon concentrations
were measured and recorded at this station. Wind speed and wind
direction were recorded intermittently.
t \ 2J \ , — ^^
J~ \
] BENCH WITH CABINETS
©
BENCH _
O DOOR (lO)
©
1 '
KEY:
CO Intake Manifold (mounted on ceiling
©
r
[
1
}
Q:
DOOR
1 i
Spc
©
BENCH
/
O
o
o
ice for Other Insl
enters through ceiling)
Data Acquisition Board
WS-WD Recorder
N0-N09-N0 Analyzer and Recorder
&- X
0. Analyzer and Recorder
Hydrocarbon Analyzer and Recorder
Space for RTI Calibration Assembly
(9 Clean Air Supply
(concealed beneath bench)
(10) Electric Panel
Hi) Exhaust Manifold; Electric
Power Strip
^2) Heating System (air
conditioner is mounted
on the ceiling)
Qj) Compressed Gas Tanks
(Outside)
(u) Telephone
Figure 16. Interior plan view of mobile laboratory at Lewisburg,
West Virginia.
1-28
-------
Figure 17. Exterior view of Lewisburg, West Virginia station.
1-29
-------
4.2 Ins trumentat ion
The Instruments used to obtain measurements of air quality
and meteorological parameters are described below. Parameters
measured at each station are summarized in Table 1. In addition,
wind speed and wind direction were monitored at McHenry, Maryland
and Lewisburg, West Virginia.
Ambient ozone concentrations were measured at three of the
stations using the Bendix Model 8002 Chemiluminescent Ozone
Analyzer, while a McMillan Corporation MEC 1100 Chemiluminescent
Ozone Analyzer was used at the Kane, Pennsylvania station.
Nitrogen dioxide was measured at the three stations identified
in Table 1 using the Bendix Model 8101B Chemiluminescent NO,
NO-, NO Analyzer. Nonmethane hydrocarbon concentrations were
£* A
determined at the three stations identified in Table 1 using the
Bendix Model 8201 Ambient Hydrocarbon Analyzer.
Table 1. PARAMETERS MEASURED AT MONITORING STATIONS
Station
Parameters
McHenry, Maryland
Kane, Pennsylvania
Coshocton, Ohio
Lewisburg, West Virginia
0~> windspeed and direction
03, N02, NMHC
03, N02, NMHC
03, NO , NMHC, windspeed
and direction
1-30
-------
An air sampling system consisting of a 2.5-cm I.D. Teflon inlet
line, a glass manifold with sampling ports, and a blower arrangement
(to aspirate sample air from outside the shelter through the manifold
system) was used at each station. An inverted glass funnel was used as an
inlet to prevent moisture and settleable particulates from entering the
system. Each analyzer sampled air drawn from the inlet manifold through a
5~urn porosity Teflon filter with Teflon filter element and Teflon tubing.
Wind speed and wind direction were measured at two sites with a
Climet Model 013-6 Wind System. The wind speed and wind direction
transmitters were mounted on 9.2-m, guyed towers.
4.3 Instrument Calibration and Maintenance
4.3.1 Logistical Considerations
Dynamic calibration techniques were used to calibrate
each analyzer at two-week intervals during the measurement period.
Data obtained from the calibrations were used to provide updated
transfer equations for converting voltage output to the physical
units, micrograms per cubic meter.
The logistical problem of calibration and maintenance of ten
instruments at four widely separated locations was approached as
follows. Cylinders containing known concentrations of nitric oxide
and methane were carried to each site requiring them and left for
the duration of the study. A collection of items necessary to
calibrate the instruments was assembled and packed such that it
could be carried in a small, charter airplane.
4.3.2 Specific Calibration Techniques
Table 2 summarizes the calibration techniques for the
pollutants monitored.
4.3.2.1 Calibration of Chemiluminescent Ozone Detector. Dynamic
calibration of the ozone analyzers was achieved by use of an ultraviolet
ozone generator. The ozone source consists of a shielded mercury vapor
lamp (20.3-cm in length) which irradiates clean (compressed) air flowing
through a 1.6-cm diameter quartz tube at a rate of 5 liters/minute.
Variable ozone concentrations (0-1 ppm) can be generated by variable
*As specified in 40 CFR 50, Appendix D.
1-31
-------
Table 2. CALIBRATION TECHNIQUES
Pollutant
Technique
Ozone
Nitric oxide/nitrogen
dioxide
Nonmethane hydrocarbons
Ultraviolet ozone generator
referenced to neutral-buffered
potassium iodide
Nitric oxide-ozone conversion
unit
Standard calibration gas
certified as to CH, content
4
shielding of the lamp envelope. As a reference method, the neutral-
buffered potassium iodide (KI) technique was used. The concentration
of ozone at each calibration point was verified by the KI method.
4.3.2.2 Calibration of Chemiluminescent N0-N00-N0 Detector. The
z x
gas-phase titration technique developed by Hodgeson and associates
8/
at EPA— was used for the dynamic calibration of the NO, NO , N0«
X £.
analyzer. The technique is based on the rapid gas-phase reaction
between NO and 0,, which produces a stoichiometric quantity of N0_.
Nitric oxide (-50 ppm) in nitrogen (N~) is diluted with zero air to
provide NO concentrations in the range 0.01 to 0.50 ppm and used to
calibrate the NO and NO channels. Nitrogen dioxide concentrations
X
(0.01 to 0.50 ppm) are produced by the quantitative reaction of ozone
with NO.
4.3.2.3 Calibration of Nonmethane Hydrocarbon Flame lonization
Detector. Calibration of the hydrocarbon analyzer was accomplished
using mixtures of stable gases prepared to exact concentration in
pressurized cylinders. Mixtures of methane in air purchased from
Scott Research Laboratories with a certificate of analysis as to the
methane concentration were used for all calibrations.
A detailed discussion of calibration methods and procedures is
given in Appendix A.
1-32
-------
4.3.3 Maintenance of Air Pollution Monitors
Maintenance of the automated instruments took place during the
regularly scheduled biweekly calibration trips. Thus, no alterations
affecting the response of the instrument (as may be caused by maintenance)
occurred between calibration periods other than normal drift or analyzer
failure. Specific maintenance included renewal of chart paper, ink,
and magnetic tape, changing the 5-ym porosity Teflon filter element
on the inlet line of each instrument, adjusting sample flow rate and
reactant flow rates to those specified by the instrument manufacturer, and
adjusting temperature controls such that the ideal room temperature of
298K was maintained as nearly as possible.
When failures occurred,, the instrument was brought back on line as
soon as possible. Specific failure time spans and reasons are summarized
in Appendix B, "Performance Characteristics and Operational Summaries for
Instruments."
4.4 Data Acquisition System
The magnetic tape data acquisition system used was the Westinghouse
Pulse-o-Matic recording system which provided 15-minute integrated
concentrations. A battery backup unit is included in the recorder to
preserve the time information on tape in the event of a power failure.
A photograph of one of the data acquisition systems is shown in Figure 18.
Each of the small weather-tight enclosures contains a one pulse
transmitter. The larger box contains the recording unit and battery
backup components.
The tapes were replaced biweekly and returned to the Westinghouse
Electric Corporation, Meter Division, Raleigh, North Carolina for
translation to a form compatible with the RTI computer facility.
4.5 Data Reduction and Preliminary Processing
The data manipulation necessary to recover the data stored on a
magnetic tape consisted of two phases:
1) translation of the tape to a form compatible with available
data processing equipment, and
2) processing the data on a computer to obtain concentrations in
3
1-33
-------
Figure 18. Data acquisition system.
1-34
-------
The tape translation was done by the Meter Division, Westinghouse
Electric Corporation. In their process, the tape is rewound and the pulses
counted between each time mark. From information supplied by the user,
including start and stop times, and times of any power failures, the actual
date and time for each interval was computed. This information,
both pulse counts and corresponding times, was printed out in
tabular form and punched on IBM compatible cards for further
processing. The data at this stage of processing were in the form
of counts and had to be divided by 1500 to determine the average
voltages.
The second phase consisted of processing the data on a computer
to produce a printout for visual examination and a magnetic tape
for use in the subsequent analysis of the data. In order to produce
a printout which contained all information within itself, certain
supplementary data had to be supplied to the computer along with the
pulse counts. These data include times when the instruments were
inoperative or not functioning correctly and linear best-fit equations
relating the voltage output of the instrument to the concentration
of the pollutant being measured. The times for instruments being
inoperative came from operator logs, calibration notes, examination
of preliminary computer runs, and strip charts. The equations were
derived from data obtained during calibrations—the known input gas
concentrations and the resulting voltage output from the instrument.
A regression analysis was performed on these points to obtain a
best-fit equation characterizing the instrument's response. The
printout produced by the computer then contained the data in pg/m .
If data were absent or invalid the number 99999 and a code letter
were inserted to indicate the reason for the loss of data. A sample
page of printout is shown in Figure 19. The magnetic tape which has
the data recorded iri computer compatible form contains exactly the
same data less extraneous headings and spaces.
1-35
-------
LJ C5 O LJ O
OOOOO OOOOO OOOOO OOOOO OOOOO OOOOO
:> o o o :r o-o-oo-a o o- > o o ;> o o o o- ooooo
•< OOOOO OOOOO O^OOOO OOOOO OOOOO
•> ooooo o-oo-oo ooooo ooooo o o o- o- o
• OOOOO OOOOO OOOOO OOOOO OOOOO
-%
(\J
OO LJCJOUC'OOOOCT
o: ^>
z>
O2 O«)-iH^-in O •H^O'^ OJCMCXJcD-O ^-p-trimif> OOOOO OOOOO
(/) O t\J 1-1 «-t OOOOOOOOOO
— OOOOOOOOOJ-
3: oaooaoc-ooo
LU OOOOOOOOOO
I u_
2?
O UCJOUC- L.IC^OLJC' C)Ut_CJC- ULJC.OCZ
fSJ
OOOOO OOOOO OOOOO OOOOO •^-CXJOCD'O CUO-OtDT-l
ooooo ooooo ooooo ooooo ojf\jc\j<\iry rv>c\jcxi»-toj
j-ot>oa ao-o-o-o o-oooo ocT'Ooo
ooooo ooooo ooooo ooooo
LJ
T
7- .....
2* CI>OC3OO OOOOO OOOOO OOOOC OOOOO OOC3OO
o
T
^
O CM
»- c ..... ....*
C T-a-roror-o rocxcvroro Kiror^roro fOfOPOforrj rorororOro roc\,f\.roro
T
cr
(~)
C
^
o ..... ..... ..... ..... ..... ....,
~^J lf>LTl*TO33 ^"TCXTT LOOCX/r^-cO LOi*\jC\jr^.r^- ^JiPf\ilO»H OOf^-CMfXi
T C3OOOO C3OOC3C3 OCDOC3C3 QOC3OO OO^OO OCDOOO-
2:
~z
-
cr 'L .....
LU O coOOOO ;Gr-.f^-:or^ ^OCDOO CDOOOO oor^r--otr.tt otriO'j>< oipoirt< oinou^< oiTioj><:
—• '"\ir\jc\it\jLU 'oto^Of'J'jj ^•^••^•^•uj iAiou>Lnuj ^•o^o-Ouj r^-r^r^r^uj
u z
" W
U- T
u. a.
•"" Q. E;
Z Q. O
2 -I CC
— < CD
O L) —
rr —t
LLi LL O
M CD Z.
o
-H
0)
o
o
u
rt
4J
MH
O
4J
?
O
J~>
0
-H
H
pu
0)
60
ft
<1J
i-H
i"
n)
00
•H
1-36
-------
5.0 RESULTS
The results of the field measurement program are presented herein,
with both primary and supplementary data included. Summary statistics,
diurnal pollutant concentration patterns, and correlation coefficients,
as appropriate, are presented below.
5.1 Primary Data (June 26-September 30, 1973)
5.1.1 Summary Statistics
Mean hourly ozone concentrations, standard deviations, and
case counts by station are shown in Table 3; the corresponding statistics
for nitrogen dioxide are given in Table 4. During this period 946, 1118,
and 1334 determinations of hourly nonmethane hydrocarbon concentrations
were made at Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg, West
Virginia, respectively. These observations were discarded when it was
learned that the hydrocarbon analyzers used were affected by water vapor,
thus giving erroneously high nonmethane hydrocarbon concentrations.
Cumulative frequency distributions for hourly pollutant concentrations
for the entire period are presented in Table 5. Maximum hourly ozone
concentrations (concentration exceeded during approximately one percent
3 3
of hours) were 320 Pg/m at McHenry, Maryland; 270 yg/m at Kane,
3 3
Pennsylvania; 340 ug/m at Coshocton, Ohio; and 250 yg/m at Lewisburg,
West Virginia.
During the field measurement program, RTI chemists became suspicious of the
high nonmethane hydrocarbon concentrations being determined with the Bendix
instrument. Their suspicions were confirmed by the quality assurance
program (described elsewhere in this volume) which indicated that the non-
methane hydrocarbon concentrations determined with the Bendix instrument
were indeed high when compared with simultaneous determinations made with
the Beckman Model 6800 Gas Chromatographic Flame lonization Detector. At
the request of RTI, Bendix Corporation personnel investigated the problem
and determined that total hydrocarbon measurements made with their analyzer
were subject to a positive moisture interference. Hence, nonmethane
hydrocarbon concentrations obtained by subtracting the methane concentration
from the total hydrocarbon concentration were erroneously high. Since the
moisture interference was not defined quantitatively, there was no basis
for adjusting and thereby recovering earlier nonmethane hydrocarbon
concentration determinations. Bendix Corporation modified the analyzer at
the Kane, Pennsylvania station in mid-October by adding a column to remove
water vapor from the air steam used for the total hydrocarbon concentration
measurement. Subsequent measurements at Kane, Pennsylvania were considered
valid.
1-37
-------
Table 3. STATISTICAL SUMMARY OF HOURLY OZONE CONCENTRATION
MEASUREMENTS BY STATION
June 26-September 30, 1973
Station
McHenry , Maryland
Kane, Pennsylvania
Co shoe ton, Ohio
Lewisburg, West Virginia
Mean hourly
concentration
(yg/m )
148.9
130.3
111.5
105.4
Standard
deviation
(yg/m )
56.0
56.8
62.7
53.5
Case count
1,622
2,131
1,785
1,663
Table 4. STATISTICAL SUMMARY OF HOURLY NITROGEN DIOXIDE
CONCENTRATION MEASUREMENTS BY STATION
June 26-September 30, 1973
Station
Kane, Pennsylvania
Co shoe ton, Ohio
Lewisburg, West Virginia
Mean hourly
concentration
(yg/m )
11.3
21.4
16.3
Standard
deviation
(yg/m")
8.0
15.4
14.6
Case count
1,869
2,043
1,699
Tables 5a,b,c, and d show the cumulative frequency distributions for
hourly pollutant concentrations for June, July, August, and September,
respectively. The frequency distributions of hourly ozone concentrations
by month are plotted in Figures 20, 21, 22, and 23 for McHenry, Maryland;
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg, West Virginia,
respectively.
The mean hourly ozone concentrations shown in Table 3 can be
3
compared with the 112 yg/m mean hourly concentration observed during the
(Text continued on page 1-46)
1-38
-------
13
W
O
O
Pi
H '-v
H CO
1Z r-
cyi
« rH
W
IS "
O 0
N CO
O
M
1*4 0)
0 .0
W 0)
a ^
0 ft
i— I nl
DNCENTRAT
ne 26 - S
U 3
!->
PH v_x
^»
P 0
O H
*3
h H
O CO
CO >H
S pq
O
SB
P PQ
CY DISTRII
E HYDROCAI
E FREQUEN
NONMETHAN
CUMULATIV
IDE, AND
XI
o
• H
in Q
0)
i-f
CO
H
•atlon
tated concentl
B
a
*J
^i
0)
4J
«
cn
11
,H
ex
n
M-l
O
Percent
*
B
01 o
C .£
5 >
-H
€
C
& 41 (U
AJ O *H
JS -H -H
cn 2: TJ
O
ane
rbon* Ozon
W 0
0) O
0 tH
0)
So) d>
S °~x
ij Q
-H -H
01
N
Q>
& §
C N
C
4J /•>
2 ^B
c a
CJ ^
C
O
o
OOOCT'CT'OOOCJOCOOOOCOOOOOOC/OOOOCfOOOC'OOOOOOOO
o
»-4
rH
TH T-t
OOC3OOO(3C3OOOOOOOOOOC3OOOOOC3OOC3OOOOOOOOOOOO
c:
-rH
T-! rH
OOOCOOOOOOOOOOOCDOOOOOOOOOOOOOOOOOOO JOOOO
o
«-t
50"**4
oooooo-o^^rv^om^^fowcMfN,^^,^^
^-•^^
^^?ss,c:s^sss^sssss^s?sssss§sss5S|^s
•H
rH
cfl
I
C
O
,£>
M
CO
O
O
01
C
c3
1
O
1-39
-------
53
W
O
O
H
M
is
W CO
53 r-
O CTv
N rH
O
O CO
CO 0)
!S C!
O 3
I~H '"'"I
H
^d 1
H vD
53 CM
W
u a)
O 3
0 i-j
i-l 53
C*J O
t) H
O H
H
o w
co ta
O 53
H O
p1 §
CQ • W
U 5
p H
O" W
S 25
P^ O
53
go
P W
P H
U XI
O
H
• O
CO
OJ
3
H
o
o
•H
CO
tl
4J
a
*
0) §
3$
43 a)
i °
• I-S
> 11
3-
g
60 do *ti
s 23
r£ 4-1 O
CO -H -H
J
aJ
O
"&
C ,0
a) ij
a -tt
0 ° 5
•H
O
a
* QJ QJ
O Q i-l
0 5 °
fl)
(U 0) CD
C 00 *O
a s|
-H •> c
c S
PC
^
g
8-5
JJ y
g 3
U
p
3
!
iCSOCiCL COO OOOCi c_
iOOCJOC OOoOOOC
oooo-^oocaoo J i:
ooooooooooo -^
OOO^-3OOOOOCJC3^
OO'^>O'^OOOC3C3O ~
CD 0 0 C ^ C~ 0 C C- 0 (_• C
o o CD n. cr cr-ocor-oc:
£»OC3^ 00"000|0_>0
0^>OO 3 ^> 0 3 0| -3 '3 3
1oocac; ciocaocaooq;
ooocn ooooooo o
oofScofooioocjiooa
'OO^ft^OLTtCJOOCaiOO-'j
O O Oi to
^ o o a
"3 7» C3 O >"3
O O O O C2
c: Ci O O C
CD a o o o
o r^ O O O
*> m FJ TH 0
c -a} * ^r f\
rO C\jl r4
O t3 C9 C9 CJ
C3 O'l5)iCft''IC
'
. . . . J
O C O O C
in * t* co a
0 C 0
o c; o
o ^> J
000
^i> o a
O 0 G3
t_ ._. t:
o cr cr
0^
O O CD
0 0 CJ
o o a
000
0 0 O
*"
IWIOW
O'O'O
O O CD
O O O
O O O
coo
o o a
O 0 O
C; c.1 C:
000
O C3 CD
O o CD
o o a
o o ;;D
0 O O
C. CL CJ
o o o
o o a
ooc
cxj ro ^
K> K) ro
% .
*•
* IB 'vj
on IB
o o a
o o a
o o o
o o o
o o o
O O «3
000
o ca cs
o o o
a o a
oo o
o o a
*<3P>9 C3
o o a
c*«s> a
ooc
oo a
o»oa
ooc
•* *• fi
0
o
o
0
c.
0
0
c
•
c
•
o
o
o
0
o
TJ
-H
rH
Cl)
1
^ cO
4J
CO
T)
Pi
o
o 43
0 rt
o
o
T)
JS
a)
fl
r^
flO F7"i
u. 4^
i
C
o
53
*
1-40
-------
[TROGEN
123
W
&
O
M
O
fn on
O f^-
c-i
CO rH
13
o «
H iH
H en
^ K.
CONCENTI
ily 1-Jul}
P-I j
^3
£>
O !3
pd o
t-4
fn H
°£S
RIBUTIONS
ElBON BY Sr
ENCY DIST
E HYDROCA
FIVE FREQU
NONMETHAN
> 55 J3
C
» a) 0)
00 OOT3
l-i O -rl
wQ 4J 0
W -H TH
i
•M O
•H -H
2 -o
o
s
M
McH
ntration
CuB/m3l
u
c
o
1 ! !
o o t; c; c c^oooc^cc-cooc (-< cJQOoe:oot;GOOcooOe»'cric;
rH
^ ^ -H
*-(
! ' i
rH H ' '1
O O O C ' O CjOCDCJCJOOiOOOCJOCDOCjOc: OOOCDOC. OOiOCD OCiC.
C* O * i l
i vH -H
SSoooSS^&S^SSSS^^^wwH^^^^"' *" r"o0cr"^c~
i ! i ! ' i
i I !
' ' t
1 i
' ' 1 i ' i |
«4sSi^SS^S^35^S^S^3|g^SS£|g5S5
1
i
O|K» o o o
0 O 0 O 0
c O o OJ c:
1
[ i
C3 O O CM O
CD JO 0 Ol 0
c, 0 c. «=, c_
>000<3o
l^SIil
•H
rH
a
rd
"O
tti
o
o
cu
c
o
1-41
-------
ITROGEN
!3
«-i
W
2;
0
N s-~-
O CO
r~^
Pn o
RATIONS C
ust 31, 1
f CONCENT
ust 1-Aug
i-J 60
pi y
£> •
P PQ
M
M 2
ftj O
(-( CQ
UENCY DIS
HYDROCAR1
VII VE FREQ
TONMETHANE
ie
& •
c
* QJ QJ
CO 00 T3
Lewisbur
Nitro
e dioxi
N
O
*
Ci
0) C
, Ohio
en Nonmeth
e hydroca
Coshocton
Nitrog
ne dloxic
N
*
0) 0
•M U
cu o
6 M
C "O
O >i
n) E; x:
P-4
o>
c n
(0 OJ QJ
U 0
•H -rH
& 'V
o
s
* OJ
>> c
Q) O
o
s
o
*H
5 *S.
CJ ^— ••
c
CJ
c,cc.(_c<-c-c, i_ci-cc-t_cct cc^c-c c •- c o c 0 tj c. tr
o c- o c cc c.e.C-c.r c c c. o i_ r_ c c-C-Cc_c_c, dcocroo
CJ
xH
t. C ^3
H
r_ (_-,, C If C t s- *3 r, CV! ^ LT Lfi f\ If KJ M O *3 f- CT t^1 ^T r- O^ f^1 U"1 C~ f C\ C
t- T- rH
C
T-H
•H
O^^^^^-*X>K> onix.r<*a3,0'X)-oaOToHor^f^r^so oo^^r
•H r-l
S :
CD O O t-t
,H .r-l i-t -rH
^O^^CO,^^^.^.^ *«^»^^^r*] "j
: ! ' l j
ii1
' i 1 '
-^sssssssjjss^sps^ssps^sss
O a --5 -J 0 0 O O 0 D
o c d- c- c o c o c: r-
o e; c_ c; c_- a o o O o
* *> w r> CM c\| c* o cs| J
t-l ' r" C| O O «9 cr
i
Ilfeliliii
•H
r-H
cfl
cd
4.J
G
O
o
o
OJ
C5
fl
4J
0)
O
a
•K
1-42
-------
w
o
§
H ^v
H ro
(2j I—
ON
* r-i
W
£2 **
o o
N ro
O
t-t
PH 0)
CO QJ
S3 4J
O fX
M 0)
H CO
u ,0
is e
O 0)
CJ 4J
>H CD
i-4 CO
Pi ^
O Si
tfi O
M
Pn H
O O
§0
I—-^ f—l
U W
o
M
• X
"0 O
LO M
Q
(1)
J2
H
C
,m ,, ~^ ^- .- , _ ~ _. - -.-^ — — — cc'OCrc^OC^OC^OOOCrc^O O'^tGl*3 CD O fO O O C.
_- - __—__„______ _^_^^t.:tLc^c;oocLOc:ocDC!Oocrc3Ooc^oC3looO'C:
co o
@ M
JO O !?
Z ,
CD CO O.D'3~J'\jrOOv-t,-H-3 TJCDa^>T_J^^^>^5J^--3'JC3JO^>^>^>C3C3OC3C3OIOOC»rD
00 GO T3
H P"d OOLTlHr^CM-OrH ^O^D^DCDOCaJDTDCDOCDCDtJOOOOOf^O'OOCaC^CDOOOC^O
•H -H *-" T-l
3
I ' I
c-o^^c;^ f^o^rc n h'a;xri^irt"^t_vctAlc"rv-T-j''a'CDiW'i3't^THiooc>c^c^cooocr
- ' . . «| - . .[ *,«'•-
Si (uc^c^O'ii fvi^cr^ocv^cxicMr^- rt M o ^j-] TH cr. ^O1 ^r c\' cv TH i (c o c; *—• c" c' o o o] o
o J30>j-.jvj>x>r»««^<-«-io«i.\iTHHjcroc.'-<_c. ocJCiOf^cr-c^iOoc^oc^coocs^LjOicsoojo
O O*H OOO'>'OtAUsf\JrH (-CJ!Cl*OC5C_*C-C"lCOOC.-OC;IC5OC3Oc^C1OOO.c:;C;C3jC>OOJC?
U 4-1 O
O -H -H
2 *O
CO
p
I j
(U CD .j> fN* rv t_rt ^ -^r r*» ix> ^r~ o o n u\ ^ t^ j? :oi C3* ^o cvt o rH 01 UN fO oj 3 o oi cs o o ^ vj o cv o o •
6
Q) O
~" O C; O Cj CL. C-> CD O O O O O O O O O C Cf O O O O O CD1 O O JD1 CJ o Oi O O C?l CD CD C3 i
•••>•»•. «•*••*•. .•!*»•••«•*••'•».[***!.••
OOC: CJ c: C^OCDC3CTOCDC>CDC^CDOC>OC'C>OCDC?C3C3OCDCDO1OOC>c:)OCr)'
o) 6
s •
C '
o
n)
^ ' 'o ' J ooo, '
aj cucu ••••'•••••••««>*r..,«4 ^ . • »^ # • * • ,•]»•» • • • «^d
rt '£'3 oro^Q'-O(\j ooooo^ooCD^jo:TOOC>JDO^DJoC3C3|ooc>ooc>0CDOOOci|>
^ f, x CD o- «r T-) , , i
4-) O
I
JCJG'*CDOtril^fOTM^l'*a3oi<^CDfs*|*-H»OfO|>OOOiOOCDjOe> f u in vfi or aT (^ T-I r*) so o ^ Q. r^ vH CD *o ^ ^' rx' T-| o CD! ci t-. c~n o o cs1 ^ cr' c~ o oooooocTjC>OC>c''oo< -^j
O ' ! 1 " ^T<»H^THTHT-rt^»He>jroC\]««fK>K)(»|»Kl
CJ
•H
c
o
ca
o
o
)-J
a)
ct)
a)
B
o
•K
1-43
-------
100,— — .
e- 40
"x
\
September
July
• August
\x N
,\. \
40
80
120
160
Ozone Concentration
200
240
280
320
360
Figure 20. Frequency distribution of hourly ozone concentrations for July,
August, and September, 1973, at McHenry, Maryland.
100
g 80 —
40
20 —
July
/'-August
\
September
-A
\
\
40
80
120
160
200
240
280
320
360
Ozone Concentration (yg/m )
Figure 21. Frequency distributions of hourly ozone concentrations for July,
August, and September, 1973, at Kane Pennsylvania.
1-44
-------
100
60
AO
20
September
July
r^- Augus
40
80
120
160
200
240
280
320
160
Ozone Concentration (ug/m )
Figure 22. Frequency distributions of hourly ozone concentration for
July, August, and September, 1973, at Coshocton, Ohio
100 —
80 —
60
40
£ 20
\
X / L September
July
40
120
160
200
240
280
320
Ozone Concentration
Figure 23. Frequency distributions of hourly ozone concentration for
July, August, and September, 1973, at Lewisburg, West Virginia
1-45
-------
period August 4 to September 25, 1972 at McHenry, Maryland. Similarly
the hourly nitrogen dioxide concentrations given in Table 4 can be compared
3
with the 14 yg/m mean hourly concentration for August 4 to September 25,
1972 at McHenry, Maryland. The NAAQS for photochemical oxidants
3
(160 yg/m ) was exceeded during approximately 37, 30, 20, and 15 percent
of the hours at McHenry, Maryland; Kane, Pennsylvania; Coshocton, Ohio;
and Lewisburg, West Virginia. In contrast, the NAAQS for photochemical
oxidants was exceeded at McHenry, Maryland during approximately 11 percent
of the hours between August 4 and September 25, 1972. At McHenry, Maryland;
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg, West Virginia the
3
160 yg/m hourly ozone concentration was exceeded on 78, 65, 46, and
39 percent, respectively, of the days for which data are available during 1973.,
On the basis of the data presented in Table 4, it is unlikely that the
3
NAAQS for nitrogen dioxide (100 yg/m ) as an annual arithmetic mean would
be exceeded at any of the three locations.
5.1.2 Diurnal Patterns
Mean pollutant concentrations for each hour of the day are
shown in Table 6 and mean diurnal curves of ozone and nitrogen dioxide
concentrations are presented in Figures 24 and 25, respectively.
While the diurnal patterns of ozone concentration generally followed that
shown in Figure 24, marked departures from that from that pattern did
occur. Thus, Figure 26 presents examples of daily ozone concentration
3
peaking at night and selected episodes of persistent high (>160 yg/m )
ozone concentration are depicted in Figure 27. Figure 28 contrasts the
mean diurnal ozone concentration curves for 1972 and 1973 at McHenry,
Maryland. In addition to higher concentrations, the 1973 curve exhibits
a strong diurnal characteristic when contrasted with its 1972 counter-
part.
Mean diurnal ozone concentration curves for three urban locations
within the study area are presented for comparative purposes in Figure 29.
5.1.3 Correlations
Correlation coefficients for simultaneous hourly ozone
concentration versus hourly nitrogen dioxide concentration within and
between stations are presented in Table 7. Cross correlation coefficients
(Text continued on page 1-56)
1-46
-------
-l I -^
hJ P^ 1 T
Pi 0 | -
h^ C
o pi : .0
o ; 2
P-4 | U
~* j ™J Q
^3 ^^ U
o w
EH p£j >'
-" i
oo ffi
Q W |
5 5
P
c/3 H
j^; pq
&j ^^
^-t ^-—s
&
a) •:
^Q
H
"^
0) O
C .i"*
at c
5 "c
o >
c*:
•"*
3 C
.
o
c
c cu a)
W OH
o MX
C 4-1 O
,-C -H -H
;n Z "d
o
CJ
c
o
0
OJ O
C r£
cd j-t
-IJ U
£= £
O >-,
t5 4^
cfl
C
0) QJ QJ
£ ao -a
nj o >H
•M 0
•H H
OJ
O
N
O
T3
OJ
>, C
)-( O
C N
QJ O
1
rt 0
c g1 1
S-H |
W C 0)
Q) -H oJ
g 00 Bl
•H Q) n3
H J3 w|
-•LTi »-Lf\ • • IA --IA '-lO • • lf\ * • lf\ "-ll^
oo cro oo oo oo; oo oca oo
r*-ro irtro •«•« roro or*) fO^- f-o*- v*
t! CM ^ J (C ^ •/ .", ^ ,-, c" ^ c" ^ ^ c" ," ^ vi V^ -«" r^ ^
,-Ofo r-o o^o ^OM NO^ -j^^o -n-\j t^cM
oar^ o o r^ o-jr-v cscar^ caor*. QOP^ o r^ aof^
00 Or>^ *3 r» •"? -3 T^3 O O) OO ^O
--as ^-.co *.ca --a> -.cu ..ao ."^o • ^ eo
•^•^r -,-r-" c\v >^ if •« vr T» \r ' irr- ir>r-
OvJ-^ ^\J ^ ,TH f\Jr-t l\J*H i\J^t OJ^f IXJr-t
^ 'j n *> -y ,A o r^. >A r^ "^v LA -«• \j J^ o ^r Ln IA ;^v 'A CM > -^r
-) vj v> r- ^ y -o .^o -u o ^ ^ * >c TV *"" .\J ^ ^
c if, o if c- if , or u r^- IT r-v tr \L IT- ^ IT-
T-4
CCJ OO OO OO CI?O OO CTO OO
r>-or- oooo TXOO> roxoco roroo> THCMCO KXCT-OO o>oao
-xi r, -or, cvj^ o^ r,o ^^ •Njr- * .t>
•«• r^ c- vci_c ircr, c^ c'^rcr oc^o trojc cvrotr *ro-r>
c^ r ** ^V °* e> ^ °" ^ K* ^ o" A" ^ o^ ^ * ir'cv*0* ^ A "
THT T-*^T o^ o^- r^^r o^" •>•*- OT
t-t -H »-< T-l
fOOD-O -A^-0 OtntJ 0^-0 f^f^O J^O^X) t\J^)CO ^OiT.OJ
*? ?? P"? ?S ^^ 2* SI? ^5
; ^ ^ ^ ^ > >
-Li 'OJ ILI ,U LU IJ 'U LU
{- (f O (/i ft/' C1 (A: ft/ C if O O" f C/>
.^'iJ ^JJ -riij rui -2 'ii ^LIJ S:LU -^i-u
f c if *t c~ ir •<("•>(/ -a f (/
c- > tr o GJ ? t/j o c~ r> (AHJ CL > c' o c- 5 c/, L_' L > if o e r c/ i_ o y tr t_~
O
Tl
n\
'S
^3
iH
U
•H
QJ
rrt
en
C
0
o
CU
0)
n3
cu
4J
• *v
cd
•U
*Tl
•H
frt
^
r}
•H
CO
cu
4J
0)
o
•H
C
•H
C
o
'T"
nj
o
o
T3
'^
CU
c
Jri
4J
§
C
0
CD
(-4 CU
cu cn
T3 O
a P.
3 !-i
>> CL,
C 0
CU -rt
4J
o aj
O M
0
0) M-l
1-47
-------
w
p
M
X
O
M
P
)^
pil]
o
o
H
h- 1
- r~-
W 01
N "
0 0
m
P1-)
O J-i
cu
CO J3
o cu
M 4-1
H a
<: cu
P5 CO
H 1
W CM
O
ss cu
0 C
0 3
^H s«— x"
,J
Pd !H
^3 <1
C^ (—5
SB
W
pr 1 p~j
O H
en Pn
2 o
o
H K
H f3
< O
M rc
W K
P U
P W
P 0
H &
CO O
P PH
o i
en &
t5 P
\^ jj
1
CU
,— 1
cd
H
[1
Ij
1
j
i
e*P
£
60
^
C!
•H
4-)
CO
r!
oncei
u
•K
OJ
flj
4J
a>
o
35 X
<0
S C
dJ
oo -o
00 O 'i-
3 -U
M 2: t
0)
1
1
o
•K
a> i
cO 1
4J 1
a) o
g i
0 11
•H
^
o
c
G aJ o
o co -a
•U O -H
u u x
O 4-» O
X -rl ••-
o
u
c
c
0
•K
a g
gf
4-1 C.
1 £
B T3
O >i
.
cO
Ci
cu a> cj
C OD T3
CO O T-
4-1 C
o c> -O
OO O-l "3 O OO O-3 C3O OO OO
L" ir> -=! -^ r IT cv L' c\i r^ T- r^ r-t w * CN. ^r -xj JD ,o rM o IN, AJ rxi ^j (M ,y ^r o -o <•
.v ^ ^ ^ r-"^" ^ ^ ^ ^
*o^ p^f o^j1 T-iLr cvio ^»"ir> vir iTiif*
-1 <-4 t-* M i~»
oc,a o o a- oor^ o c-- rv oocc ooa oor-. ooo
OO CO OO OO OO OO OO OO
*•*•!•- o-a) »»o
ifi"o ^c\i rvir^. 'o-o o^o tj\o r^io cO(-«*
..en ..-n .^S «.S ^ °Io* .-o ^ .ox ^"15
*"* *^
^T 'O'O tlfN '^|jT> fOfN. O O CO C\J OfN
rf' ^" ^^ rt10 *" ^i11" T^10 ^^
llj . II til ,U JJ LU IJ QJ
c (/i C- tr/ f1 t/> c1 c/) c^ t/"1 a (/i C' if tzi ; OJ ^ UJ .Z IU ^ >JJ 2 UJ ^ UJ
-ff c: c/ u
o a*Htvj(n*"iA>o
r *^ *^ T-* r r' *-> t-1
J^j
O
M-l
T)
CU
3
rH
O
c
•H
cu
n)
M
4-1
a
o
u
cu
CO
ni
a
cu
^_j
td
cd
T3
•H
rH
cd
£j
•H
CD
CU
4J
cd
O
•H
C
•H
CJ
0
"£
cd
o
o
TJ1
cu
C
42
cu
g
Q
w
^4 CU
a) ca
T3 O
3 !-i
!>% CX
4-1 C
C O
CU -H
1 t
O rd
0 l-i
O
CU M-l
!H -H
*
1-48
-------
o
w
n
M
O
M
O
w
o
o
Pd
H
M
IS1 -
O 0
O )-i
QJ
CO O
SS S
f*~) QJ
H 4-1
H PJ
-<
P <:
o o
S
w
EH pd
0 H
05 fn
21 O
0
M PS
H P3
< 0
M Pd
W PC
Q U
-t
S
• -
to
OJ QJ
oo -a
t-i P X
3 4J 0
QJ
O
0
c
QJ O
C XI
ra t-i
OJ 0
B ^
O >
o y; X
-C
o
C
C QJ QJ
o on T:
4-J O *•"
O 4-1 C
JC -H -H
o
o
a
0
N
•K
c
QJ O
C Xi
flj ^
4-J U
QJ O
S t-i
c -a
O tl
Cd
D-
C
QJ at a
to o •>•
£ n >
•H -r
C
o
N
0
S
QJ
C "
0) O
?
31
rt o
/-\ 00
H C
n -H
UJ C C
at -H c
S oo B
-H (U 03
H J3 m
0 0 ^0 rc,r- tOO cc.^> r,c-.0 0 C %0 C,0%C
OO OC1' OO OO C"'O C^(T» OCJ
cotNjo r^T^""* o*^!^ oofo firycxi oi^cv ^c>cv
i/\(o -o < -OPO >oro CD * oi*> f>ir
THi-< T-lr-l t-H »-( »-!*-* t--lT-1 ^-t »-t i-l *H
cjo^ otar** oof*- oor*» CJCDPN. oc3*o caor*^
C3 O 15 O O O •"> ,~J O O O T> O O
<-r r-"00 r'fx*00 w* c^ * f-%"°° r^" ^ ** a* ^ *" c^ r^ "
>r .(Mrt ^ > o 3 p- •«• in in -o m o vr ^
it-, n ,n f> ^ -^ * «r u o o co ^ v> tou (MIT- «HI/N
, * rH -4 •-! iH *-* T-I
Ci c* o C' CT T-( r. CD TH c o e-j o cr <-• o c. o c^ tr> cn
CD O C. C3 <~1 O O CJ C~> CJ O O O CT.
*oojr> ^-ino T-t-oo T-*O-O inroao in -r oo *«-<«-co
£ « roc> ro-o <%jO
^tJ^ ir>cjc^ r-oc in TO u, r-r^ room tooir
^H ^t »-I i-t *-t -^* i-»
Or- r«*f^ * rs. oaj cor^ r^-in oo
t^- u~ fv. ir r*- iTi ^ri ir y^ ii" LTi ti u"i in
tH r-« rt '~t T-I »-l T-4
,JLJ 'U 'H 'U ,LJ UJ UJ
o t/i t cn i vj r~- n c/i
„- u ^ u * UJ ' UJ -j ixi ? UJ ^ m
*I((/i ij C f/, •* l~\ W< •« £ (/i ft C V -N* ^
TH IT
vH
CD CD C
0 0
c» flO «>
«H
CC1 .-* 0^
CM m
»--<
a.,,0
M aO
^
!U
O t/)
.2 LU
^( C'1 */l
(V'
13
a)
•x)
r-l
u
C
•H
O
o
Q)
cn
n)
a)
^
4-1
n)
4J
a)
•-O
n)
cn
QJ
u
•H
d
•H
a
o
cd
O
O
V-i
0)
0
n)
I
C
o
£3 •
CO
!-i 0)
0) W
13 O
c a,
3 t-i
3
>, Cu
)-l
4-J C
C O
Q) -H
4-t
o n)
• c=:
o C
o
0) MH
£1 C
H -H
•K
1-49
-------
180
L60
140
120
e
00
o
•H
JJ
(fl
O)
g 100
80
60
40
X—X,
\
\,
\
X— X^~/
G>
\,
\
\
X Mcllenry, Md.
O Kane, Pa.
A Coshocton, Ohio
• Lewisburg, W. Va.
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EDT)
Figure 24. Mean diurnal ozone concentrations at McHenry,
Maryland; Kane, Pennsylvania; Coshocton, Ohio;
and Lewisburg, West Virginia, from June 26 to
September 30, 1973.
1-50
-------
30
20
00
c
o
o
c
o
CJ
1 0
LU
-A-
O Kane, Pa.
A Coshocton, Ohio
* Lewisburg, W. Va.
__^
"^•— •
A°VQ
Minimum Detectable Limit
0
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EDT)
Figure 25. Mean diurnal nitrogen dioxide concentrations at
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia from June 26 to September 30, 1973.
1-51
-------
a
M
C
O
•H
0)
o
c
o
o
0)
e
o
N
o
210
200
190
180
170
160
150
140
130 -
120 -
110 -
100 —
8/23
Coshocton, Ohio
8/24
McHenry, Md.
7/26
100
12 14 16 18 20 22 24 2
Time of Day (EOT)
8 10 12 14
Figure 26. Examples of night-time ozone concentration maxima at
McHenry, Maryland and Coshocton, Ohio during summer, 1973.
1-52
-------
g
M
O
•H
C
0)
c 220
o
u
160 yg/ra )
at McHenry, Maryland; Kane, Pennsylvania; and Coshocton, Ohio
during August, 1973.
-------
180
160
E
60
c
o
•H
4J
C
OJ
o
o
140
120
100
80
\
\
June 26-September 30, 1973
\
\
X
August 4-September 25, 1972
"\
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EDT)
Figure 28.- Mean diurnal ozone concentration at Garrett County
Maryland Airport for June 26 to September 30, 1973,
and August 4 to September 25, 1972.
1-54
-------
60
3.
c
o
•H
140
120
100
80
c
-------
Table 7. CORRELATION COEFFICIENTS FOR SIMULTANEOUS HOURLY OZONE
CONCENTRATION VERSUS HOURLY NITROGEN DIOXIDE CONCENTRATION
WITHIN AND BETWEEN STATIONS
Nitrogen dioxide
Ozone
McHenry
Kane
Coshocton
Lewisburg
Kane
0.097
(1540)a
-0.166
(1818)
0.044
(1764)
-0.158
(1582)
Coshocton
0.020
(1586)
-0.192
(1922)
-0.102
(1775)
-0.050
(1627)
Lewisburg
-0.133
(1491)
-0.221
(1579)
-0.047
(1566)
-0.084
(1655)
rt
Number of pairs
for simultaneous and lagged hourly ozone concentrations between station
pairs are shown in Table 8.* The selection of one-, two-, and three-
hour lags is based on an inspection of the mean diurnal ozone concentration
curves shown in Figure 24.
Correlation coefficients and differences in coefficients significant
at the .05 level for various sample sizes are shown in Table 9. Since
a serial correlation approaching .9 was determined previously for
hourly ozone concentration data lagged by one hour at McHenry, Maryland,—
Table 9 contains values for both serial correlation and no serial
correlation assumptions. Since the assumption of serial correlation is
consistent with the observed changes in pollutant concentration from
hour to hour and provides a more conservative test of significance, the
correlation coefficients reported in Tables 7 and 8 were compared with
those in Table 9 that assumed serial correlation.
Thus, in Table 7, none of the coefficients is significant at the
.05 level with the exception of the Kane ozone-Lewisburg nitrogen dioxide
comparison which is marginally significant. Since many of the nitrogen
*Correlation coefficients were computed using the BMDX 84 "Asymetrical
Correlation with Missing Data" program.
1-56
-------
P C/2
<
00
^i
3
n
en
•H
s
CD
a
o
•u
0
O
cn
o
o
QJ
qj
|
j
o 3
Pi ! >
U 0
'
• o
00 W
o
, 1 ^.jl '
o3 '
H :
|
>i
c
OJ
PC
o
CO
rM
,H
O
CO
™
*-)
3 O
o
00
CO
CO
L~M
rH
o
CO
CN
^
0
1
C
O
Id -ri
d) w
X ccj
II -H 4J
fc U)
^- co r- ) O cri ON
CN in I-- -3" ^O CN
•wD ^~ "-^ r~*- co in
o ^ o ^ o ^-^
-^s s-^ s~^.
ON f-^. CO CO m CN
co in
• iH • iH • i-H
O ^ O ^-" O ^^
CN t-- ,H
-------
Table 9. CORRELATION COEFFICIENTS AND DIFFERENCES
SIGNIFICANT AT THE .05 LEVEL
Serial correlation (.9) No serial correlation
Number of •—•
r ri~ro lrl lri~rol
pairs II 1121 II |12|
1450
1475
1500
1525
1550
1575
1600
1650
1700
1750
1800
1900
.230
.228
.226
.224
.223
.221
.219
.216
.213
.210
.207
.201
.326
.323
.320
.317
.315
.312
.310
.305
.301
.296
.292
.284
.051
.051
.051
.050
.050
.049
.049
.048
.048
.047
.046
.045
.073
.072
.072
.071
.070
.070
.069
.068
.067
.066
.065
.064
dioxide concentrations were below the minimum detectable concentration,
caution should be exercised in assigning importance to any of the
coefficients reported in Table 7. In contrast, all of the ozone-
ozone comparisons (Table 8) are significant at the .05 level. Considering
the station spacing, the correlation coefficients are surprisingly large.
However, none of the zero lag correlation coefficients shown in Table 8
is significantly different from any other.
5.2 Supplementary Data (October 1-November 2, 1973)
In addition to the contractually required measurements at Kane,
Pennsylvania, additional measurements were made at McHenry, Maryland;
Coshocton, Ohio; and Lewisburg, West Virginia. Those measurements are
also reported herein.
1-58
-------
5.2.1 Summary Statistics
Means, standard deviations and case counts of hourly ozone
concentrations, by station are shown in Table 10; the corresponding
statistics for nitrogen dioxide are given in Table 11. At Kane,
Pennsylvania, 163 cases produced a mean hourly nonmethane hydrocarbon
3
concentration of 126.4 yg/m with a standard deviation of 34.4. These
data were obtained following modification of the hydrocarbon analyzer
to remove the moisture interference and are considered valid. In
addition several grab samples were collected in Tedlar bags by RTI
personnel and submitted to Chemistry and Physics Laboratory, National
Environmental Research Center, Research Triangle Park, North Carolina,
for hydrocarbon analysis by gas chromatography. The analytical
results as well as the comments of the analyst may be found in Appendix C.
Cumulative frequency distributions for hourly pollutant concentrations
are presented in Table 12 and plotted in Figure 30. Maximum hourly
ozone concentrations (concentrations exceeded during approximately one
3 3
percent of hours) were 240 yg/m at Kane, Pennsylvania; 170 yg/m at
3
Coshocton, Ohio; and 155 yg/m at Lewisburg, West Virginia.
During this period, mean hourly ozone concentrations are lower at
all stations than during the June 26 to September 30 period. Mean
hourly nitrogen dioxide concentrations are similar during both periods.
The NAAQS for photochemical oxidants was exceeded during approximately
19, 3, and <1 percent of the hours at Kane, Pennsylvania; Coshocton,
Ohio; and Lewisburg, West Virginia. It was not exceeded at McHenry,
Maryland; however, it is noted that only 59 hours of data were obtained
3
at that station. The NAAQS for hydrocarbons [160 yg/m as a three-hour
concentration (6-9 a.m.) not be exceeded more than once a year] was
exceeded twice at Kane, Pennsylvania.
5.2.2 Diurnal Patterns
Mean pollutant concentrations for each hour of the day are
shown in Table 13. Mean diurnal ozone, nitrogen dioxide, and nonmethane
hydrocarbon concentration curves for Kane, Pennsylvania for October 1-31,
1973 are shown in Figure 31. Due to the small number of cases for each
hour (maximum of 7), little significance should be attached to the
nonmethane diurnal concentration curve. In Figure 32 it can be seen that
(Text continued on page 1-67)
-------
Table 10. STATISTICAL SUMMARY OF HOURLY OZONE CONCENTRATION
MEASUREMENTS BY STATION
October 1-November 2, 1973
Station
McHenry, Maryland
Kane, Pennsylvania
Coshocton, Ohio
Lewisburg, West Virginia
Mean hourly
concentration
(yg/m3)
87.0
111.3
73.9
81.8
Standard
deviation
(yg/m )
19.4
52.9
37.1
34.3
Case count
59
729
558
739
Table 11. STATISTICAL SUMMARY OF HOURLY NITROGEN DIOXIDE
CONCENTRATION MEASUREMENTS BY STATION
October 1-November 2, 1973
Station
Kane, Pennsylvania
Coshocton, Ohio
Lewisburg, West Virginia
Mean hourly
concentration
(yg/m )
13.2
23.6
15.8
Standard
deviation
(yg/m )
13.2
12.9
15.1
Case count
731
559
782
1-60
-------
I
!
§
H
g ^
W H
O^ :
ONS OF 0^
ovember <
NCENTRATI
tober 1-N
o u
o o
>-<
Pi O
p H
0 ^
CO pq
M O
pq <
M O
^ 5
co n
M >-<
C K
>H W
2 FREQUENC
NONMETHAN
:UMULATIV1
SIDE, AND
0
H
• 1=)
(TNI
rH
-•
C T3
O >
0) 0)
O -H
^ X
AJ O
•H -i-l
CJ
O
hane
arbonA Oz
0) O
E *-*
C -r)
o >,
!Z -C
CU QJ
OCT3
4J O
•H -r-
tz, -o
QJ
C
O
N
O
C
0) O
-U O
OJ 0
pi M
o >,
c
Nitro
zone dioxi
o
a
tration
/m3) Ozc
c «
0) 3
o >-
o
u
c. n c» o o o C' C' o CL o o o c,' cs o cr? c~> c o cz> ci cr <- c; o o cj CD
C, O Ifl CV TH t-1
T-*
crCDr^r-i«-)irvc~icr f^^-^ririr^^-^ hotv^-^oc., ocrotncrcrcrc:
„
T-1 T-i
*H t-4
M -H
s§sggg§ssss^?;^sKss;^ai^rH^^ oooo
C; O* if" (\i TH
T-J
•*H -rH
c- c: £. C er r o a I-- IT K C\' T-;
•^1 rHl t-l vH -H rHj
c~ c c> c~ e c- c.) r_ cj c- c o tr tr c^ c- cr, c- c • cr o C? o r? o cr • o c:
cd
i
p!
O
^3
S-i
B)
O
O
M
•n
0)
fl
OJ
§
O
1-61
-------
o
•H
4=
CO
PM
0)
C
C 60
O M
4J 3
U 43
O CO
JC -H
to !s
o a>
U i-4
o <3 •
ao
o
•H
J-l
cO
n
4-J
OJ
O
fi
O
o
N
O
csi
M
0)
•i
OJ
o
•H
M M
Q) (-4
,0 -rl
O >
4-1
U 4J
o tn
•rl 60
J-l M
n) 3
n -Q
4J tn
C -H
Q) £
O 0)
O T3
d
a) cd
C
o •«
N o
O -H
£1
>, o
3 (3
O O
,G -U
O
4-( O
O 43
CO
C O
O O
•rl
U ••>
3 CO
42 -rl
•H fi
M cd
4-1 >
CO rH
•H >>
-d to
C
>> CJ
a a)
a PM
cu
3 «
a4 a)
a) a
uoj 3 BJ: 311301103
A-[jnoH jo
60
•H
1-62
-------
!
w
n
M
X
o --,
M I ,
0 ; -i
• •
S3
W I'M
0 i- <
O i! '
OS 1 • -,
H I ' i
r— 1 1
53 3 •
x-x ',
"co ,, ;
W r- •
*-r —. '! >
<< CT\ |i
0 --)
Nl !
o - |i -•)
CNJ j
P4 f
o <* • i
a)
en 43 i1
S3 e ;;
O QJ i|
M > !!
H O
5 £ 1'
PS 1 !]
H -H
^*
w s-j
U OJ
53 ,£>
O O
u 4J ~
0 >."B
>-< O I! -C
nJ ^ M
PH •— '
53 53 n
CO J g
K 1-1 , -H
Hi! ^»
r ^ ll <3
PM H 1'
p fa i
< w
H 53
co '
.Q
cd 1
L— t '
i|
c
0) O
C XI
03 l-i
X! fl
u O
0) O
(a >-*
C T3
O >-
z x:
cfl
>•
3 C
Q) QJ
OO T3
OC O -H
^ »-< X
3 -U O
XI -H -i-t
M Z T3
•H
&
QJ
(-J
0)
C
0
N
O
-K
C
QJ O
C Xl
ctf Vj
X! cQ
*-i U
QJ O
0 tJ
O t^
o z x:
•H
x:
o
c
C Q) QJ
O 00 T)
•U O *H
O V-< X
O -P O
x: -H -H
en Z T3
0
G
C
0
N
O
c
Q) O
C Xi
nj n
X! 0)
U O
QJ O
C T3
O t^
Z X
CO
P-1
c
Q) QJ QJ
C OO T3
rt 0 -H
W P X
•P O
2S TD
OJ
Cl
O
N
O
"3
QJ
>i C
^ o
C N
W
u
£
4-J U-l
n3 o
/-v 00
H P!
O -H
W C QJ
•— ' C I-H
0) -H D
6 60 £
•H d) CT
H JD K
(- c, c cr c? c:> <_< C-- c, 0 C-. c o <^ c^ c^ c-1 ca 0 t. 0 0 d
s« .. -• •• •• •• •• ""
or- o cr. o-o o-o oc »o o-o »t-
(>a.o=>o<:'''>?
p.cyorc-00^.-
0,»0»»!><>°>
r^cioc-r-o^C'
'
^-r,io -omro r^oio r-^'X>« innio » r» w -^<-tio -oi-ro
»^ »«Ki »-ro ..f^ . • ff ••re ••r'3 ««r;
^, !r, ' »in « irv cvnr, ' cv o ' CMT> f -o ^ in
;;,-, ^1^ -^v^ TH-r-l ^t-l rH^l ^1^ -H-H
J5r^tV Ki-0(\J (NJf>c\l fOlfNM CVIK'(\' UNC-CVI (\'CC(V O-K>rv
..ro ..ro ..~5 ..ro --rO . • IO ..to ,*,"-f
«- c; Or" 1 J OJ OL U- U U. TH >-l 0. < C-
JS.CM «rj O-H «^H in CM ^'3O C3C5O OOC3 OOO OOO OOO
00 00 >0 OO OO t»C5 O>0 0>0
o,aO'O>o<'>Oa
»o-»o.»»o><*
o-o-t^o-o-o-oo
oo-o-oo*""^
(rr^^r i-'f)^ ^fvT f. cr "j tp^T- \c, ov T-IO^T c^^^r
. . ry ..(M • . CM ••!» ..(V .-CM ..tV) ..CM
ir. TH r- CM r~- f> a, » i^. in r- -f> o- o- tr o-
fg^ ^J,H CVrt CMrH CMrH CMrH CMT< CMrH
3r^-» rvJ.J-1- JN-O « M ^o -M ra a> ;> -r J ^ -H
O o^-
Jjro r-jpo n v r^in >Mfo CMCVJ CMCM toro
^JTHT-tTHTHT-tTHT-i
*OCOT-* -DCVJ^H OCVJ-rH THlOtH -«-r^TH «HrHTH 'OTOr-* COrOt-*
^-^' Air\ ^ro srcvi tj-cvi iniA tn^r iAro
,H^ THrH THrf THTH THT^ rft-l TH »-) -H-rH
,.ro ..10 •«« ."10 ..10 . • 10 ..to ..K)
(, r_ i- u- r- o. -r- -H i TH o c. o; c o- r-io -o if> ro o TH ro incviro
• ,1 fo'r-" oro lf\'r> 'MCM iTll" « -0 OCNJ
o T- c T cr c c rvi o cv
^ CVItOTlO'O'^00
,- C f (- CT C c
0
M-H
'O
g
•H
en
eu
4J
cd
CJ
•H
T)
•H
C
o
^Q
S-i
03
O
o
\-\
Tj
>1
,£1
0)
CJ
r
^3
4-J
eu
o
K •
eo
S-i 1)
a) ec
T3 O
C OH
3 M
3
>> ft
j_)
•u C
C 0
0) -rl
4J
O cd
O S-i
o
el) M-*
^.s
•K
1-63
-------
z
w
O I
o
Pi
H
^
w .-->
Z ro
O fv
C^O C7>
O rH
CL, r.
O cvl
CO M
Z 0)
O 43
M e
H OJ
PS O
H Z
Z 1
W rH
U
Z M
O CO
CJ ,Q
0
>H -P
hJ 0
Pi 0
£3 v^
0
K Z
0
Pn H
0 H
<
CO H
Z to
0
i — i >-<
H M
<3
M Z
> 0
W M
Q Pi
~j
<4
§ O
< pi
o o
Z >•<
*-f* *-r*
<1< I-U
H
CO W
Q <
Z HI
3 c
QJ QJ
oo -a
00 O -H
U r-l X
3 -u O
Xi -H -H
co :z; T3
-H
3
QJ
,_J
0)
c
0
N
O
*
C
QJ O
C X
« rH
x: re
4-1 U
QJ O
C "O
0 r^
o z &
•H
rC '
o
a
c: a) a)
O 00 TD
iJ O -H
O r-l X
XI "H 'H
en z -o
3
QJ
0
N
O
c
QJ O
G Xi
re r-l
x: cj
4-1 U
QJ O
e ^
C TJ
0 X
rS rC
re
a-
•• c
CD QJ 0)
CJ OO T3
re o -H
& r-l X
•u O
•H *H
2 T3
QJ
c
O
N
O
"d
S
rN QJ
x c
r-l O
C ^
QJ O
tf
u
£
4-J 14H
re o
^. oo
H C
O -H
w c: QJ
v- C t-t
(U «H &
B 00 £
H rO a:
00
^..^ .. -^ J. . .. . i ' «. «*
C-o oo oc- o-c^ oc-' cro or- o>o
Of>oi>oooo
<; 0 C' O C T- O O
00*000 ooo
( C .^ O O L. C O
ooro rvrHro orDro ryr->M orvro caojro fooro r-irooj
• « PT-, -^^ • • rc • • KI ••ro --to *»ro ••to
rvin -o-o ^orv m'T rooj "ooj rooj roro
rH r-t r-l rH rH rH rH rH rH r-J rH rH rH rH rH rH
r- f o. • PW CT ir,<~coK>c:
..^ ..-o ..-o ..-o ..-o ..to «-ro «-ro
^c r- r >_NJ c*c ^ri_ '4"r C-CT fnc-
•*CM in :M rvoj LAOJ JCM r-noj rHOJ c\ir-o
rH -i-H r-l rH
OOCD C3OO C3 O CD C3C3O OC*».O C3C3C3 CD«3C3 r3iC3O
OO OO O»~3 O"3 O-^i OO OO OO
CTO CC O-OO-O
oo^oo oo o*
OOOOOSOO O
T- f> •> -> O
c c" v c* tr *"• n \f rr, o rv --D c^j ^r .\j vj rv ^ ro rv --\ .-o T-i ^r ro j fo o rv. n ^
• • (\- • t O -J M O if! 'MA -O rH — 1 ^ \-H O fv ,O
rrH ""-rH Mr4 *£rH r-~Oj O-O< C' c r-o^ m -H -c r-tirr^ cc-rHrv ocorv
(CO- rv^-( *»rv inco rv-* ooo rHin rv-*-
fO1^) U>rO fJf\J OJOJ OCM -rHfO OJ1*" rHfO
rHr-l r-'rHrHrH-H -H
^•corH orvra O-^T'--» oooo r-*^rH rvtnrH incyo in*o
* • fo • ^r) ».f* ••r', -'ro « • K> • • ro ••ro
(v.\| O-4- sT'O IOPO ~Z>rH OrH -O«H; (-vO
•HrH rHrH rHrH -rH-H r-trH rH tH rH
rjrOrH vcrvcj ^ir-tr & tr o ^10^ o,po-rH crnrOfO x>inoj ^^TOJ rvrHoo roo-o rvooj >oa>CM
^•rH OrH *TO ^0 rHO O«O JO ^T OJin
r- r>r-, fr> CTT c*5 of c o
;=:-;:?•>-->:>
, U _iJ J U OJ J t U UJ
i_O ft/ it/. crTt/u rvr"!^!: crc/1^: CTTTCOU
oo •HojroTtn'-o
( T-f T- »-• i—r r- T-1 r-'
n
o
M-<
no
(U
t3
3
r— j
O
a
•H
fll
vl/
M
rrt
vu
CO
4-1
0
3
O
O
Q)
CO
cd
O
CD
^C
4J
Cfl
4J
cd
T3
^
•H
, |
cfl
pi
•rl
CO
Q)
4J
Cfl
Cj
•H
6
\-4
•H
(-
O
,Q
VJ
cfl
o
o
)_4
TJ
>i
rC
§
Cfl
,d
4J
(1)
e
pi
o
88 Co'
H 0)
QJ CO
T3 O
C CX
3 H
3
>*. CX
Lj
4J C
PI 0
Q) -H
4J
O Cfl
• B
O !-<
O
a) M-I
& G
H -H
1-64
-------
J3
W
O
O
rv>
t-^i
H
M
a
A
W /—N
S CO
0 r^
C**3 ON
O r-l
ft, „
O
PS O
H S
£5 1
W ^H
u
5*" LJ
^-^ M
o cu
o ,0
o
>-l 4-1
Pd O
1 — 1 x_^
H- ' ^^
o
PC S3
, o
PM 1— 1
°3
oo H
& W
O
M >H
H M
•*rr*
H Z
i i ^
> 0
W PQ
d r*j
3 O
_<* -.j
- W
4-1 O
C M
0 X
0 0
'-•' I-I
00°
i— f
Cfl
H
CO
5
00
3.
c
o
•H
CO
4-J
C
0)
o
c
o
u
•K
CU
c
CO
X
4-1
B
0
2 X
£
St C
OJ
oo T:
be o •
u i_<
3 4-1
J3 -H •
W 3 TJ
•H
qj
hj
c
*
0)
C J3
CO M
J2 cO
4J U
OJ
£
C "
O >•
O 55 J3
•H
_c
O
C
c u cu
O 00 -O
JJ 0 -H
U MX
O 4-1 O
JS -H -H
« IS ID
0
O
c
o
N
O
c
cu o
C xi
« M
42 cfl
4-J C
<
•u o
•H -H
a -a
0)
C
o
N
O
T3
H
, c
% °
C N
M °
u
S
(0 O
^^ 60
H C
Q -rt
w a oj
^^ C ^H
(U -H P.
B bO 6
•H QJ cfl
H XI W
oo c; ooo oo o ooe ooc; c_c.c; ooo o o o
o>c o^er o-o o-o o^c o>c ONCO o-c,^
ooo^ o»o^o»o^o
o t> t> o- o^ o o- ON
OOO* ON o ON ON O
OGCC'COCO
'**'
NO^CM ino:M OKI<>J OTH^I rnco'M in^oj nmcM v-fo CM
* • FJ . »r^ *«ro ..jo ••K • *K> • • ro • -ro
roro ^o^- oro ONO o \t\ rHin o*in r-.m
^ ^H rHrH CMrH CV ^H CV'TH CV1 rH -i-trH rHrH
^^^ v^o r^-^r*- ro ic. cr f\j ^ r^ *c c~~ TH TH ON TH r--. m TH ,
• • *.M »»ro "»CM ««ro B.-O -"t-o **ro • • ro
U'ro ^rc^ TH«3 K>IT sou- <• ir rocv NC^
TH" *O rHfM rHCM OCM ONfM >X>CM OOCM fvCM
T-* TH TH TH
! '
OC3CS OOO OOO OOO OOO OOO OOO OOO
•* •• •• •• •• •• * • . •
OO O*0 O*O OO O*0 O>O O«O ONO
cr o oo'ONO'O'ON
O^ON OS Ch ON ON OO^
O-O t> ON ON O1* ON O^
^^ON ^OONO^^
vi-CNro ^-ITK. MT-rHfO THtxi-j OrHfO K>foro CMrj^ f^r^ro
*» M "N C»» CM 0 "O (D rO rH .M T\ f^ t^ *r
THI^ rnr^ cro crro ONM o^ro CCCM r-vcvt
rH -H -H r-t
ONTHr-. iTr*-r^ ojONr^ •^•ONr>. inTHf^ rocor1*- o-CMt^ ^-^rr>.
• * *• •• •* •• •• •» * *
r-ro o^- rooN ir^in to-o >o o ^r^ ONTH
rH rO CMfO CMCM CMfO CMrO CMK> CM rO ^HfO
rH TH rH tH rH rH rH T-l
mrocD >or^o r^r^o NO-VO rxr>.o aorDto »omo ONCMO
r^^M TJ^H 'MrH ^-CNJ CMCM CMro' -«-ro COT
^H ^n ,H,_, ^^ ^^ ^^ ^^ ^^
'rr^o incccL K>THC, o^ en o vino TH er, ON inina- inroo
• •ro »«to »»io B»ro »«ro • •CM *«CM ««ro
^ V CX K T t.- If IfN (y, y^ ^ K „, c f^ Q,
^^ ^TNO ro in ^\iin rxjin rHin c?Xi o<cM ^NOCM r-rocM rHincM ror-cM
•• •- •• •- . . •• ». . .
foco inro or^ THK> aoro CMO^ ^-CM r^co
cc T-. r c\. r r\ r^ K t^ f, c to c c- >- v, c.1 i/o c. er x trt o r- y tfi o cr :• tfi o
^ OOCKOrHCMrO-Nr
J-l
o
M-l
'O
OJ
13
3
t i
o
C
•H
0)
VH
cfl
cfl
4J
C
o
O
cu
CD
cfl
CJ
a)
jn
4J
• *\
cO
1 1
cfl
r^j
T3
•H
rH
cfl
>
CJ
•H
CO
0)
IJ
cd
a
-H
1
•H
r;
*-i
o
fl
^_l
ctf
O
O
J_)
rQ
t>v
f^^
r;
r1— '
CU
c
Cfl
x:
4->
CU
^
O
3 •
CD
!-i CU
CU CO
T3 O
r~* r^
t-4 [-^
dl,
r-*
^
>-, &
^
4J C
c o
Q) -H
4_J
O nj
O i-i
O
CU M-H
rd a
H -H
•K
1-65
-------
160
140
120
I— O
O
100
c
o
•H
4-J
a}
J-i
4-J
C
O)
O
C
O
u
80
60
40
20
'Ox
G>
*
O Nonmethane hydrocarbon
• Ozone
CD Nitrogen dioxide
"Minimum Detectable Limit'
0
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EOT)
Figure 31 . Mean diurnal concentrations of ozone and nitrogen
dioxide from October 1-31 and nonmethane hydrocarbon
from October 16-31, 1973 at Kane, Pennsylvania.
1-66
-------
L80
160
B
60
140
c
o
•H
•U
cfl
M
4-1
C
OJ
g 120
o
100
June 26-September 30, 1973
October 1-31, 1973
80
0000 0200 0400 0600 0800 1000 1200 1400 1600 1800 2000 2200 0000
Time of Day (EOT)
Figure 32. Mean diurnal ozone concentrations at Kane,
Pennsylvania for June 26 to September 30 and
October 1 to 31, 1973.
mean ozone concentrations at Kane, Pennsylvania during October 1973 are
lower at all hours of the day than during the June 26-September 1973 period.
Mean diurnal curves of ozone and nitrogen dioxide concentrations
at Kane, Pennsylvania, Coshocton, Ohio; and Lewisburg, West Virginia
are shown in Figures 33 and 34, respectively.
5.2.3 Effect of Change of Season
The primary reason for extending the monitoring period
at Kane, Pennsylvania was to examine the effect of change of season,
in particular the onset of freezing temperatures, on ozone concentrations.
1-67
-------
160
140 —
120 —
bO
3-
c
O
•H
1J
cfl
c
0)
O
C
O
u
100
Kane, Pa.
Coshocton, Ohio
• Lewisburg, W. Va.
0000020004000600080010001200140016001800ZOUOZ2000000
Time of Day (EOT)
Figure 33. Mean diurnal ozone concentrations at Kane, Pennsylvania;
Coshocton, Ohio; and Lewisburg, West Virginia from
October 1 to November 2, 1973.
1-68
-------
30
20
e
too
o
c
0)
c 10
o
o
O Kane, Pa.
& Coshocton, Ohio
• Lewisburg, W. Va,
,_
Minimum Detectable Limit
.000 T200 1400 T60D IK
0000 0200 0400 0600 0800 1000 T200 itoo itoD T^DD 2&TO 2^00
Time of Day (EOT)
Figure 34. Mean diurnal nitrogen dioxide concentrations at
Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia from October 1 to November 2, 1973.
Tt Ls known that hydrocarbon species capable of participating in the
photochemical generation of ozone are emitted by trees and other forms
of vegetation. It was speculated therefore, that with the death of
leaves due to frost, the emission of hydrocarbon species from trees
would decrease and further that such a decrease might lead to reduced
ozone concentrations. In Figure 35 mean daily ozone concentrations and
minimum daily temperatures for September and October 1973 are compared.
Freezing temperatures occurred on September 19 and 21 (the usual date
for the first killing frost is September 23) and again in October,
yet higher mean daily ozone concentrations occurred during October.
Obviously, no simple relationship between freezing temperatures and
ozone concentrations exists.
1-69
-------
300
290
280
273
270
v\
w
Freezing
\/ IA Ak
v \ \rv\
180
160
140
120
E
00
100
80
60
40
20
II I I I III I I I
1 5 10 15 20 25 30 1 5 10 15 20 25 30
September October
Figure 35. Mean daily ozone concentration at Kane, Pennsylvania
and daily minimum temperature at Bradford (Pennsylvania)
Regional Airport for September and October 1973.
1-70
-------
6.0 CONCLUSIONS
The principal conclusions arising from the statistical examination
of the data collected during the summer of 1973 are as follows:
3
1) The NAAQS for photochemical oxidants (160 ug/m for one hour
not to be exceeded more than once per year) was exceeded during
approximately 37, 30, 20, and 15 percent of the hours at McHenry,
Maryland; Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia, respectively from June 26 to September 30, 1973.
2) Nitrogen dioxide concentrations were at or near background
levels at Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg,
West Virginia.
3) Mean diurnal ozone concentration curves for the four
stations are very similar, varying mainly in magnitude of hourly
ozone concentration.
4) High correlations (.468-.678) significant at the .05 level
exist between simultaneous hourly ozone concentrations measured
at the four stations. The differences in the ozone-ozone correlation
coefficients are not significant.
5) There is no apparent lag effect, i.e., changes in ozone
concentration at one station do not lag behind changes in ozone
concentration at the other stations for the lags investigated.
6) The occurrence of high ozone concentrations at nonurban
locations is widespread, affecting a four-state area in eastern
United States.
1-71
-------
REFERENCES
1. Investigation of High Ozone Concentration in the Vicinity of
Garrett County, Maryland and Preston County, West Virginia.
Research Triangle Park, N.C. : Research Triangle Institute,
January 1973 (Also issued as Environmental Protection Agency
Report No. EPA-R4-73-019).
2. Mount Storm, West Virginia-Gorman, Maryland, and Luke, Maryland-
Keyser, West Virginia, Air Pollution Abatement Activity.
Air Pollution Control Office Publ. No. APTD-0656. Research
Triangle Park, N.C.: Environmental Protection Agency,
April 1971.
3. Richter, H. G. Special Ozone and Oxidant Measurements in Vicinity
of Mount Storm, West Virginia. Research Triangle Park, N.C.:
Research Triangle Institute, October 1970.
4. Ripperton, L. A., H. Jeffries, and J. J. B. Worth. Natural
Synthesis of Ozone in the Troposphere. Environ. Sci. and
Tech. .5, 246-248, 1971.
5. Miller, P. R., M. H. McCutchan, and H. P. Milligan. Oxidant Air
Pollution in the Central Valley, Sierra Nevada Foothills,
and Mineral King Valley of California. Atmos. Environ. 6^,
623-633, 1972. ' ~
6. Wunderle, J. Personal communication to W. D. Bach, Jr., 1973.
7. Beard, C. Personal communication to J. J. B. Worth, 1973.
8. Hodgeson, J. A., R. K. Stevens, and B. E. Martin. A Stable Ozone
Source Applicable as a Secondary Standard for Calibration of
Atmospheric Monitors. Air Quality Instrumentation, Vol. 1,
John Scales, ed., 149-150, ISA, Pittsburgh, Pa, 1972.
1-72
-------
APPENDIX A
CALIBRATION METHODS AND PROCEDURES
1-73
-------
APPENDIX A
CALIBRATION METHODS AND PROCEDURES
Dynamic calibration procedures were used to calibrate all analyzers
used during the field measurement period. Biweekly calibrations were
performed on each instrument using the procedures described below.
A-l General
Because the four stations were at elevations above mean sea level
adjustments to the data were necessary to reduce values to reference
condition of 25°C (298K) and 760 mm Hg. Adjustments to volume measure-
ments were made using the following equation:
P 298
V = V x -=— x —=22—
R 760 t + 273
where
V = volume of air at reference conditions, liters,
R
V = volume of air at sampling conditions, liters,
P = barometric pressure at sampling conditions, mm Hg, and
t = temperature at sampling conditions, °C.
Table A-l summarizes the sampling conditions assumed for
each site.
Table A-l. ALTITUDE-PRESSURE RELATIONSHIP FOR SAMPLING SITES
Station
McHenry, Md.
Lewisburg, W.Va.
Kane, Pa.
Coshocton, Ohio
Altitude above
mean sea level
meters
885
705
630
354
feet
2,900
2,301
2,060
1,160
Room
tempera-
ture,
°C
25 ± 2°
25 ± 2°
25 ± 3°
25 ±3°
Baro-
metric
pressure
mm Hga
682.5
697.7
704.1
727.7
Volume of 1
liter at ref-
erence condi-
tions (25°C,
760 mm Hg)-liters
0.90
0.92
0.93
0.96
Derived from Table p. 9-4, Handbook of Air Pollution, PHS Publication No.
999-AP-44. "Barometric Pressure at Various Altitudes."
1-75
-------
The same adjusted volume was used each time a calibration was
performed.
The general procedure followed during a typical calibration of any
analyzer was as follows:
1) The portable calibration unit (described below) and Spectronic 20
spectrophotometer were turned on and allowed to warm up for
30 minutes. During this time, the potassium iodide (KI)
sampling train was assembled and the volume flow rates of
analyzers, the sampling train, and the calibration unit were
checked and adjusted as necessary with a wet test meter and/or
a bubble flowmeter.
2) The analyzer sample inlet line (or calibration gas inlet line
in the case of the hydrocarbon analyzer) was disconnected from
ambient air and connected to a source of zero air. The time
of each operation was noted both in the instrument logbook and
on the strip chart recorder. The instrument was allowed to
sample zero air for a period of time sufficient to establish a
valid zero output. This was best discerned by observing and
noting the digital voltmeter output and the strip chart
recorder trace. Then a known concentration of calibration
gas was introduced (~80% of full-scale response) and the
readings recorded after the signal stabilized. No adjustments
were made to any part of the analyzer system prior to completing
this step.
3) If the need for adjustment was indicated by zero and/or span
drift or maintenance was required; it was done; then a dynamic
multiple point calibration of the system was performed by
introducing successive pollutant concentrations approximating
10, 20, 40, 60, and 80 percent of the operating range of the
instrument. From the record of the instrument output, a
calibration curve was constructed for use in data reduction.
Steps (2) and (3) constituted the preliminary and final
calibrations, respectively.
4) The analyzers were again connected to ambient air.
1-76
-------
A-2 Ozone Analyzers
A dynamic calibration system producing ozone by ultraviolet
A-l/
irradiation of oxygen—— was used to calibrate the gas-phase chemilumi-
nescent ozone analyzers. The ozone generator consisted of a shielded
mercury vapor lamp (20.3-cm in length) which irradiated clean compressed air
flowing through a quartz tube (1.5-cm in diameter). By varying the length
of the lamp exposed to the air and the total flow of compressed air
(usually set at 5.0 1/min), ozone concentrations from zero to
3
approximately 1 ppm (1960 Ug/m ) were produced,
A portable calibration unit consisting of a regulated power supply,
zero air source, calibrated rotameter, ozone generator, mass flowmeter
for nitric oxide mixtures, and a glass manifold with sampling ports
was assembled. This unit was transported from site to site for
calibration of each ozone (as well as nitrogen dioxide) analyzer.
A schematic diagram of the ozone calibration system is shown in
Figure A-l.
In order to obtain a reference measure of the ozone output of
A-2/
the calibration unit, the neutral-buffered Id analysis method—-
was used for each calibration point. Figure A-l shows the arrange-
ment of the two impingers containing the absorbing reagent. The
analyzer and the bubbler train sampled simultaneously from the glass
manifold. The volumetric flow (~1.0 1/min) through the KI sampling
train was determined with a calibrated wet test meter as shown in
Figure A-2.
The stepwise calibration procedure was as follows:
1) Install the ozone generator, allowing it to warm up for
30 minutes. Check and set all volume flowrates with a wet
test meter. Set up KI sampling train and adjust sampling
rate to 1 1/min.
2) Flush the system with a large concentration of ozone by
adjusting the sleeve to uncover the UV lamp 15 cm. After
several minutes, close the sleeve entirely and flush with
zero air.
1-77
-------
CO
£1
o
6
0)
4-1
to
C
o
•H
AJ
cfl
O
0)
C
o
N
O
I
<
a)
&0
•H
1-78
-------
«t
UJ
Id
o
a
•rH
cd
(-1
4-J
00
c •
•H en
rH C
CL, O
B -H
en -H
•a
rH C
!*i O
O
M-l
O i-H
td
c c
o o
•H -H
4-1 4-1
CT3 fd
V-i )-i
,£> 0)
rH O
cfl
a M
ai
^-J T3
o a
MH 3
en cu
3 4-1
P. O
50
•H
1-79
-------
3) Connect the ambient sampling line from the analyzer to the
calibration manifold. After 10 minutes, record the zero point
in mv output and chart divisions.
4) Pull the sleeve to uncover 1.50cm of the UV lamp. Allow the
output to stabilize for 10 minutes and then connect KI
bubbler train. Sample long enough to produce a detectable
color change in the neutral buffered KI solution.
5) Establish the slope of the spectrophotometer calibration
curve. With a 2.54cm cuvette, the spectrophotometer employed
in this study gave a slope equivalent to 1.02 yg 0,,/ml
absorbing reagent/absorbance unit (a.u.). The same cuvette
was used for the entire study. The spectrophotometer was
A-2/
calibrated with standard iodine solution on two occasions :
at the beginning of the study and midway through the study.
6) Measure the absorbance of the KI solution at 352nm using the
spectrophotometer. Repeat the KI bubbler procedure and obtain
a second absorbance value at the same sleeve setting. If the
agreement is reasonable (+ 5%) for identical sampling conditions,
proceed to larger sleeve settings and repeat the procedure.
3
7) Compute the concentration of ozone in yg/m for each sleeve
setting as follows: ^
,.-"£>-"^
a) Determine the volume of air sample^f>y the KI bubbler
assembly and correct it to re^€rence conditions of 25°C
and 760mm Hg as followsu
~3
3
V = volume-'of air at reference conditions, m ,
R
V = volume of air at sampling conditions, liters,
P = barometric pressure at sampling conditions, mm HG,
t = temperature at sampling conditions, °C, and
-3 3
10 = factor converting liters to m .
3
b) Solve for yg/m as follows:
, 3 (absorbance)(1.02 yg On/ml/a.u.)(10 ml KI reagent)
0 yg/m = 3
VR
1-80
-------
For example:
Station: Kane, Pennsylvania
Sleeve setting: 2 cm
Volume sampled under site conditions: 101
Pressure: 704.1 mm Hg
Temperature: 25°C
KI absorbarice: 0.105
(0.105 a.u.)(1.02 vgO /ml/a.u.)(10 ml)
0 ,ug/ni = ^ = 115.2 yg/in
0.0093 m
8) From the instrument output in mV and the ozone concentrations,
prepare a calibration curve for the analyzer.
A-3 Nitrogen Dioxide Analyzer
The NO-NO-NO analyzer was calibrated by gas-phase titration.^—^
ic X
The technique makes use of the rapid gas phase reaction between NO and
0 to produce a stoichiometric quantity of NO .
After the preliminary zero and span checks, the first step in
the final calibration is the introduction of zero air into the
analyzer. After 10 minutes, a zero reading is taken on the NO, NO ,
and NO channels.
x
Before transporting the NO calibration gas cylinders to the
field, the NO concentration of the contained calibration gases was
A-3/
verified using the technique of Hodgeson and associates.——" The
procedure consisted of titrating an NO concentration of 1.0 ppm
with successive concentrations of ozone (0-0.8 ppm) produced by
an ozone generator referenced to the neutral-buffered KI procedure.
The resultant NO detector outputs, after stabilization at each
titration points (i.e., 0.0, 0.1, 0.2, ... 0.8 ppm ozone added),
were plotted as concentrations ppm (y-axis) versus 0_ concentration
added, ppm (x-axis). A straight line drawn through the linear
portion of the titration curve was extrapolated to the x-axis. The
concentration at the x-axis intercept, C', was the 0 concentration
equivalent to the initial diluted NO concentration. The cylinder
NO concentration was then calculated as follows:
1-81
-------
N° FNO
where
C _ = cylinder NO concentration, ppm,
F _ = measured NO flow, ml/min,
C' = equivalence point 0.,, concentration, ppm, and
F = total clean air flow, ml/min.
The NO portion of the analyzer was calibrated by dynamic flow
dilution of the cylinder gas. This was accomplished by metering
the NO from the cylinder through a calibrated mass flowmeter and
then into the dilution system of the ozone generator. To calibrate
the NO portion of the analyzer, a constant NO concentration of
3
approximately 940 yg/m (0.5 ppm) was produced by dilution. Ozone
was added in increments from the generator. Decrements observed
on the spanned NO detector, are then equivalent to the N0_ concentra-
tion produced by the 0,. source. Since the NO- produced was equivalent
to 0,, consumed, the calibrated 0,, source served as a calibrated N0?
source when NO was present in excess. After adequate time (-10 min)
for stabilization at each point, the mV output of each channel was
recorded.
The NO concentration was deduced from the decrease of the NO
signal, and a calibration curve relating NO concentration arid
analyzer mV output was constructed.
A-4 Hydrocarbon Analyzer
Nonmethane hydrocarbons were determined at three stations using
the Bendix Model 8201 Ambient Hydrocarbon Analyzer. Calibration was
accomplished using hydrocarbon-free air as zero gas and mixtures
of methane in hydrocarbon-free air as span gases. The calibration
gases were purchased from Scott Research Laboratories who certified
the contents of each gas bottle.
By appropriate valve switching, it was possible to use the
methane to calibrate the total hydrocarbon (THC), methane, and
nonmethane hydrocarbon (NMHC) channels. The zero and span calibration
steps were as follows:
1-82
-------
ZERO
1) Place sample valve (#1) and backflush valve (#2) on off
position and meter/output switch in electrometer position.
2) Observe recorder baseline while component gates are open
(i.e., CH light on and THC light on).
3) Adjust zero control for the THC and CH to zero by observing
recorder output ot meter while switched to appropriate
output.
4) After one complete cycle, adjust nonmethane zero control
for zero indication on recorder or meter.
5) Return valves #1 and #2 to auto position.
6) Connect zero air cylinder for dynamic zero; flow rate
of 200 ml/min. Switch operating mode switch to calibrate.
7) Compare mV readings with previous zero data and record;
discrepancies suggest contamination.
SPAN
1) Connect cylinder containing highest concentration to the
calibration inlet. Set flow rate at 200 ml/min.
2) After at least two cycles, set the span control for THC
and CH, to the appropriate mV reading.
3) After completing the span operation on THC and CH , switch
output to electrometer position.
4) Observe strip chart for one cycle. Manually override valve #1
to the off position for the duration shown on the strip chart
in Figure A-3. This will cause the CH, output to equal the
THC output.
5) Adjust the nonmethane span control to give the appropriate mV
reading with the output switched to nonmethane.
6) Run additional calibration cylinders and record mV readings,
but do not adjust span controls.
7) Recheck nonmethane calibration as outlined above for each
calibration level.
1-83
-------
Cycle time =
6.7 in .
-rr= 3. 35 mm
2 in/min
THC light ON
#1 ON
Auto zero On
#1 OFF, #2 On
Manually switch
valve #1 to OFF
Auto zero ON
#1 ON
ZERO
#2 OFF
Figure A-3. Calibration cycle for nonmethane hydrocarbon analyzer,
-------
REFERENCES
A-l Hodgeson, J. A., R. K. Stevens, and B. E. Martin. A Stable Ozone
Source Applicable as a Secondary Standard for Calibration of
Atmospheric Monitors. Air Quality Instrumentation, Vol. 1,
John Scales, ed., 149-150, ISA, Pittsburgh, Pa., 1972.
A-2 40 CFR 50, Appendix D,
A-3 Hodgeson, J. A., R. E. Baumgardner, B. A. Martin, and K. A. Rehme,
Stoichiometry and Neutral lodometric Procedure for Ozone by
Gas-Phase Titration with Nitric Oxide. Anal. Chem. 43.
1123-1126, 1971.
1-85
-------
APPENDIX B
PERFORMANCE CHARACTERISTICS AND
OPERATIONAL SUMMARIES FOR INSTRUMENTS
1-87
-------
APPENDIX B
PERFORMANCE CHARACTERISTICS AND
OPERATIONAL SUMMARIES FOR INSTRUMENTS
B-l Instrument Performance Characteristics
Minimum detectable concentrations, ranges, and precisions for
the air quality monitoring instruments used in this study are
summarized in Table B-l.
B-2 Operational Summaries for Instruments
Figure B-l provides an operational summary for each air quality
monitoring instrument at each station.
Table B-l. INSTRUMENT PERFORMANCE CHARACTERISTICS
Instrument
Bendix Model 8000
Chemiluminescent
Ozone Analyzer
McMillan MEG 1100
Chemiluminescent
Ozone Analyzer
Bendix Model 8101 B
Chemilumines cent
NO-NO 0-NO Analyzer
2 x J
Bendix Model 8201
Ambient Hydrocarbon
Analyzer
Parameter
°3
0Q
3
NO
NO
NO*
THC
CH4
NMHC
Minimum
Detectable
Concentration
/ 3
yg/m
9.8
9.8
6.1
9.4
53
93
ppm
0.005
0.005
0.005
0.005
0.005
0.080
0.140
Upper Limit
Ug/m
392
980
615
940
6670
6670
6670
ppm
0.2
0.5
0.5
0.5
0.5
10
10
10
Precision
(% of indicated
concentration)
+ 0.5
+ 0.5
+ 1.0
1-89
-------
tfl
>
3
,-Q
CD
•H
O
c
O
4-1
U
O
f.
O
C
0)
rc
o")
O
U
H
2
W
r~ CO o-s O
CN CM CN CO
cu dJ
-------
cfl
>
00 O
3
en
S
o
o
f
w
o
u
o
u
60
C
•H
0)
cx
o
0)
E
in
C
o
u
-------
rt
>
O
o
03
C
O
0)
C
ex)
m
O
w
u
o
\N
O
*t
M
1
vo \N
o
o
M
U
-
CntnCD0303C/30303t/3Cn0303030303030303COCDC0030303tn0303CnO)0303
oc
C
•H
0)
ex
o
o
C
0)
e
03
I—i
I
00
C
0)
ex
o
C
cu
a
^
4-1
03
C
C
O
•H
—4
CO
CJ
I
C
o
•H
03
C
0)
i-l
J-l
UD
•H
en
to
C
o
•H
4J
CO
S-i
0)
(X
O
d
o
CJ
II
m
s
t>0
•H
1-92
-------
60
0)
i-l
u
u
CJ
u o
u
60
c
cu
D.
O
QJ
e
3
W
C
6C
c
to
t-l
-------
00
Va
u
U
rt
)-i
0)
P*
O
Lewisburg
Ohio
O
U
U
U
no
nstrumen
C
0)
Coshoc
cd
PH
C
03
o
U
O
z
U
u
O
O -, U
O
o o
operating
nstrumen
f.
C
•H
t-i
O
cfl
O
•H
4J
CO
(-1
0)
t-i
C
0)
re
CJ
Calibration
4-1
C
o
O
I
m
0)
M
d
u
PH
ooooooooooooooooooooooooooooooo
4_JJ_14-J4-l4JiJJ-)4-J4-)-UW4-J4-l4-)-U-UtJ4J^J^-l4-14-)4-J4JJ-J4_)U4-J4-)-U4-J
oooooooooooooaaoocjaooocjaUoouooo
ooooooooooooooooooooooooooooooo
ion
ns
1-94
-------
APPENDIX C
HYDROCARBON ANALYSIS OF GRAB SAMPLES
1-95
-------
APPENDIX C
HYDROCARBON ANALYSIS OF GRAB SAMPLES
This appendix presents the results of gas chromatographic analyses
of grab samples performed by the Chemistry and Physics Laboratory, National
Environmental Research Center, Research Triangle Park, North Carolina.
C-l Analytical Results
The stations, dates, and times at which grab samples were collected,
the dates and times at which they were analyzed, and a sample code
designation are shown in Table C-l. Components comprising the hydro-
carbon content of each analyzed sample are given in Table C-2. Each
component concentration is given as parts per billion carbon (ppbc).
3
To convert ppbc to yg/m , multiply the component concentration by
0.67-
C-2 EPA Analysts' Comments
"Three gas chromatographic procedures were used to determine hydro-
carbon composition. Each individual peak observed has been identified by
at least one hydrocarbon species. (The only exceptions are four
unknown peaks observed in sample N-8). There is extremely high confi-
dence that the components representing the C -C._ aliphatics are correctly
identified as aliphatic hydrocarbons. However, the peaks identified as
the C.--C _ aromatics have known retention time interferences from halo-
genated and oxygenated compounds. Therefore, the confidence of the
correct identification of the peaks observed on the aromatics chromato-
gram is not high. Terpene hydrocarbons such as a- and 3-pinene have
retention times in the region of some of the aromatic peaks.
"The component indicated by the asterisk in Table C-2 is most
probably acetaldehyde. This species is indicated by the peak geometry.
Acetaldehyde and 2-Methylbutene-2 have similar retention times.
However, the acetaldehyde peak tails severely while the corresponding
olefin does not. Acetaldehyde has been identified in the urban atmosphere
and is a photochemically produced oxygenate of irradiated urban air.
"The very high concentration of the ethylene observed in samples N-l
through N-6 is an indication of ethylene pollution of the nearby
1-97
-------
ambient air by the exhaust of the ethylene-ozone chemiluminescent
analyzer. Samples N-6 through N-8 are the usual levels of ethylene in
rural ambient air.
"The C -C_ hydrocarbons are observed in most rural air samples. The
higher concentration of the C.-C paraffins in some of the samples would
indicate a local source such as the gasoline evaporative emissions from
a nearby automobile.
"The very high concentration of toluene is difficult to explain.
A nearby source is expected. Interfering halogenates or oxygenates
having similar retention times as toluene are not known. The analyst
doubts that natural sources of an unidentified component having a
similar retention time as toluene could explain the magnitude of the peak.
"The acetylene/CO ratio is somewhat consistent in the samples with
the exception of sample N-8. Since it is difficult to imagine any other
component having a similar retention time to acetylene on the silica
gel column, the acetylene must have come from some nearby source such
as an acetylene torch.
"When samples are collected in Tedlar bags, care must be taken not
to expose the outside surface of the bags to high concentration of any
hydrocarbon component or solvent. The 2 mil thickness of Tedlar is
permeable to most of these components. "
1-98
-------
Table C-l. GRAB SAMPLES COLLECTED FOR HYDROCARBON ANALYSIS
RATION
Lewisburg, W. Va.
Kane, Pa.
Kane, Pa.
Lewisburg, W. Va.
Coshocton, Ohio
Lewisburg, W. Va.
McHenry, Md.
Coshocton, Ohio #1
Coshocton, Ohio #2
DATE AND TIME
OF SAMPLE COLLECTION
10-16-73
1400-1420
10-17-83
1015-1040
10-17-73
1045-1115
10-16-73
1215-1250
10-4-73
1100-1150
10-5-73
1300-1380
10-2-73
1300-1330
10-24-73
1200
10-24-73
1200
DATE AND TIME
OF SAMPLE ANALYSIS
10-18-73
1400
10-18-73
1500
10-18-73
1600
10-19-73
10-11-73
1530
10-12-73
1300
10-12-73
1400
10-26-73
1330
10-26-73
1430
CODE
DESIGNATION
Nl
N2
N3
N4
N5
N6
N7
N8
N9
1-99
-------
Table C-2. HYDROCARBONS IDENTIFIED IN GRAB SAMPLES
SAMPLE
COMPOUND
Ethane
Ethylene
Propane
Acetylene
Isobutane
n-Butane
Propylene
iso-Butylene
trans-Butene-2
Methyacetylene
cis-Butene-2
iso-Pentane
n-Pentane
Pentene-1
2-Methylbutene-l
trans-Pentene-2
cis-Pentene-2
2-Methybutene-2
Cyclopentane
2-Methylpentane
3-Methylpentane
Hexane
2 , 4-Dimethylpentane
Me thy cyclopentane
1 cis 3 Dimethyl-
cyclopentane
2,2,4 Trimethyl-
pentane
Trans 3 Dimethyl-
eye lopentane
Toluene
N-l
10.3
133.0
5.7
4.0
2.1
8.5
2.0
2.2
1.3
0.7
8.2
3.8
0.0
0.0
0.0
0.0
7.9
3.8
2.1
2.8
N-2
10.2
97.4
6.5
4.1
3.3
8.9
2.5
2.4
1.0
0.7
8.1
3.7
0.0
0.0
0.0
0.0
4.0*
3.4
1.4
2.1
53.2
CONCENTRATION in ppbc (vol.
N-3 N-4 N-5 N-6
14.7
172.6
11.0
5.9
5.8
21.7
1.7
2.7
1.3
0.8
19.7
8.3
1.1
1.1
1.4
2.0
2.1
9.9
4.3
5.1
189.9
7.3
275.0
5.4
3.7
4.2
14.7
0.2
1.1
0.2
17.7
7.2
0.9
1.0
1.0
8.0
2.5
2.7
219.2
36.4
396.9
26.4
3.9
9.9
30.0
0.5
1.3
1.3
0.6
33.1
12.4
0.7
0.9
1.1
3.1
12.3
3.4
6.2
1.0
1.3
13.4
375.0
17.6
2.2
11.3
45.2
1.1
3.9
2.0
1.0
55.8
27.3
1.6
1.2
3.2
2.6
2.4
4L.6
19.0
19.5
3.0
8.5
11.7
117.5
/vol . )
N-7
6.8
4.5
4.7
3.2
9.1
74.0
0.7
3.0
1.5
1.0
126.2
26.0
1.4
1.2
2.2
1,0.0
4.0
33.6
9.3
8.0
>11.2
4.6
19.9
4.7
1850.0
N-8
31.4
7.0
18.0
63.0
20.3
61.4
0.5
6.4
4.4
2.0
55.2
20.6
1.0
1.7
3.8
2.6
4.0
21.9
7.6
9.3
>12.0
4.9
255.6
N-9
34.8
8.7
19.3
11.5
14.1
42.4
0.8
4.5
3.0
1.2
35.6
18.8
0.8
1.0
1.6
0.6
1.3
14.7
5.4
6.4
>8.0
935.2
1-100
-------
Table C-2. HYDROCARBONS IDENTIFIED IN GRAB SAMPLES (cont'd)
SAMPLE
COMPOUND N-l
Nonane
Ethylbenzene
p-Xvlene
m-Xylene
o-Xylene
n-Decane
Isopropylbenzene +
Styrene
n-Propylbenzene
m+p Ethyl toluene
1,3,5-Trimethyl-
benzene
tert-Buthybenzene +
o- Ethyl toluene
sec-Butylbenzene +
1,2,4-Trimethyl-
benzene
Unknown
1,2,3 Trimethyl-
benzene
n-Butylbenzene +
p-Diethylbenzene
Sum of Unknown
Peaks
Methane 1630.0
Carbon Monoxide 530.0
N-2
10.
3.
3.
10.
5.
7.
1.
3.
1.
0.
6.
2.
1.
1.
1560.
440.
8
2
0
2
4
5
1
7
0
4
4
2
7
4
0
0
CONCENTRATION in ppbc (vol. /vol.)
N-3 N-4 N-5 N-6 N-
10.0
1.2
1.0
3.4
1.8
3.4
0.2
1.8
1.0
0.5
4.8
1.4
0.0
0.0
1550.0
380.0
7
2
2
7
3
7
1
6
1
1
8
1
1
4
1510
320
.1
.1
.0
.7
.5
.5
.0
.7
.9
.4
.4 13.8
.5
.6
.2
.0 1690.0
.0 378.0
9
9
23
11
7
3
11
4
3
7
2
1510
310
.0
.6
.2
.4
.6
.2
.8
.4
.6
.7
.2
.0
.0
4
4
11
6
9
1
5
1
1
9
1
1510
310
7
.0
.2
.2
.0
.2
.5
.0
.7
.2
.2
.8
.0
.0
N-8
13.5
23.7
6.1
30.0
19.0
36.8
38.9
16.6
15.3
8.4
3.4
4.4
37.0
1660.0
430.0
N-9
9.8
5.7
2.9
10.0
6.8
11.1
7.6
4.3
4.2
4.8
1.5
4.4
1730.0
500.0
1-101
-------
INVESTIGATION OF OZONE AND OZONE PRECURSOR
CONCENTRATIONS AT NONURBAN LOCATIONS IN
EASTERN UNITED STATES
Part 2. Quality Assurance Program
by
C. E. Decker
W. C. Eaton
T. M. Royal
J. B. Tommerdahl
Research Triangle Institute
Research Triangle Park, N. C. 27709
Performed as Subcontract No. 33-73
from PEDCo Environmental Specialists, Inc.
Contract No. 68-02-1343
EPA Project Officer: E. C. Tabor
Quality Assurance and Environmental Monitoring Laboratory
National Environmental Research Center
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N. C. 27711
May 1974
-------
1.0 INTRODUCTION
Under a program initiated in June, 1973, the Research Triangle Insti-
tute conducted an investigation of ozone and ozone precursor concentra-
tions at non-urban locations in the eastern United States (EPA Contract
68-02-1077, RTI Project 41U-848). This was an outgrowth of a field study
of atmospheric ozone concentrations conducted in Garrett County, Maryland,
and Preston County, West Virginia, during the summer of 1972. Results of
the 1972 summer study indicated that transport of ozone rather than local
synthesis reaction (i.e. , photochemical ozone production) is responsible
for the unusually high ozone concentrations observed in this rural, moun-
tainous area. The 1973 summer field program, which consisted of data
collection (4 months), analysis, and interpretation, was designed to
evaluate the occurrence and extent of high ozone concentrations at
nonurban locations. The measurements made at the respective sites were:
0^ - McHenry, Maryland; 03, N02, NMHC - Kane, Pennsylvania; Coshocton,
Ohio; and Lewisburg, West Virginia. The purpose of the present study was
to conduct a quality assurance program relative to the air quality measure-
ments being conducted under Contract 68-02-1077.
Calibration of instruments is necessary for the successful pursuit of
any study in which sample analysis is employed. Calibration of instruments
in the field, because of logistic problems, is usually more difficult than
calibration in the laboratory. In the above-mentioned study the difficulty
was further compounded by having several widely separated field sites at
which atmospheric measurements were made.
2-3
-------
In the 1973 summer program, the instruments at each station were
calibrated on an orderly biweekly schedule. These calibrations provide a
check on accuracy of data and, if the data are accurate, allow a valid
comparison of both data and instruments at the various stations. In field
studies, however, a back-up system of checks and calibration is always
desirable, especially as a test of comparability of data taken simultaneously
at different sites.
A quality assurance program was designed to provide a direct test of
the comparability of data obtained from the four widely separated monitoring
stations. The program was carried out by taking a separate set of ambient
air analyzers, calibration system and calibration gases in the RTI Environ-
mental Monitoring Laboratory to each station and sampling from a manifold
with an intake common to both the mobile and the fixed station. The
instruments in the mobile unit were calibrated by a different team from
that performing the scheduled calibration of the fixed stations. Simul-
taneous sampling was conducted for a minimum of 40 hours at each of the
sites. Signals from the instruments in the fixed site and mobile laboratory
were recorded simultaneously on the data acquisition and on-line processing
system located in the Environmental Monitoring Laboratory.
Thus, the behavior of instruments at each site was compared to the
behavior of an additional set of instruments located in the Environmental
Monitoring Laboratory which for this study served as a primary standard or
reference point for each of the gaseous pollutants being monitored. The
Environmental Monitoring Laboratory was equipped with the best available
instrumentation for monitoring the pollutants in question. In addition,
2-4
-------
the equipment was operated, calibrated, and maintained by highly
qualified technical personnel, and all calibrations were conducted with
extra care under essentially laboratory conditions. In one step this
served to verify both the validity of the environmental data being
generated at the fixed station and the comparability of data from the
various sites. At the same time the precision of a set of similar instru-
ments was checked by running them side by side on a common sample. Analysis
of the resulting data provides a quantitative assessment of the individual
station's performance and the relationship of the measurements among the
four monitoring sites.
A description of the mobile Environmental Monitoring Laboratory and
the quality assurance program are presented in Sections 2.0 and 3.0,
respectively. Data comparisons, results of statistical analyses, and a
summary of results are presented in Sections 4.0, 5.0, and 6.0. The
calibration systems and detailed calibration procedures are described in
Appendix A. Data for ozone and nitrogen dioxide from the fixed and mobile
site analyzers and supplementary data (i.e., sulfur dioxide and hydrocarbons)
obtained from analyzers Located in the Environmental Monitoring Laboratory are
tabulated in Appendix B.
2-5
-------
2.0 MOBILE LABORATORY AND EQUIPMENT
2.1 Mobile Monitoring Laboratory
The RTI Environmental Monitoring Laboratory used in this study
is a self-contained, 31-foot motorized vehicle custom-built for air qual-
ity monitoring. It was outfitted with the latest complement of air qual-
ity monitoring instrumentation, a self-contained calibration system for
each pollutant analyzer, and a digital magnetic tape recording system
coupled to a minicomputer capable of online data processing and print-
out of real-time air quality data in physical units and storage of the
raw data on magnetic tape. The motorized van has a self-contained motor-
generator for generation of electrical power; a controlled environment
(heating and cooling); an ac voltage regulator; a glass and Teflon mani-
fold system for passage of air samples to the monitors; a tower for
mounting various meteorological sensors; and storage space for compressed
air and gas tanks.
Two interior views of the vehicle are shown in Figures 1 and 2; ex-
terior views of the mobile unit may be seen in the photographs of the
setup at the respective sites.
2-6
-------
I •»•
Figure 2.
2-7
-------
2.2 Air Quality Analyzers and Calibration System
For this study, the mobile laboratory was equipped with
analyzers capable of continuous measurement of ambient air concentra-
tions of ozone (0 ); oxides of nitrogen (NO ); (NO), (NO ); total sulfur
j X Z
(S); carbon monoxide (CO); methane (CH ); total hydrocarbons (THC); and
nonmethane hydrocarbons (NMHC). All air pollutants except NO and NMHC
are measured by direct detection. Nitrogen dioxide and NMHC are computed
by subtracting NO from NO and CH, from THC, respectively. Table I gives
X 4
details for each analyzer.
Table I AIR QUALITY ANALYZERS
Instrumentation
Bendix 03
Model 8002
Bendix NO, N02, NO
Model 8101-B X
RTI 0
Model 525
Bendix Total Sulfur
Model 8300
Principle of Detection
Gas phase chemiluminescence
Gas phase chemiluminescence
Solid phase chemiluminescence
Flame photometry
Beckman CH4, CO, THC Flame ionization
Model 6800
Bendix CH4, CO, NMHC Flame ionization
Model 8201
Bendix CO
Model 8501-5FA
Nondispersive infrared
Range
0-0.2 ppm
0-0.5 ppm
0-0.5 ppm
0-1.0 ppm
0-10.0 ppm
0-10.0 ppm
0-20.0 ppm
2-8
-------
Apparatus and associated equipment utilized in the mobile laboratory
for calibration of ambient air quality analyzers included the following:
apparatus to perform manual iodometric analyses (neutral buffered KI Method),
KI sampling train, spectrophotometer (Spectronic 70), and ultraviolet ozone
source; gas phase titration system for producing N0? by oxidation of NO to
N00 with ozone; and certified cylinders of CH and CO in zero air.
The contents of compressed gas cylinders used for calibration purposes
and their contents as certified by supplier's analysis are presented in
Table II.
Table III presents a synopsis of pollutants measured by instruments
carried in the mobile laboratory and methods used for calibration. Included
are the precision and accuracy associated with the calibration techniques.
Table II CALIBRATION GASES
Cylinder
Instrument Content
Bendix NO NO in N_
x /
Beckman 6800 CH. in Air
4
CH. in Air
4
CH, in Air
CO in Air
Supplier
Scott
Scott
Scott
Scott
Scott
Supplier's
Analysis
50
99
4
1
95
.2
.6
.64
.6
.9
ppm +
ppm +
ppm +
ppm +
ppm +
2%
2%
2%
2%
2%
2-9
-------
Table III
CALIBRATION TECHNIQUES
Pollutant
Ozone
Nitrogen Dioxide
Nitric Oxide
Sulfur Dioxide
Carbon Monoxide
Methane
Total Hydrocarbons
Calibration Technique
UV-Ozone Generation Referenced
to Neutral-Buffered KI Proce-
dure (as published in Federal
Register, April 30, 1971
Edition)
Gas Phase Titration (as
published in Federal Register,
June 8, 1973 Edition)
Dilution of Standard Cylinder
Gas (NO in N2)
NBS Permeation Tube
Standard Cylinder Gas
(CO in Air)
Standard Cylinder Gas
(CH4 in Air)
Precision/Accuracy
of Calibration
Method
Estimated
Accuracy +_
Precision of
Calibration,
.+ 2%
Estimated
Accuracy.
Precision of
Calibration,
Accuracy of Component
Analysis ......... ±
Precision of
Calibration
+ 2%
Accuracy
Precision of
Calibration,
Accuracy of
Analysis
Precision of
Calibration.
.+ 2%
Accuracy of
Analysis..
Precision of
Calibration
2-10
-------
2.3 Data Acquisition System
The Environmental Monitoring Laboratory data system consists
basically of a digital magnetic tape data acquisition system and a mini-
computer. The signals from the respective analyzers are scanned at five-
minute intervals. These sampled voltages are digitized and recorded on
the magnetic tape in computer compatible format; in addition, the scanned
data are introduced into the minicomputer where the appropriate transfer
functions are applied to each of the analyzer signals and the resulting
values in physical units are printed out. An example of the data out-
put for a five-minute scan is shown in Figure 3. The five-minute data are
accumulated and at the end of each hour a summary which includes the five-
minute data plus the hourly average is printed out as shown in Figure 4.
The output signals from the respective analyzers in the fixed station
were coupled into the mobile laboratory data system so that all signals
were sampled and recorded by a common system. This was accomplished by run-
ning a multiple-pair shielded cable from the mobile laboratory to the fixed
site station and picking up the signals at the input to fixed station magnetic
tape recording unit.
The system is so designed that all operating mode information and
calibration values may be entered into the system by digital mode switches
and via the teleprinter keyboard, respectively. Any changes in instrument
status are automatically compiled and calibration information (i.e., trans-
fer functions) is also updated automatically via the minicomputer.
Backup data recording equipment consists of strip-chart recorders
for each of the analyzer outputs and a digital printer which prints out
the same data that are recorded on magnetic tape.
2-11
-------
-.905
. 01 3
-. 787
9.370
. 0 PI /;
-.106
-. 1 nr-j
T-T i 03
HEM 03
PEN MO
EFN MO?
BEN T?
PE;: THC
PEX CH4
BEX CO
PEN 03F
EZWN0F
TEN M02F
PEM TIICF
BF.\T UMHC
EEM CH4
7,1973 1
-.176 -.050
.019 . 000
-.000 . 000
9.393 .000
-.000 -.000
-.106 -.610
3. 184-13. 163-
33.7093
102.7035
-1.151 I
2. 698?
6336.4707 T
5973. G0B5 T
98.8340
-25.2240
-6. 605?
1239.5124
2^3.4337
1040.3730
/
2:10 HOURS
-.002 . P"0
-.063 .000 15.
.02* -.006
.140 .021
-.001 -.000 -.
-.608 -.513 -.
13. 171-13. 1 70
UG/M
UG/M
UG/M Prin
UG/M Qutpi
TTG/M < $$$
TTG/M is o.
TTG/M
UG/M
TTG/M
TTG/M
TTG/M
TTG/M
UG /M
Printout of
the.
voltage measured on the
/ 65 channels
of the
/ Data Acquisition System
/
0 0 0 1.339
027-15.065
001 .038
158 .001
992 -.106 -.
511 -.512
0|70 . 000
!? ?' 0 —.00]
000 .000
013 .212
g^q -.609
607 -.106
tout of up to 20 Instruments
at in Engineering Un,its.
indicates Instrument Output
Pfline.
773
00 9
107
1 06
Figure 3. Sample data output for five-minute scan.
2-12
-------
Code for Status of Instrument
EASTERN REGIONAL OZONE STUDY
OCTOBER 5/1973
(Left Blank for Ambient
Data Point)
RTI 03 BEN 03 BEN NO BEN M02 BEN TS BEK THC I
TIME UG/M UG/M UG/M UG/M UG/M T TIG/M
20 0 90. 106. -1.
20 5 102. 102. -1.
2010 95. 99. -1.
2015 88. 96.
2020 82. 96.
?025 89. 102. -1.
2030 92. 101. -2.
2035 86. 95. -1.
23 H?, 86. 88. -3.
2045 79. 89. -2.
2050 89. 93. -2.
2055 91. 97. -2.
A LEPAGE 89. 97. -1.
t t
Hourly Averages
Printed Once per Hour Automatically
14. SSSSSSSX 1C92.
1?. $S$$S$$X 1106.
12. $$$$$S$X 1112.
14. $$SSJS£X 1086.
17. S$$$$$$X 1155.
11. $£$$$$SX 1137.
7. $$$$$£$X 1110.
1 6. SSSSSSSX '1 1 14.
9. £$$S£f$X 1123.
4. SJSSSSSX 1130.
6. JSJSSSSX 1093.
10. SSSSSSSX 1098.
1 1 . SS?$S$S 1113.
t
$$$ - Indicates less than 9
Valid Data Points During Pas
Hour, Hence, no Average is
Computed .
Air
3 EX CH>
UG/M
1001
099
Iff 43
less
998
1(?27
1 1 63
If IS
1 275
1071
1015,
101 5
1237,
t
Figure 4. Sample data output for hourly summary.
2-13
-------
3 .0 FliLD QUALITY ASSURANCE PROGRAM
3.L Procedure
The purpose of the program was to conduct a quality assurance
program relative to the air quality measurements being made at the four
stations operated under the referenced Contract. The specific locations
and respective site descriptions are given in Section 3.2. The pollutants
measured in the four stations were 0-, NMHC, and NO" . Operation of the
network consisted of daily checks by local personnel at each station and
biweekly maintenance and calibration of all analyzers by contractor
personnel. A single portable calibration system was utilized throughout
to calibrate all Oo and N02 analyzers during this study. Continuity was
maintained throughout the measurement program by using a single calibration
system and operator. The quality assurance program was designed to provide
a direct test of the comparability of the data obtained from the four
widely separated monitoring stations, within the constraints of time and
funds.
The mobile air monitoring laboratory described in Sections 2.1 and 2.2
was used in this quality assurance program. The program was conducted over
a relatively short period of time in order that the personnel and the
techniques they utilized would have minimal variability. Consideration was
given to the type of data, number of samples, period of comparison, and
calibration sequence in establishing the general field procedure presented
in Figure 5. The time indicated for each segment in the schedule is the
approximate time allocated for equipment warm-up, calibration, and comparison
and data acquisition, both prior to and after calibration of the fixed site
analyzers.
2-14
-------
O
I—I
<1 O
PH CO Q
pq M O
U
o
U
W
PH
CO
O
25
O
O
CO Q
MO
I—I PH M
i-l
W O
PH CJ
PH
Pi
W
PH
Q O
W M
X H
M H M
PS
ffi
CN
VVVXXVN
PS
§
U M
M H
'
PO
o
s§
3 H Pi
U co O
W P-i ta
CO CO
H
Pi
o
PH
CO
O
CO O
M O
Pi h-1
-------
Tho site visitation schedule was as follows:
August 27-30, 1973 — Lewishurg, West Virginia
September 5-7, 1973 — Kane, Pennsylvania
September 14-18, 1973 — Coshocton, Ohio
September 21 - October 3, 1973 — McHenry, Maryland
October 5-7, 1973 — Lewlsburg, West Virginia
A more detailed presentation of the sampling times and order of events is
given in Section 3.3.
2-16
-------
3.2 Location and Description of Sites
Lewisburg, West Virginia
The Lewisburg site was located immediately adjacent to
Greenbrier Valley Airport near Lewisburg, West Virginia. Elevation
above mean sea level is approximately 705 m (2301 ft.). The fixed site
station was a portable laboratory owned and operated by the Bendix Corpo-
ration. It was located on a hill overlooking the airport. Exposure
to the air was good in all directions. Figure 6 shows the environ-
mental monitoring laboratory at the Lewisburg location and the arrange-
ment of the sampling inlets.
i' >
Eigure 6. Lewisburg, West Virginia site,
2-17
-------
Kane, Pennsylvania
The northernmost station was located in Kane, Pennsylvania.
The borough of Kane is surrounded on three sides by the Allegheny
National Forest. The specific location of the fixed station was the
industrial arts room of the Kane Area Senior High School. The school
is situated at the highest point in the area, 630 m (2060 ft.) above
mean sea level. Initially, instrumentation was located on tables inside
the industrial arts room; after school started in the fall, the equipment
was moved to an EPA supplied mobile van just outside the industrial arts
room (in the far right-hand part of Figure 7). The air inlet was at a
point 1.8 m (6 ft.) above the building roofline and was at the same point
for both periods. The mobile laboratory was parked immediately next to
the building and a Teflon sampling line attached to the entry port of the
fixed station (see Figure 7).
Figure 7. Kane, Pennsylvania, site.
2-18
-------
Coshocton, Ohio
The Coshocton site was located at the North Appalachian
Experimental Watershed facilities near Coshocton, Ohio. The complex
of buildings is 354 m (1160 ft.) above mean sea level. The acreage
surrounding the station is used mainly for farming. There are no ob-
structions to air movement. The fixed station was located in a room
on the second floor of a structure known as the engineering building.
It was not possible to bring the mobile laboratory as close to the fixed
station as was desired. In order to sample air from the same inlet as
the fixed station, a 1 inch 0. D. Teflon line was mounted on a cable as
shown in Figure 8.
figure o. i^osnocton, Ohio site.
2-19
-------
Garrett County Airport, McHenry^_Mary_lanE!
The Garrett County Airport is at an elevation of approximately
885 m (2900 ft.) above mean sea level. The site provided excellent exposure
for ambient air monitoring instruments. The only pollutant measured at
the fixed site was ozone. The ozone monitor, strip-chart recorder, and
data acquisition system were located in a small workroom at the east end of
the hangar. It was possible to position the mobile laboratory quite close to
the fixed site. A short length of Teflon tubing connected the instruments
in the mobile unit to the ambient air inlet point. This arrangement is shown
in Figure 9.
Figure 9. Garrett County Airport, McHenry, Maryland site.
2-20
-------
3.3 Summary of Data Acquisition at Each Site
At the time of the mobile laboratory visits, each of the
fixed sites had been in operation for at least six weeks; thus, several
biweekly calibrations had been performed. Figure 10 summarizes the
times of data acquisition by the fixed stations as well as the dates
of the mobile laboratory/fixed site comparison periods.
Lewisburg, West Virginia, was the first site visited. The Kane,
Coshocton, and McHenry sites were then visited in that order. The visit
to Garrett County was the longest; part of this period was one of unat-
tended operation of the mobile laboratory and fixed site. A second, short
visit was made to Lewisburg to re-examine a hydrocarbon analyzer which
malfunctioned during the first visit. Figures 11, 13, 15, 17a, 17b and
21 in Section 4 show in more detail the events that occurred during the
comparison periods.
2-21
-------
T
ai
43
o
4-1
o
o
0)
1
4J
a,
,
P3 M
OO
CO
CM
cfl
PM
CU
I
:oo
CSI
x™
e
o
4-1
o o
O -H
cn o
o
c_>
>CX>^
1—^rr
60
3 •
43 CO
CO >
•H
C
O
•iH
4-1
Cfl
r4
43
C
o
cn
•H
B
o
cfl
u
CO .
O 43
CO
£**» rJ
0)
CU
•H
PQ
cO
n
d
o
•H
4J
cfl
U
C
o
CO
r4
cu
a,
o
TD
CU
X
•r-t
CD
0)
4->
Cfl
Q
C
O
•H
CO
CO O
rJ ^H
43 M
•H CD
rH PL,
^ C
o
"O CO >,
C -H T3
cfl (H 3
Cfl 4-1
C cx in
o g
O 4-1
4-1
CU
CO
cO -H
rJ X
U 4)
_ rH
Cfl -H
W O
Cfl
H Cfl
O !H
O D
O
rH
CU
60
•H
2-22
-------
4.0 DATA COMPARISON
As was stated in Section 3.1, the pollutants measured for comparison
were 0.,, NMHC, and NO-. Examination of the fixed site data indicated
that, except for brief excursions, the N0~ levels were very low; in fact,
near the limit of detectability of the chemiluminescent analyzer. In the
case of nonmethane hydrocarbons (NMHC), the Bendix 8201 Hydrocarbon
Analyzers at the fixed site suffered from a positive moisture interference
in the total hydrocarbon (THC) measurement which affected the NMHC values.
This interference was discovered, confirmed during the comparison period,
and later corrected. Thus, no data exist for comparison between hydro-
carbon measurements from the mobile laboratory and fixed sites.
For the above reasons, the data comparison and statistical analysis
of the NO and NMHC measurements are not presented. The following sections
will be concerned with comparison of ozone data. Figures showing detailed
calibration and comparison periods at each site and graphical comparisons
of the On concentration during the same periods are presented in Sections
4.1 to 4.5.
On some of the graphical presentations a third plot is drawn
representing the 0,, concentration from the fixed site analyzer as recorded
by the Westinghouse Pulse-0-Matic data acquisition system located at the
fixed site. In most instances these concentrations "track" quite well
with the 0 concentration computed by the Mobile Laboratory's data acqui-
sition system and computer. It should be pointed out that the two systems
collect data in different manners. The Westinghouse system integrates the
signal from the fixed 0^ analyzer for a fifteen minute interval. The
2-23
-------
Hewlett-Packard system aboard the mobile laboratory collects an
instantaneous value every five minutes; the averaging time is essentially
the response time of the analyzer; a fifteen minute value would be the
average of three such values. Because of this difference in data
collection methods, small variations would be expected for fifteen minute
averages, hourly averages, and daily averages.
2-24
-------
4.1 LEWISBURG, WEST VIRGINIA (August 25-31, 1973)
The first site visited by the mobile laboratory during the comparison
study was Lewisburg, West Virginia. Figure 11 is the first of a group of
figures that show chronologically the events that occurred during the
comparison of instrument performance in both the mobile and fixed stations.
In general, the activities at the fixed site are listed on the left; those
of the mobile laboratory on the right of the time column. Within the time
column, calibration periods are indicated by cross-hatching: mobile labora-
tov, right slanted; and fixed station, left slanted.
The warmup period for the mobile site instruments began on the evening
of August 26. Due to problems associated with the line voltage regula-
tion system, the comparison period began somewhat later, on the evening
of August 28.
Both the fixed and mobile site analyzers were calibrated on the after-
noon and evening of August 29. A second comparison period followed,
this time after dynamic multipoint calibration of both the fixed and
mobile analyzers.
The ozone analyzer in the mobile laboratory underwent a zero and
span operation at 0930 on August 30. The transfer equation changed
slightly, so a third comparison period for 0 analyzers began from this
point until all analyzers went offline at 0915 on August 31 and the
comparison period ended.
Figure 12 presents graphically the comparison between the ozone
analyzers at the fixed and mobile sites before and after the calibration
periods mentioned above. Based on hourly averages, the fixed site ozone
analyzer indicated slightly higher ozone concentrations than those from
9_0 C.
-------
2000
(8-28) 2400
0400
0800
Fixed-Site
Instruments On-Line
for Comparison
(8-29) 2400
0400
0800
Fixed Site Instrument
Calibration
1200
(8-30) 2400
Fixed Site Instrument
Calibration
1200
2000
(8-31) 2400
Fixed Site Analyzers ~T
0915 L
Mobile Lab Analyzers On
for 24-Hour Wanaup (1500)
— Warmup - Mobile Lab Analyzers
Mobile Lab Analyzers Calibrated
(2100-2330)
Fixed Site/Mobile Lab Comparison
_Mobile Lab Analyzers Calibrated
(1315-2120)
• Fixed Site/Mobile Lab Comparison
Ozone Analyzer Zero and Span
(0930-1015)
Mobile Lab Analyzers Calibrated
(0925-1535)
-Fixed Site/Mobile Lab Comparison
Mobile Lab Analyzers
-Offline
Figure 11. Fixed Site/Mobile Lab Comparison
August 27 - September 1, 1973
Lewisburg, W. Va.
2-26
-------
s
cd
l-i
bO
O
c-O
(3 03
cd £> **
S-J • O
3 IS on
en
S-) CM
•H w en
,H -H p
cd U 60
3 0) 3
O* i-J
-------
the mobile site ozone analyzer. After calibration of both analyzers, the
ozone concentrations from the fixed site were slightly below those indi-
cated by the mobile site.
The Beckman 6800 Air Quality Chromatograph located in the mobile site
was not operational during the period August 25-31, 1973, thus requiring
a second comparison period for Lewisburg. The abnormally high nonmethane
hydrocarbon values obtained with the Bendix 8201 Hydrocarbon Analyzer
raised questions with regards to accuracy and validity of the data from
the analyzers. An interference was suspected and the manufacturer was
asked to investigate the problem.
2-28
-------
4.2 KANE, PENNSYLVANIA (September 4-8, 1973)
The second site visited during the comparison study was Kane,
Pennsylvania. The warmup period for the mobile site analyzers began
at 1200 on September 4, 1973, and the comparison periods before and
after fixed site calibrations extended from 2300 on September 5, 1973,
to 1200 on September 6, 1973, and from 0100 on September 7, 1973, to
0800 on September 8, 1973, respectively. Final calibration of the mo-
bile site analyzers was completed at 1030. (See Figure 13.)
Figure 14 indicates that the fixed site ozone analyzer measurements
were somewhat less than those indicated by the mobile site ozone analyzer
both before and after calibration.
2-29
-------
Fixed Site 03
Analyzer Equipment
Failure (0800-1200)
(9-4)
1600
2000
(9-5) 2400
0400
0800
1200
1600
2000
(9-6) 2400
0400
0800
1200
1600
Fixed Site Analyzers
Calibrated
03 (1700-2300)
N02 (2300-0100) (9-7)
Fixed Site NMHC
Analyzer Brought
On-Line After Repair
(0100)
Fixed Site NMHC
Analyzer Calibrated
(1100-1500)
2000
2400
0400
0800
Fixed Site Analyzers
Off-Line
(9-8) 2400
0400
0800
Mobile Lab Analyzers On
For 24-Hour Warmup (1200)
Data
Data
Data
Data
Data
) — Fixed Site
Warmup - Mobile Lab Analyzers
Mobile Lab Analyzer
- Calibrated and Brought
On-Line (2050-0005)
-Fixed Site/Mobile Lab Comparison
Mobile Lab N02 Analyzer
•Calibrated and Brought
On-Line (1020-1220)
_Mobile Lab Analyzers
Calibrated (2145-0200)
Fixed Site
Fixed Site/Mobile Lab Comparison
Mobile Lab Analyzers
- Calibrated and Off-Line
(0915-1
Figure 13. Fixed Site/Mobile Lab Comparison
September 4-8, 1973
Kane, Pennsylvania
2-30
-------
s
cd
M
60
O
>-i
P-l 00
cd r-~
OJ -H ON
o c I-H
C cd
cd > •>
!-i iH CC
3 >% I
en en vo
W C
<3 c n
0) 0)
>,Pn ^5
J-> S
•H « 0)
iH 0) 4J
cd C ft
d
CC !
C/3
60
•H
ui/Srt
2-31
-------
4.3 COSHOCTON, OHIO (September 14-19, 1973)
The third site visited during the comparison study was Coshocton,
Ohio. The warmup period for the mobile site analyzers began at 1300
on September 14, 1973 and continued until 1400 on September 15, 1973 at
which time the calibration of the mobile site analyzers began. Because
of a failure of the fixed site ozone analyzer immediately prior to arriv-
al of the mobile van, no comparative data exist between ozone monitors
prior to the fixed site biweekly calibration on September 15, 1973.
Prior to this calibration, the PM tube of the fixed site analyzer was re-
placed and the instrument was allowed to warm up for approximately 24
hours. Thus, the only comparison data available are from 1400 on Sep-
tember 18, 1973 to 1100 on September 19, 1973. (See Figure 15.)
Figure 16 indicates that the hourly averages of the fixed site
ozone analyzer were less than those indicated by the mobile, van. How-
ever, on September 19, 1973 around 0800, the ozone concentrations
"crossed" and the fixed site ozone hourly averages were greater than
those of the mobile van.
2-32
-------
Fixed Site Analyzers
On-Line (0-, inoperative)
(9-16)
Fixed Site 0, Analyzer
Repaired (1600)
Fixed Site Analyzers
Calibrated (1625-2220)
NO,, HC
Fixed Site 0, (9-18)
Analyzer - Preliminary
Calibration
Fixed Site Ozone Analyzer
Calibration (1020-1225)
(9-19) 2400
Fixed Site Analyzers
Off-Line
Mobile Lab Analyzers On
' for 24-Hour Waraup (1300)
- Warmup - Mobile Lab Analyzers
Mobile Lab Analyzers
"Calibrated (1420-1740)
. Mobile Lab Analyzers
Calibrated (1225-1405)
Bek HC, CO
Mobile Lab Analyzer
-Calibrated (1115-2210)
- Fixed Site/Mobile Lab Comparis
Mobile Lab Analyzers
. Calibrated (0900-1300)
and Off-Line
Figure 15. Fixed Site/Mobile Lab Comparison
September 14-19, 1973
Coshocton, Ohio
2-33
-------
i-I
•H
CO
T)
0)
C
0)
0)
•H
O
g
oo
0
0)
0)
CO
O
Csl
3
O
0)
co g
CN -H
H
00
V-l
0)
a)
4-1
CX
-i
txo
o
a\
a)
o «
C O c^
tfl -H r-(
^ ^J I
3 O 00
CO H
CO -
< C l-i
O CO
>> -U 42
4-1 a S
•H o cu
H ,n 4-1
nj co &,
3 o cu
ex u co
60
•H
o
CM
o
o
o
00
o
CN
o
ju/grt
2-34
-------
4.4 GARRETT COUNTY, MARYLAND (September 21-October 3, 1973)
The mobile laboratory arrived at Garrett County, Maryland on
September 21, 1973, for an extended period. Part of this time the mobile
laboratory (and the fixed site) operated unattended. Figures 17a and 17b
present the events for the fourteen day period. Since several calibrations
were performed on the mobile van analyzers, there are several possible
periods for comparison of the fixed site versus mobile site ozone measure-
ments. The hourly averages for these comparison periods are shown in
Figures 18, 19,and 20. Figures 18 and 19 also show the hourly averages
from the Westinghouse Pulse-0-Matic data collection system that was
located at all the fixed sites and analyzed independently of the mobile
laboratory data. The agreement of this data with the fixed site data
(as processed by the computer on-board the mobile laboratory) is excellent,
thus lending credence to the reliability and accuracy of both the Westing-
house Pulse-0-Matic data system and the processing system used to compute
the data from the fixed sites during the rural oxidant study.
2-35
-------
Fixed Site Ozone
Instrument Brought
On-Line (1100)
(9-21) 1200
1600
2000
(9-22) 2400
0800
(9-23) 2400
0800
1200
(9-24) 2400
0400
0800
200 7
Fixed Site Ozone
Analyzers (1500-1900)
Ca1ibration (No
Transfer Funt-tion
Change)
2000
(9-25) 2400
0400
0800
1200 -
1600
2000 -
(9-26) 2400
~ Data
Mobile Lab Analyzers On
For 24-Hour Warmup (1200)
• Warmup - Mobile Lab Analyzers
Mol)l le Lab Analysers
C.i] ibratod and Brought
On-Line (1215-1610)
-Fixed Site/Mobile Lab Comparison
Mobile Lab Analyser Calibration
(Ozone Instruments)
(0925-1025)
-Fixed Si to/Moblie Lnb Comparison
Mobile Lab Analyzer Calibrations
(Hydrocarbons)
(1210-1300)
-Fixed Site/Mobile Lab Comparison
Mobile Lab Calibrarions
(1420-1545)
- Fixed Site
Figure 17a. Fixed Site/Mobile Lab Comparison
September 21-25, 1973
Garrett County, Maryland
2-36
-------
Fixed Site Ozone
Analyzer Biweekly -
Calibration
(1025-1235)
Fixed Site
Analyzer Off-Line
(1400)
(9-26) 2400
0400
0800
1200
1600
(9-27) 2400
(10-1) 0000
0400
0800
1200
1600
2000
(10-2) 2400
0400
0800
1200
1600
2000
(10-3) 2400
0400
0800
1200
1600
Data
Data
•Fixed Site
Mobile Lab Instrument
Calibrations (Hydrocarbons)
(1635-1755)
•Fixed Site/Mobile Lab Comparison
Mobile Lab Calibrations
(Ozone, N02>
(0900-1255)
Fixed Site/Mobile Lab Comparison
Final Mobile Lab Calibrations
(1100-1515)
and Off-Line
Figure 17b. Fixed Site/Mobile Lab Comparison
September 26 - October 3, 1973
Garrett County, Maryland
2-37
-------
c
a o
OJ -H
> *-"
n)
a) h
•r-l
CO
•a
JS
eo
aj c
I I I
OJ
•i
ex
a)
CM c/}
CM Vj
3
a
-i CN
cfl
O ' ro
C •> CNJ
cti ^
(-1 4J «
3 C eg
en 3 eg
w o
<1 O Vj
0)
s-< a.
cfl QJ
OO
3
60
•H
o o
CM O
o
OO
o
o
o
CO
ui/3rt
-------
0)
I-l >-l
•H 0)
C
>
0)
X 43
•H O
"O T3
VJ
tfl
i
0)
OJ
33
O
tN
cxl
O
O
cxl
O
OO
o
cxl
O
o
o
OO
O
4-J
QJ
0)
U3
o
60
C
CO
(U
I I I
o
Cxl
cxi a)
4J
ex
a)
o to
Cx|
CN
in
CXI
O (-1
rH <])
CO 4-1
fX
0) ,-v
co en
(1)
•H
H
Cxl
Cxl
O
CXI
o
t-i C i-H
60 n)
O rH •
P-i ^J cxl
tO t^
11 | i ,
3 C , 4-1 ,£2
4-J 4-1 E
•r( 0) OJ
a)
M
3
bO
-ra/8rt <
2-39
-------
•H
t/1
•8
•H
,0
CO
M
,0
n
u
ja
. rt
\ ^
H- T
»
/
f
'
QJ
H
,0
. o
\ s
^
\ •"
\ 0
cl
O
•H
4-*
\ n)
/ ^
/ -°
/ 'rH
rH
n)
U
-
-
X— >
o
o
CM
rH
1
o
o
oo
o
*"
_
,H
CM
rH CM
M
, ON
PL, n i-(
O
C "
M 4-1
3 Cl
CO 3
co o
>, 4J t-l
4J 4-1 a)
•H a) &
iH ^ O
cfl M 4J
3 cfl O
O1 O O
o
CSJ
60
•H
o
o
CM
o
00
o
vo
o
CM
O
O
o
00
o
o
CN
m/8rt «
2-40
-------
4.5 LEWISBURG, WEST VIRGINIA (October 4-8, 1973)
A second comparison period was initiated at Lewisburg, West
Virginia during the period October 4-8, 1973, mainly to obtain addi-
tional hydrocarbon data. The mobile and fixed analyzers were calibrated
at about the same time on October 5; thus, the comparison period for
ozone is restricted to data after calibration. The study ended on
October 8, 1973. (See Figure 21.)
Figure 22 shows that excellent agreement was obtained between
the mobile and fixed ozone analyzers for the two-day period.
Comparison of total hydrocarbon and nonmethane hydrocarbon mea-
surements from the Bendix 8201 and the Beckman 6800 analyzers indicated
an elevated response of the Bendix analyzer to ambient concentrations
of total hydrocarbons. Moisture was suspected as the interferent and
subsequent tests at Bendix confirmed the suspicion. This cpnclusion and
the confirming evidence uncovered at Lewisburg and other sites were re-
sponsible for invalidating all hydrocarbon data collected at the four
sites during the rural oxidant study.
2-41
-------
(10-4) 1200
1600 -
2000
(10-5) 2400
0400
0800
Fixed Site Analyzers
Calibrated (0845-1100)
Fixed Site Analyzers
On-Line (1540) 1600"
2000
(10-6) 2400
0400
0800
1200
1600
2000
(10-7) 2400
0400
0800
1200
1600
2000
(10-8) 2400
0400
0800
1200
- Data
1600
Fixed Site Analyzers
Off-Line (1800)
'S//////77S-.
- Data
Data
'///////TZ
Mobile Lab Analyzers On
for 24-Hour Warmup (1400)
Warmup - Mobile Lab Analyzers
Mobile Lab Ana]yzers Calibrated
and Brought On-Line (1110-1615)
Fixed Site/Mobile Lab Comparison
Mobile Lab Analyzers Calibrated
(1055-1455)
Fixed Site/Mobile Lab Comparison
Mobile Lab Analyzers Calibrated
(0950-1050)
Fixed Site
Mobile Lab Analyzers Off-Line
Comparison Ends (1800)
Figure 21. Fixed Site/Mobile Lab Comparison
October 4-8, 1973
Lewisburg, W. Va.
2-42
-------
T3
0)
X
01
4-1
•H
X>
O
O
CN
cs
o
o
O
00
o
-------
5.0 STATISTICAL ANALYSIS
Selected data obtained during the quality assurance program at
each site (Appendix B) were analyzed to determine the comparability
and relationships between the fixed and mobile site ozone measurements.
The results of these analyses are summarized in Table 4 and include the
correlation coefficient, mean, range, standard deviation about the mean,
ratio of fixed to mobile means, estimated bias, percent relative bias, and
the number of observation periods both prior to and after the calibration
of the fixed and/or mobile site analyzers.
In general, there were two comparison periods as shown in Figure 5.
The first comparison period was for approximately 16 hours and occurred
after dynamic calibration of the mobile site ozone analyzer and before
the regular biweekly multipoint calibration of the fixed site ozone
analyzer. The second period of comparison was approximately 24 hours
and occurred after the fixed site calibration. In some cas.es, there were
other comparison periods following additional calibrations of the mobile
site analyzer.
Code numerals were assigned to the mobile (M) and fixed (F) site
ozone analyzers in Table 4 to denote different comparison periods and are
as follows:
M-l Comparison period following first calibration of mobile site
analyzer.
M-2 Comparison period following second calibration of mobile site
analyzer.
M-3 Comparison period following third calibration of mobile site
analyzer.
F-l Comparison period preceding calibration of fixed site analyzer.
F-2 Comparison period following calibration of fixed site analyzer.
2-44
-------
W
W H
o w
O P
CO
O
M
t/>
H
Q W
W hJ
Pd M
M PQ
sss
CO
H
2
y
< M <*
rJ W ft
§ &
fa S
o --.
o
M CO
<; <
p3 ^
a
•K
H <
CO M
W pq
*
• O
>r_-
rX-
W X
Q I-H
fa
S3 W
O >-:
-*
,_4
H "
pej O
S
Q
rj £3
O M
co O fa
S °
. r"*! M
PM W hJ
§ Pu M
O M
0 S
H
M
CO
cn CM CM
m sr ON
en ON o
CM rH rH
1 1
Cn rH O
CM 00 ON
rH O O
ON rH in
r— oo rH
rH 1 rH
+ 1
O en T-H
ON rH CM
en rH vo
-a- oo oo
rH o -a-
cn rH VO
r- -a-
o cn cn
CM VO CM
1 I 1
VD ON r-H
en rH en
m ON
vo ON en
rH VD CM
1 1 1
r-» oo r^
CM CM en
CM r-. CM
vo cn VD
ON en o
rH
cn co VD
OO rH r--
r- -a- rH
rH
rH ON vO
r-- VD rH
00 ON ON
ON ON ON
o o o
rH CM CM
rH CM en
»
rH
en
00 00
rl CM
• 3
pO 4-1
co eg
•H 3 en
& 00 1 —
cu 3 ON
rJ < rH
VO ON
VO CM
rH
O VD
rH rH
1 1
o -a-
ON 00
O 0
m o
en O
rH CM
1 i
oo m
00 vo
rH CM
m CM
-O
fj S
jS 4-1 en
co o. r^
O OJ ON
O CO rH
H en
CU
4-1 A M
4J S CU
CU S J3
rl 4J O
rl O, 4J
CO CU O
O CO O
rH
oo
CM
'
\
vO
ON
d
CM
en
1
00
en
cn
rH
r~l
en
r^
in
rH
1
en
CM
rH
in
rH
1
CM
CM
VD
rH
r--
00
-a-
en
-a-
ON
ON
o
CM
CM
«,
co
00 vO
rl
3 rl
n CD
CO f>
•rl o en
S 4-1 I--
CU 0 ON
hJ O rH
o
o
cn
g
00
3.
II
CO
j-i
•rl
a
F-)
•K
!Z
"oT
rH
•H
O
£
1
•a
cu
X
•rl
fa
**-s
II
ca
CO
•H
_O
•a
CD
en
S
•H
1 j
CO
W
CO
•a
o
iH
rl
CD
a.
§
•rl
4J
CD
4-1
3
a
•H
e
m
H
UH
O
•
o
c
II
2
CO
CO
CD
a
a)
wC
4J
<4H
O
O
•rl
4J
eO
Crf
II
jj*
-»^
fa
O
•H
4J
m
K
C
rt
CU
*
T3
CD
.a
pn
x^.
1
^
4.
t
C
C
I
fl
(.
^,
c
c
*r
4
a
i
•i
i
c
r-
i
0
c
cd
CD
S
CD
rH
•H
o
S
^^
-*
J
u
J
4
U
X
S
0
d
•i
3
j
>
•i
j
9
H
U
i
2-45
-------
Each observation period used in the comparison consisted of a pair of
fifteen minute ozone averages, one each from the fixed and mobile sites.
Four comparison periods were available for Garrett County, three for
Lewisburg, and two for Kane. Due to the failure of the fixed ozone
analyzer at Coshocton, immediately prior to the arrival of the mobile
laboratory, the only comparison period for that site was after repair and
calibration of the fixed site analyzer and calibration of the mobile site
analyzer.
Brief summations of the comparison results between measurements of
ozone from the fixed and the mobile sites are presented in the following
paragraphs. Each summary pertains to the time frame indicated in Table 4
for each site. Certain qualifications should be restated prior to presen-
tation and interpretation of the comparison results. These are as follows:
1) Two independent monitoring systems sampling from a common
sampling point were used to generate each pair of da'ta
points.
2) Identical brand-name ozone analyzers were used where
possible.
3) Independent test atmosphere generating systems were used
to provide calibration concentrations to the respective
analyzers in the fixed and mobile sites.
4) Each ozone calibration concentration was verified by
manual iodometric analysis.
5) Individual differences were minimized and continuity was
maintained throughout the quality assurance program by
using qualified personnel to calibrate the ozone analyzers
in the respective stations.
2-46
-------
A. Lewisburg, West Virginia (August 28-31, 1973)
The data presented in Table 4 show excellent correlation between the
fixed and mobile site ozone measurements both prior to and after the fixed
site calibration. Correlation coefficients ranged from 0.98 to >0.99.
The ratio of the fixed to mobile mean values, however, was 23 percent
higher prior to the calibration of the fixed site analyzer. A possible
3
reason for this estimated bias (18 yg/m ) could be span drift of the fixed
site analyzer since the previous biweekly calibration. Upon calibration
of the fixed site analyzer, there was a reversal in the mean ratio; however,
there was also a marked reduction in the range of the ozone measurements.
Explanations for the change in bias immediately after calibration of the
fixed site analyzer could be related to differences in calibration systems
and differences in techniques of individuals performing the manual iodo-
metric analyses during calibration. Also, analyzers in the mobile
laboratory operated on regulated line voltage, whereas the fixed site was
not equipped with a line voltage regulator. After a second calibration of
the mobile site analyzer, the ratio of the means was 0.90.
B. Kane, Pennsylvania (September 6-8, 1973)
Excellent correlation was obtained at Kane between the fixed and
mobile site ozone measurements both prior to and after the calibration
period. On the average, the mobile site analyzer indicated 10 to 16
percent higher ozone values than the fixed site analyzer with biases of
3
-13 to -20 pg/m for the respective periods.
2-47
-------
C. Joshocto.i, ,)hdo ('-' v'em-e; 18--J.1', 19,3;
'Ihe correlarlon L^> .. te. ; '•"or the oi\r -o upar (son period at
Cosboi- Lot. was 0.'^7 ,~i J tn - r.t o o" the r.eaj A'-'" 0.^7. An averag
3
bias of -10 PT 'n wa« xMoLlf' between th^ !.-'o •' K ""-zers.
e
U. Garret t Co1 nty, Mar/, iu > .^eptenibrtr 21 -- ,ir'.-,ber 3, 1973)
The data presente-. -•' >\ TabU 4 slow PXCC! 1 ?. i! . ocraldtion between
("he fixed and mobile si*"-1 ..iial.'^ers with tae Correlation coefficient
ranging from 0.97 to O.'^B,, The ratio of means aid not change signifi-
cantly after calibration of either the fixed cr mobile site analyzers,
and the average bias for the entire period was less than 5 percent.
E. Lewisburg, West Virginia (October 6-8, 1973)
The return visit to Lewisburg coincided with the regular biweekly
calibration of the fixed station analyzers. Excellent correlation was
rbtained between the two measurements with the mobile site analyzer
raiding approximately 4 percent higher on the average. The difference
between the fixed and mobile site analyzer measurements was much less
than during the first comparison period (see A).
2-48
-------
6.0 SUMMARY AND CONCLUSIONS
The purpose of this program was to provide some measure of the
comparability of the data being generated at four fixed stations during the
rural oxidant program conducted by RTI during the summer of 1973. The
objective was accomplished by comparison of measurements obtained by instru-
mentation housed in a mobile laboratory with those obtained from the four
fixed stations under routine operating conditions, both prior to and after
calibration of the fixed site analyzers. The RTI Environmental Monitoring
Laboratory, equipped with a full complement of air quality analyzers,
calibration systems, data acquisition system, and on-line computer capable
of processing data in real-time, was transported to each of the respective
sites and was set up adjacent to the fixed station; comparative data were
obtained for the measurement parameters from both the fixed and mobile
site analyzers. The duration of the comparison period varied somewhat due
to the operational status of the analyzers in the respective stations. In
general, the time frame presented in Figure 5 (Section 3.1) was adhered to.
Air sampling inlets for both stations were located at a common point, so
that identical samples of ambient air were provided to both sets of
analyzers. Data were processed in real time for both sets of analyzers
3
(i.e., fixed and mobile), so that instrument outputs in yg/m could be
examined immediately. This feature was necessary for preliminary comparison
of data and check on the monitoring system. Following the comparison
periods at each site, the Environmental Monitoring Laboratory was returned
to Research Triangle Institute and the data subjected to statistical
analysis.
2-49
-------
Due to the constraints previously mentioned regarding extremely low
NO- values and the invalidation of all hydrocarbon data from the fixed
sites, the only pollutant considered for the comparison evaluation was
ozone. Statistical analysis included computation of correlation coeffi-
cients, means, ranges, standard deviations about the mean, ratio of fixed
to mobile means, and estimated bias during each comparison period, both
prior to and after calibration of the fixed and/or mobile site analyzers.
Excellent correlation was obtained at each site between the fixed
and mobile site ozone measurements both prior to and after each calibration
period. Correlation coefficients ranged from 0.97 to >0.99 within a site
and comparison period. Level differences were observed between measure-
ments from the fixed and mobile site analyzers at each site as indicated
by the estimated bias of the measurements. The average estimated bias
2
ranged from 18 to -20 yg/m over the entire comparability program (i.e.,
over the 11 comparison periods at the four stations). The greatest
relative bias existed at Lewisburg (23%) prior to the calibration of the
fixed site analyzer, and the least relative bias was at Garrett County (8%).
After calibration the relative bias at Lewisburg decreased considerably,
indicating that the bias in the measurement was related to span drift of
the fixed site analyzer since the previous calibration, to a faulty calibra-
tion, to variation in the manual iodometric analyses used to determine the
ozone calibration concentrations and/or to a combination of all three. With
respect to the Kane and Coshocton sites and the Lewisburg site after cali-
bration, a consistent negative relative bias of 10 to 16 percent was
observed. The positive relative bias at Garrett County and Lewisburg
2-50
-------
(2nd visit) between the fixed and mobile site measurements averaged
approximately 3 percent. In the above discussion, the mobile site data
were used as a measure of the absolute concentration of ozone. It is
recognized that this assumption is not entirely correct; however, these
data should be more precise and accurate than the fixed site data because
they were essentially obtained under controlled laboratory conditions.
Pertinent observations that can be drawn from the data are as follows:
(1) The high correlation between the fixed and mobile site
ozone measurements (i.e., 0.97 to >0.99) confirms that
the ozone analyzers at each site were measuring the
same variable—but not necessarily generating the same
numbers.
(2) Level differences (bias) did exist at each site between
the fixed and mobile site ozone measurements. The esti-
mated bias over all sites and comparison periods ranged
from 18 to -20 yg/m3 [(E Fixed - Mobile)/N].
(3) The average relative bias over all sites and for all
comparison periods was less than 10 percent. Excursions
from this limit did occur; this information was used to
identify problem areas and take corrective action.
(4) Bias between the fixed and mobile site ozone measurements
generally decreased immediately after calibration and/or
recalibration of the fixed or mobile site analyzers.
Bias between the fixed and mobile site measurements could be due to many
factors, such as instrument performance (drift, response, etc.), operator
variables, calibration techniques and frequency, environmental factors
(temperature and line voltage variation), and concentration level of
ozone.
The overriding consideration for this program was to provide a means
for determining the comparability and interrelatability of data obtained
from four widely dispersed fixed air monitoring systems. This was
2-51
-------
accomplished by systematic comparison to data from the mobile air monitoring
facility, which was considered to be more reliable due to better environ-
mental control, sophistication of equipment (i.e., calibration and data
acquisition systems, data processing facilities) and experience of personnel.
This approach is novel in that it provides a mechanism for validation of the
operation of the entire monitoring facility, not just the analyzer itself.
This includes the air sampling system, the performance of the air quality
analyzer, the data recording system, and the system utilized to process and
output the data.
Recommendations for future work involving field measurement programs
include the use of a mobile air monitoring facility as a quality assurance
tool to verify and validate data from network air monitoring stations.
Proper use of a mobile facility to determine the bias of individual stations
would allow for comparison between pollutant levels in different parts of the
country with better confidence. Increasing the frequency of the comparison
periods for field sites to the mobile air monitoring facility would give an
estimate of the precision of the estimated bias.
2-52
-------
APPENDIX A: CALIBRATION SYSTEMS/PROCEDURES
2-53
-------
APPENDIX A
CALIBRATION SYSTEMS/PROCEDURES
1.0 CALIBRATION SYSTEMS
1.1 Ozone
A dynamic calibration system as described by Hodgeson et al. and
published with the National Primary and Secondary Ambient Air Quality
2
Standards was used to calibrate the ozone instrumentation evaluated
during this study. Briefly, the ozone source consists of an 8-inch
ultraviolet mercury lamp which irradiates a 5/8-inch quartz tube through
which clean (compressed) air flows at 5 liter/minute. Ozone concentra-
tions from 0 to approximately 1 ppm (1960 yg/m ) can be generated by
moving the shield and exposing various lengths of the lamp. Although
1>
the UV 0« generator has been shown to be quite stable and reproducible,
the neutral-buffered potassium iodide technique was used as the
2
reference method. A permanent calibration setup consisting of a zero
air source, calibrated rotameter, UV generator, and a glass manifold
was installed in the laboratory facility and calibrated by the manual
neutral-buffered potassium iodide procedure periodically during the
study. A diagram of the calibration system is shown in Figure A-l.
An identical but portable system was used to calibrate the ozone analy-
zers in the fixed stations during this study. All calibration concen-
trations were verified by manual iodometric analysis.
1.2 Nitric Oxide/Nitrogen Dioxide
Due to problems associated with long-term use of N0_ permeation
tubes and the need to routinely determine the efficiency of the carbon,
stainless steel, or molybdenum converters (which reduce NO^ to NO), an
alternative procedure (gas phase titration) was used for routine dynamic
calibration of the chemiluminescent NO-NO -NO analyzers. The gas phase
X Z,
titration technique is based upon application of the rapid gas phase re-
4 5
action between NO and 0_ to produce a stoichiometric quantity of NO- '
NO + 03 »- N02 + 02
N0 >- N0 + hv .
2-55
-------
e
0)
(3
O
•H
M
Xi
•H
Ct)
O
S-i
-------
Nitric oxide from a cylinder of NO in N« (50 ppm) is diluted with a
constant flow of clean air to provide 1.0 ppm and used to calibrate the NO
and NO cycles of the chemiluminescent NO-NO -NO analyzer. By incor-
X X i£
poration of a calibrated ozone generator in the calibration apparatus
upstream from the point of NO addition, precise NO concentrations can be
generated by oxidation of NO to NO- with 0 . A schematic diagram showing
the component parts of the calibration system is presented in Figure A-2.
As long as a slight excess of NO is present, the concentration of 0_ added
is equivalent to the concentration of NO consumed and is equivalent to the
concentration of NO generated.
A general description of this calibration scheme is presented in the
following paragraphs. Primary calibration of the NO concentration in the
pressurized cylinder containing nitrogen as a diluent is accomplished by
titration of an NO concentration of 1.0 ppm produced by dilution with
successive concentrations of ozone (0-0.8 ppm) generated by an ozone
generator which has been referenced to the neutral-buffered KI procedure.
The resultant NO detector outputs after stabilization at each titration
point (i.e., 0.0,0.1,0.2, 0.8 ppm ozone added) are plotted in ppm on
coordinate graph paper (y-axis) versus 0_ concentrations added, ppm (x-axis)
A straight line is drawn through the linear portion of the iteration curve
and extrapolated to the x-axis. The concentration at the x-axis intercept,
C', is the 0 concentration equivalent to the initial diluted NO concen-
tration. An example of a typical gas phase titration curve is presented in
Figure A-3. The concentration of NO in the cylinder can then be determined
as follows:
c
N° FNO
where
C = cylinder NO concentration, ppm,
F = measured NO flow, cc/min,
C' = equivalence point 0_ concentration, ppm,
Fn = total clean air flow, cc/min.
2-57
-------
CALIBRATION SYSTEM
ADJUSTABLE SLfLV[
MASS FLOWMCTLR
(0-50 cm /mm)
\A_ ^v^f^w^
MIXING
BULB_ MANIFOLD
i!=^"~ §=C "~>- VENT i
v_--__^— ^fH-lrf-vr '
Figure A-2. Gas phase titration system
2-58
-------
1.0
I I
EQUIVALENCE
'POINT
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
0- Concentration (ppm) (0~ Generator)
Figure A-3. Gas-phase titration of NO with 0 .
2-59
-------
Once the NO concentration in the cylinder has been determined, this cylinder
can be used over its lifetime to provide a working standard for routine
calibration ; however, to assure validity of data, the NO concentration
should be verified at one-month intervals.
During routine calibration, the NO and NO channels of the chemilumi-
X
nescent NO-NO -NO analyzers were calibrated by dynamic flow dilution of
X ^
the NO in nitrogen cylinder gas. To calibrate the NO output channel and
to determine the converter efficiency (i.e., efficiency of reduction of
N0~ to NO), a constant concentration of 0.5 or 1.0 ppm of NO is produced
in the flow system. Ozone is then added in increments from the variable
0 source. The incremental decrease of the NO measurement is then
equivalent to the concentration of NO produced by the gas phase titration
reaction. In this scheme the calibrated 0_ source becomes a calibrated
NO source when NO is present in excess.
6.1.3. Hydrocarbons
Calibration of hydrocarbon instruments was accomplished utilizing
standard calibration gases certified by the supplier. For this study
cylinders of methane in air were obtained from Scott Research Labora-
tories. Several concentrations ranging from 0 to 10 ppm of CH, were used
for calibration purposes. In the absence of acceptable cylinders of zero
air (i.e., CH, < 0.01 ppm) an alternative procedure utilizing electronic
zeroing was used in lieu of dynamic zero. The following concentrations
of methane in air were used to calibrate hydrocarbon analyzers:
Cylinder Certification by
Supplier # Manufacturer (- 2%)
Scott Research Laboratories A , „
1.2 ppm
B 4.2 ppm
C 7.5 ppm
2-60
-------
2.0 GENERAL CALIBRATION PROCEDURE
A general procedure applicable to the dynamic calibration of any
analyzer was utilized in this investigation. Procedures and features
789
developed during the previous studies ' ' such as mode switch inputs
describing instrument operational status and automatic entry of calibra-
tion data on magnetic tape were combined with on-line computer computa-
tion of transfer equations by linear regression analysis of calibration
data and processing of air quality data in quasi real-time.
The basic step-wise procedure employed for dynamic calibration of
all air quality analyzers was as follows:
(1) Verify operational status of each analyzer prior to
beginning calibration.
(2) Connect instrument inlet line or instrument calibration
inlet line, as the case may be, to the manifold of the
calibration apparatus or for hydrocarbon instruments, di-
rectly to cylinders containing calibration gas.
(3) Allow instrument to sample zero air (i.e., air minus the
pollutant of concern) for a period of time sufficient to
establish a valid zero output. Indicate the proper
manual entry status code for zero and average the instrument
output for zero input concentration for at least
15 minutes.
(4) Introduce a pollutant calibration concentration equal to
approximately 80 percent of the operating range and adjust the
span of instrument as required upon initial setup of the
instrument. This adjustment is normally required only
upon initial setup of an instrument or if excessive span
drift occurred during the evaluation period. Minor
adjustments in the span of each instrument can be performed
by the on-line computer more easily than by manual adjustment
of the span control knob, except for cases where drastic
changes in span occurred. Omit Step (4) except on initial
setup of analyzer.
2-61
-------
(5) Introduce successive pollutant calibration concentrations
of 10, 20, 40, 60, and 80 percent of the operating range of
the instrument being calibrated. Allow sufficient time to
establish a valid instrument output for each calibration .
concentration, and average the instrument output for that
input calibration concentration for at least 15 minutes.
Indicate the proper manual entry status codes for multi-
point calibration and proceed to the next higher calibration
concentration and repeat the sequence of events for multi-
point calibration.
(5) Return the instrument inlet line to the ambient air sampling
manifold and compute the transfer equation, which relates
pollutant concentration input to instrument voltage output,
for each instrument. This function was automatically
computed at the end of each calibration by the on-line
mini-computer employed during this instrument evaluation
program.
The frequency of calibration performed during this investigation
varied from daily (mobile site) to biweekly (fixed sites). All calibra-
tion concentrations were verified by manual iodometric analyses (neutral-
buffered KI procedure).
2-62
-------
3.0 DETAILED CALIBRATION PROCEDURES
3.1 CALIBRATION PROCEDURE FOR BENDIX OZONE ANALYZER
1) Set up ozone analyzer according to manufacturer's instructions.
2) Allow sufficient time for warmup stabilization period.
3) Introduce zero air and establish zero baseline and record mV outputs.
4) Introduce an ozone concentration equal to approximately 80 percent
of the operating range. Adjust span control as required and
record value.
5) Introduce ozone concentrations of 10, 20, 40, 60 percent of the
operating range and record respective mV outputs without additional
adjustments. Determine the ozone concentration for each calibration
3
point in yg/m using the neutral-buffered KI method as described in
the April 30, 1971 Edition of the Federal Register.
6) Establish a line of best fit using method of least squares for the
calibration curve.
2-63
-------
3.2 CALIBRATION PROCEDURE FOR BENDIX NO-NO -NO. ANALYZER
x 2
1. Turn on ozone generator and Hastings Mass Flowmeter and allow them
to warm up for approximately 15-20 minutes.
2. Connect instrument inlet line to manifold of calibration apparatus.
Connect compressed air cylinder to calibration apparatus and allow
zero air (air filtered through charcoal) to flow through system.
Connect NO/N? gas regulator to cylinder and evacuate with pump to
prevent the formation of N0« in the regulator.
3. Allow analyzer to sample zero air for 5-10 minutes. Average zero
output for NO, NO , and NO. channels for at least 5 cycles (i.e.,
X /t
5 minutes). If instrument zero deviates from zero by more than
+10 millivolts, readjust to zero. Indicate proper entry codes for
zero and/or zero adjust on magnetic tape.
4. Prepare 0.1 ppm NO concentration by appropriate dilution- by metering
sufficient NO (5 cc/min) in nitrogen into 5 liter/min diluent stream.
Allow 5 minutes for system to equilibrate and then average first
calibration point until instrument output has equilibrated. Enter
proper codes on magnetic tape using mode switches. Enter calibration
concentration on magnetic tape using 10-turn potentiometer. Both
NO and NO channels are calibrated simultaneously.
x
5. Proceed to next calibration concentration and repeat step #4 (i.e., for
0.1, 0.2,0.4,0.5 ppm NO, etc.).
2-64
-------
6. After completing multipoint calibration with NO, generate NO
concentrations as described in the previous discussion. Enter
proper codes and calibration concentrations on magnetic tape
using mode switches and 10-turn potentiometer. Repeat for
additional calibration points for NO channel (i.e., 0.1, 0.2,
0.4, 0.5 ppm N0~, etc.).
7. Return sample inlet line to sample manifold.
2-65
-------
3.3 CALIBRATION PROCEDURE FOR BENDIX MODEL 8201 AMBIENT HYDROCARBON
ANALYZER
ZERO
SPAN
1. Place valves #1 and #2 to OFF position and meter/output switch
to electrometer position.
2. Observe baseline while component gates are open (i.e., CH, light
on and THC light on).
3. Adjust zero control for THC and CH, to zero by observing recorder
output or meter while switched to appropriate output.
4. After one complete cycle adjust nonmethane zero control for
zero indication on recorder or meter.
5. Return valves #1 and #2 to auto position.
6. Connect zero-air cylinder for dynamic zero utilizing calibrated
capillary to appropriate flow rate of 200 cc/min. Switch oper-
ating mode switch to calibrate.
7. Compare readings with previous zero data and record. Any
discrepancies indicate contaminated plumbing, regulator and/or
cylinder of gas.
1. Connect cylinder containing largest concentration to calibration
inlet with correct flow rate of 200 cc/min.
Caution; Avoid interchanging regulators and cylinders to prevent
cross contamination.
2. After several cycles (at least two) set the span control for THC
and CH to the appropriate reading.
3. After completing span operation on THC and CH, , switch output to
electrometer position.
2-66
-------
4. Observe strip chart for one cycle. Manually override valve #1
to the OFF position for the duration shown on the strip chart
example. This will cause CH, output to be zero and cause the
nonmethane output (THC-CH ) to equal the THC output.
5. Adjust the nonmethane span control to the appropriate reading
with the output switched to nonmethane.
6. Run additional calibration cylinders and record readings but
do not adjust span controls.
7. Recheck nonmethane calibration as outlined above for each
calibration level.
-------
REFERENCES
1. Hodgeson, J. A., R. K. Stevens, and B. E. Martin. A Stable Ozone
Source Applicable as a Secondary Standard for Calibration of
Atmospheric Monitors. In: Air Quality Instrumentation, Scales, J.
(ed.). Instrument Society of America, 1972. p. 114-128.
2. Federal Register. National Primary and Secondary Ambient Air Quality
Standards. Environmental Protection Agency. 36: 8186-8201, April 1971.
3. Ballard, L. F., J. B. Tommerdahl, C. E. Decker, T. M. Royal, and D. R.
Nifong. Field Evaluation of New Air Pollution Monitoring Systems:
The Los Angeles Study. Interim Report. Research Triangle Institute,
Contract No. CPA 70-101, National Air Pollution Control Administration,
1971.
4. Fontijn, A., A. J. Sabadell, and R. J. Ronco. Homogeneous Chemilumi-
nescent Measurement of Nitric Oxide with Ozone. Anal Chem. 42: 575-
579, May 1970.
5. Hodgeson, J. A., K. A. Rehme, B. E. Martin, and R. K. Stevens.
Measurement for Atmospheric Oxides of Nitrogen and Ammonia by
Chemiluminescence. Presented at 65th Annual Meeting of, Air Pollution
Control Association, June 1972.
6. Federal Register. Ambient Air Quality Standards: Reference Method for
Determination of Nitrogen Dioxide. Environmental Protection Agency.
38: 15174-15183, June 1973.
7. Ballard, L. F., J. B. Tommerdahl, C. E. Decker, T. M. Royal, and L. K.
Matus. Field Evaluation of New Air Pollution Monitoring Systems:
St. Louis Study, Phase I. Interim Report. Research Triangle Institute,
Contract No. CPA 70-101, Environmental Protection Agency, 1971.
8. Decker, C. E., T. M. Royal, J. B. Tommerdahl, and L. K. Matus. Field
Evaluation of New Air Pollution Monitoring Systems: St. Louis Study,
Phase II. Interim Report. Research Triangle Institute, Contract
CPA 70-101, Environmental Protection Agency, 1971.
9. Decker, C. E., T. M. Royal, and J. B. Tommerdahl. Field Evaluation of
New Air Pollution Monitoring Systems. Final Report. Research Triangle
Institute, Contract CPA 70-101, Environmental Protection Agency, 1971.
2-68
-------
APPENDIX B
DATA TABULATION
2-69
-------
APPENDIX B
DATA TABULATION
Selected data collected at each site during the comparison
period are summarized in tabular form in this section. These data
generally correspond to the time frame when comparisons were made
between the mobile and fixed site ozone analyzers at each of the four
sites. Two data comparison periods are available from the Lewisburg
and Garrett County sites and are included. Additional data are avail-
able from the mobile van at each site during the comparison period
but were omitted when valid comparisons could not be made between
ozone measurements from the fixed and mobile site analyzers. As pre-
viously stated, direct comparisons between nitrogen dioxide and hydro-
carbon data from the fixed and mobile sites were not computed due to
extremely low values obtained for nitrogen oxides at all sites and
deficiency in the design of the Bendix 8201 hydrocarbon analyzer which
invalidated all hydrocarbon measurements from the fixed stations.
2
The data in the following tables are hourly averages in yg/m .
The instruments in the mobile unit and fixed sites are indicated by
(M) and (F), respectively.
2-71
-------
XXXXXXXXXXXXXXXXXXXXXXXX
H
SH
g
H
CO
W
§
N
O
SJ
S5
8
M
O
s
M
H
CO
3
M
H
PG
2
0
o
erf
EM
O
m
P-*
cy*
.-I
CO
CTi
CN
H
CO
g
§!
^5
1— 1
a
M
g
M
>
H
CO
o"
Ed
g
CO
1— 1
I-I
>-.
I
3
CO
cd
cd
Q
cu
60
cd
M
1
VJ
3
O
K
M
cu
4J
§
O
•H
•§
CJ
CU
a
1
14
M
0
CJ
G
•H
O
4J
cd
1,
H
4J
C
Qj
O
c
o
a
t-
£2
o
<
o
r-H
c/>
53
O
i
o
o
s
§
o
H
H
2!
O
hH
H
I-I
23
tj
Q
M
X
O
g
Fd
^H
J
g
^
25
3
CJ
pq
Q
i—]
X
0
M
a
Q
H
X
O
Cd
O
N
O
i-3
^
25
^
1
O
W
pq
w
1
2:
o
w
0 rHi— ItHi— IrHiHrHrHiHrHCNCNCNCNCN
5
01
•H
£>
cd
•-t
*rH
cd
cd
4-J
O
C
«(
Q
ti
o
•H
4>)
cd
N
1-1
J
•H
•a
4J
co
d
•H
p
CU
N
(H
cd
<
X
0)
3
O
O
IH
O
CU
•H
cd
O
4-J
d)
3
T3
CO
4->
M
cd
J3
CJ
a.
Tj
1 i
CO
e
0
i!
$
c
*Tj O
cu
4-1 CU
cd r-i
"^ pQ
•H C cd
rH 0 iH
cd t-l -H
> 4J cd
C cd >
•H l-i cd
CTJ »iH Cd
4J iH JJ
cd cd cd
QUO
* o
-------
O r-j
u Q
xxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxx
xxxxxxxxxxxxxxxxxxxxxxxx
CM rH rH CN
CMCNCMCNrHrHrHrHrHiHcNoncNCNi-l'-l
R
HrHCOCOrHrHrHrH
1-trHCNrHiHrHrHrHCMtM'H
O
CO
OJ
1
•a
rH
•H
cd
>
cd
4-1
O
c
M
O
g
•H
4->
cd
N
•H
,__(
-H
n *rj
cd 0)
4J 4-1
(D Cd
« 2 g
•rl ^1 U
Cd -rl
N -H M
Cd 4J rH
a cd cd
•Sou
3
tx
B
o
o
14-1
0
(U
3
1
•H
cd
m
o
4-1
(U
3
T3
Cfl
1-J
cd
o
ex
•H
}-l
4-1
CO
s
o
M
4-1
X
a
o
0)
rH
rH j-J
•H -H
cd ca
cd a)
rH
nj -H
4J XI
i— (iHiHrHi— (i-Hr-liHrHrHCNCN|CNCNCM
X * O < rH
2-73
-------
m
ooooooooo
xxxxxxxxx
xxxxxxxxx
xxxxxxxxx
OOt-tCMCNIOOCNfM
01
rH
•8
•H
•H
«
cd
O
C
M
O
d
o
•H
4-1
cd
N
•H
43 13
CO 01
4-1 4J
co cd
T3
.s
rl
0)
4-1
3
CX,
6
O
O
4-1
O
cd -H -H -H
M > U [fl CO
d) C 03 >
N -H l-l Cd
-------
H
O
H
xxxxxxxxxxxxxxxxxxxxxxxx
1
H
W
1
O
GIONAL
S
§
.1. EASTE
H
pi
O
§
w
o
M
C/3
^
QUALITY
ro
rH
oo
I
w
nn
(SEPTETfl
2
3
PENNSYLV^
^
w
$
s
HM
x
n)
w
tfl
4J
rt
Q
(1)
00
to
>
X
^t
M
O
a
r cubic meter)
HYDROCARBONS1
0)
o.
o grams
o
•H
S B
ration in
NITROGE
4J
C
(U
O
c
o
o
u
M
p^
H
h- 1
jgj
W
O
w
s
w
o
a
w
H
TOTAL
DIOXIDE
w
M
pq
S
Q
M
fe
w
o
M
X
o
J>]
S3
O
o
w
o
a
P3
g
fo
N
M
n
0
Jj^
1
•s
O
a
CD
N
JD T)
rt 111
to n)
T3
l-l
QJ
4-1
a
P.
o
o
H-l
o
0)
>-i
3
o
4-)
a)
p
T3
o
ex
M
4-1
CD
6
o
cl
o
n
a)
0) -H -H •"
H > 4-1 tS 03
N -H H n! 1)
r-l (0
OJ
n)
pel
O
£ n3 tti trj
< D U Q
a
0)
en
2-75
-------
J_l
d)
4-J
ro Q)
I-*"* r-J S
Q (-T-,
H ^ -H
CO „ >-, X)
W
W ^ i^ ^3 C
^ M > ^ ^
tU ^ ^ MO
^U ^ P 'H
• eye o 4J
M Q K rt
f-l ^ 4-1
erf w" -<
W i-J
Z Q <
O M Z
S O
4-1
O
C
4-1
N
D
O.
B
O
o
QJ
l-i
3
3
BJ
•s
o.
•H
O
U) H) rH
C -H d rt
•H rH O rH
« -H -H
tfl -4J rH 4-1 XI
e it) ffl rt o
•5 p u p 5;
l-Ht-HrHrHi— liHi-HrHrHcSICNCN(NCN
X *
< rH
2-76'
-------
xxxxxxxxxxxx
IH
<1)
•'""*> 4-J
CO 0)
§"""" 0\
rH O
H -H
°° 2 00 M 3
W -H
C/) HH r— J X
M ^ C/l MO
t> g 3 -rH
• C3-53 O 4J
M W M to
p-l M
H 4-1
• ** c
<*j W S
53 O
i G
« O
o
rt
w
N
rH >-i
w H
53 Q H 55
9 ^
<2 ^ W
CJ Z
Q 53 2
Q 2 H
^ S^i pJ
K O S
w
s
^
H
O
H
W
a M
w w «
SHH S
X
H O
M M
53 « O
R
M
Pn
W
,-]
M
U O
M W S
Crf Q
H M
M X
53 O O
PM
W
t-H
«
W O
py. *y<
o
o
o
rH
fa
5
o
a
CO
r^.
o\
i — i
CX>
Q)
"0
0)
CX
(11
C/3
xxxxxxxxxxxx
xxxxxxxxxxxx
CMtNOOOOi-HOr-ICMCNX
KT u j uj IN ^J vij tr\ ^¥ i
rHOCOCOCMOOOC^XXXX
CO
>
to
g
•H
4-1
to
N
•rH
f, TJ
CO 01
I I ) I
CO CO
T)
C -rl
CO
U-l
a)
3
T3
to
o
a.
B
o
4-1 CO "1
CO >
>H CO ^
•H CO •'-f
rH 4-1 43
CO CO O
O Q S
X * o
-------
H J
O D
H co
OOOOCNCN-*CNOOOO*
W
Z Q
9*
2 o
<-> S
X^£>OOu"irOOOvOrHvOrHO* moXXOOC-OcNvOOO
rH i—I i-H rH t-H rH iH rHrH
"£
^ 0)
CO 4-1
r~, CU
Q rH
& CJ
H » vH
CO tTi >, .JD
w |7 | 3
0 OrH 1 r.
NO 3
W tH >TJ >
^ti i * rH c3
P3 <2 ss MO
to o 3 -H
H ^B 3 3
O IH
H sc! 4J
co C
Pi O
CJ CJ
13
O
O
^
rH W
CO N
Z fH
O J
^ l S
S U
W
pq
s-/
Z
w ro
O Q
O HH
Pi X
H O
M M
z n
o
M W
M X
Z O
z
0
N
o
w
1
0
w
£3
<4
H
g
»-J
d~ LO I-O to ""^J" "*s~ i*"** oo
rH
CT»
rH
„
oo XXXXXXXXXOCJOCJcovOvOo^
t-H vD r~- \D CO
cu
CO
p, tHrHrHiHrHrHrHrH
d)
CO
X i— 1 CTi r-1 O\ rH iH
X r- 00 O i-l 10 O
oo vo CTS oo r^. LO
O\ CTi ON CTi CTv C^
O OO r~ rH O vO rH
O O O O CT* rH CM
o o o o a\ o o
rH rH rH I-H iH rH
,
u~) O O O O O O
<* oo r- xo -
n)
4->
0
a
tH
o
c
o
•H
4J
CO
N
•rf
|
•H
•a
4-t
to
(3
•H
M
4-l
^
d
o
CO
rH
c-3
O rH
•H -H
4-1 Cfl
cfl >
5 CO
rH 4J
CO Cd
0 0
CO
•H
co
CO
, I
•H
43
O
1^
X
2-78
-------
Q
"
H
CO
W
o
Nl
REGIONAL 0
§
w
H
CO
M
H
•
pd
jq
I
CO
, [
(SEPTEMBER
o
|_|
W
o
a
O
H
U
O
Cd
CO
O
U
M
QJ
4J
0
O
•H
>>,£>
M 3
id
|
p
CO
nj
4J
td
Q
CO
00
td
M
01
r-l
M
3
O
SB
0
M
0)
crograms p
•H
0
a
•H
c
0
•H
4-1
]
°
TOTAL
w
M
CQ
Q
M
F*4
w
Q
M
O
H
O
N
O
W
M
PQ
O
a
Q
S
M
M
m
Q
s
o
S
g
O
M
r-^
CT,
rH
rH
H
0)
4J
(X
CO
OOOOOOO>-<(N
ooooooooououux
•s
rt
o
c
n)
N
o T)
rt cu
-
(1
0
O
CJ
>
4-1 0)
co n) i—i
•H rH O i-H 4-'
(fl -H »H ""
Vi > 4-1 n) «
cu cJ td J*
N -H M nj . .o ^
iH n) -H td -H
Cd 4-1 r-t 4J -^
c td
-------
p
2
w3
Z PC
5 e>
M O
O Pd
eu
z o
^^ j**l
M 3
O pa
SS
H £
CO rH
W "5
;— 1
• o1
M
H
^~\
CO
r~-
o\
i— 1
^
CO
CM
.
w
PP x-V
O M
EH QJ
O 4J
p
^ 0
<4 iH
vo >-t 3
CM cd O
1 3
CO B ^
CM 3 0)
00 ft.
W cd m
§ rt 3
W Q ^
H 00
P>, 0) O
W 00 H
co cd o
0 >
% < a
•< -H
M >>
>-l rH C
PH rl O
< 3 -H
X O 4J
33 n)
• ^t
H C
g -•
0 H
PQ -^
5 5
O
Q "f£\
£ c_>
M
PQ
N— '
W
r
H
1
O
J2
W
e
s
«!
E~i
o
H
|g
sa
H O
H M
Z «
W
I-H
M
^
O
I-H
Pn
M M
F-5 S
^ K
Ss ^D
3
H
"f
^*
o
$
L""
1*1
Z
N
O
H
o
x
o
e
t*
***
fjj
s
OtrH rHCNlCNi-HlOvOfO
S^^SSS^S^JnrnSS
rH rH t-H rH rH rH
-K*-lc-)C*ONv£)O*-Kr^-co-)c
m VD sr CM
in CM rH rH f~~
in "^* co co CM
•K****OOO*-K OO*
rH r-H rH rH rH
* rH
O t^v ON rH rH
i i j
c^ * r*^ * ON
CM .H CO
"•3" 1 — 1 ON
CM CM ON
O * O * ON
rH rH
CO CO OO
m co co
0*0*0
rH i— 1 rH
ON CM
XI X X X X
rH rH rH rH r-H
X X X X X
co ON
CD C
N -rl
rH* «
Cd 4J
(- Cd
•5 Q
X *
C
o
•rH
4-J
at
M
n
•H
rH
cd
o
0
.
4-J
3
a.
s
o
o
o
01
>-J
•H
frt
vu
4H
O
4-1
0)
•a
ca
M
cd
42
O
CX
•H
t-l
i-l
V>
e
O
I-l
14-1
>,
rH
g
01
cd a)
'"j *j
cd co
-------
^-1
Q
5
EH
C/3
W
tz>
a
ts3
0
t~]
<]*
J2^
o
t->
CJ
2
pM
W
H
21
•
>-t
H
Pd
y
<;
rH*
3
0
«
^^
^
a)
4-1
CU
g
o
•rH
XI
CJ
^1
CU
a.
co
2
tu
}^
00
O
o
1
c
•H
O
•H
4J
[0
4J
O
a
o
o
*~^
H J
O D
H CO
/— \
erf
W
N
"Id J
S3 Q
*****u~l****
CO
CO
CO
x o o *
X CJ CM *
iH
****ON****
* * *
o
CO
* o * * * *
co r^ 4-1 tO CO
cu d to >
N -H t-l tO 4)
CO
n)
CO tO to
Q CJ Q
X * CJ
rH
•H
XI
2-81
-------
OrHCJr- l
i-t^DVDCT\vOCNiHCT\-3-
CM CM
rHcMrHCMCMi-HCM
,—1
•»
c*">
1
CM s-*
M
(Xf: Q)
W t->
PQ CU
>-i O 0
P H
S 0 CJ
H O -H
CO >-, JQ
S P M 3
W
W jn to •< a
H H ^ -H
CO (— t >^
Si ^ § 7l §
£D H M fl
S C!
ptj d)
* U
C o
fe o
g ^"^
Q
F-l
r^
S
o
X- \
prf
CO N
Z tH
0 J
K fZ
^^ ^N
O
p S
bn ^2
pj Q_J
W
PQ
^—^
U
1
g
0
PL]
5
H
u
§
H
O
EH
*3
g o
Q M
H 0
M M
Z P
i-l
M
0
^
P
^
M
U
M W
dJ P
H M
Z O
H
Z
O
N
0
Ct]
M
cq
0
S
P
G
H
M
CC
S
P
E
h-
PJ
c
^r^OCNlc^OU* CNJ-K -K * -K -K
VD i/i
I
O
o
0)
t-l
CU
•a
CO
O
CX
§
c
o
C
O
4-1 CO
CO
0)
3 <~l
CO
4J rH *J J3
CO CO CO O
P U Q SI
CO >
n)
CO
X *
2-82
-------
XXX XXXXXXXXX
xxxxxxxxxxxxxx
iH
CO
CN '^N
C^ ®
g «
>-< o S
5 o u
H o "H
S Q *-" 3
fi3 2 j^. cd o
23 K !? 0
0 0 1 rl
N O ^o 3 CD
O Pi CM M 0-
p-l 1
H to n) <2
<5 W (x, 4-1 g
3; CJ CO co
O Z p^ Q t-i
M M W
3 00 g CO O
CO f-i M *H
§ <* §3 >
Cl4 £H ry] _, 33 CO
r_j ^j i i
^ 3 c
P3 d*
„ o
^-i G
H °
z
0
CJ
£_)
5
s
Od
o
H i-J
O P
H CO
P?
M
W J
o M i
§ 0
CJ 0 <
CJ
W
o
rH W
CO M
a >H
O J
< §
0
o z
Q S
JH ^
ffi CJ
W
cq
1
rH
&j
g
O
^
w
1
H
LiJ
JEJ
TOTAL
2
w w
O Q
O rH
rt X
H 0
M M
Z «
M-J
M
PQ
o
S
C
)_
CJ
M W
PH Q
H M
M X
a o
J-t-
K
p^
o
s
Q
[i
X
^~
P1^
w
^
o
o
^
i —
C
2
C
£
^
p>
C
• c^ cj^ 10 co in
OOrHOO •X-KHC-K^C'K-K-K-K rHCO
QJ CJ ^o ^^ if) o*\ CM co ^^ LO CM CM r*** r*^« r~H *^" c? *^J* *^ o^ •"") r*^ \^ co co r^*
CM CO rH r-t i— 1 rH
. .
cu Q)
'o 'o
J-J 4-)
O CM CO "^T i/1^ \O r*^> CO O*\ C^ rH CN CO "^ O *—\ OJ CO *^" LT^ vO r^« CO O"» C3 rH CM CO -i«^"
OrHrHrHr-HrHrHiHrHCMCMCMCMCMO rHrHi— li-HrH
0)
i — 1
•§
rH
•H
CO
>
CO
4J
O
C
^
o
o
•H
4-J
CO
N
•H
rH
•H
r£) T)
CO d)
4J 4-1
cfl cO
T3
C -H
•H i — 1
CO
M >
(U C
N -H
rH CO
CO 4-1
X *
01
4-1
o
CJ
O
J-J
£-J
rH
•H
Cfl
MH
O
4-1
0)
T3
in
4-1
CO
o
p.
•rH
^4
| I
cn
6
o
)H
MH
^
rH
C
O
rH
.Q
C "co 0)
O rH 4-1
•H -H -H
4-i co in
co >
IH CO 0)
^ n rH
•H CO -H
rH 4J ^3
O Q S
CJ <3 rH
2-83
-------
W
a
o
N
o
_>
o
wP
u
CO
I
co
CO
H
co
3
Q
ou
QJ O
60 M
cd o
O1
PQ
CO
C
(U
o
o
d D
O 3
H CO
a Q
O M
M x
3 o
§
o
Q
N
frf >
rH O rH 4-1
cd -H -H -H
> 4-1 n) co
rH Id -H Cd
Cd 4J rH 4-1
cj cd cd cd
< a o o
x * o <
2-84
-------
o^ M
^"*" j_j
| •> o
H | -H
CO I/-) K** r^
W ) cd o
eg p,j (H -H
W PH > -] O
'""^ N^i
•^ pt^
H J
o 5
H co
x^x
Pi
w
N]
rH >H
W J
S3 n <
O H Z
pp X ^
3 O
^ S5 S3
o o <
S g
u
H
«
rH W S3
CO N O
O J
O 53
O g ^
Q ^ rH
S I 1
w
^^-^
I_J
H
O
H
S
W W pq
° M §
S ><
H O
a p o
KN
M
fa
W
1-1
P9
0 0
M W g
Pi Q
H M
M X
S3 0 P
M
W
>-)
H
«
S3 S
O
O
rH
pi
CD
§
QJ
3
C
•H
C
O
0
en
S
>J
0)
O
4J
u
o
r~- oo vo O
OOOOOOOOrHrHrHiH
rHCNJ.-IOOOOOOOOO
i— liHl— (i— 1
OOrH
rHi—IrHrHrHi—IrHCNCNCNCMCN
3
a
o
CJ
m
o
0)
M
3
3
T3
a
a
•H
E
o
c
o
r^ ^o
CO cu
4J 4-1 0)
0) CO rH
d -H c co cu
•H rH O rH 4->
CO -H -H -H
t-1 > 4J CO CO
0) C ct)
N
CO 0)
Hi
. CO -H
CO 4-1 rH 4-1 ,Q
C CO CO CO O
< Q O Q «g
X * U << rH
2-85
-------
CJ
H W
l-l X
j:^ o
Q
fa
OU^r~-vomo\C^v£)T-4CT\ OO
r^.r^vor^vo-O cvj
OOOOOOOOOO XXUUCJo*
co
O
i-Hi-lr-IOOO XXOOUO-K * * * rHi-Hi-l-K
,0 "V
cd 01
4J 4J
cn td
T3
C -H
M
a)
4J
I
U-<
O
01
cd
o
s
ro
4J
CJ
ex
•H
1-1
4-J
CD
I-I
4-t
01
rH
.0
OO
LTJ
ai
•8
4J
CJ
O
C cd a)
O rH 4J
td -H -H fl
V4 > 4J cfl 0)
o) c cd >
N -H l-l cd -1
rH Cd -H Cd 1-1
Cd 4J t—t 4J fQ
C cd cd cd o
< a u Q S
X * O
2-86
-------
INVESTIGATION OF OZONE AND OZONE PRECURSOR
CONCENTRATIONS AT NONURBAN LOCATIONS IN
EASTERN UNITED STATES
Part 3. Airborne Ozone Monitoring Program
by
J. B. Tommerdahl
C. E. Decker
L. A. Ripperton
J. J. B. Worth
Research Triangle Institute
Research Triangle Park, N. C. 27709
Performed as Subcontract No. 33-73
from PEDCo Environmental Specialists, Inc.
Contract No. 68-02-1343
EPA Project Officer: E. C. Tabor
Quality Assurance and Environmental Monitoring Laboratory
National Environmental Research Center
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N. C. 27711
May 1974
-------
1.0 INTRODUCTION
During the period of July through October, 1973, the Research
Triangle Institute conducted an investigation of ozone (0 ) and 0
precursor concentrations at nonurban locations in the eastern United
States under EPA Contract 68-02-1077, RTI Project 41U-848. This field
program was designed to evaluate the occurrence and significance of
high 0 concentrations at nonurban locations. The program consisted
of the measurement of 0 , nitrogen dioxide and nonmethane hydrocarbons
at three of the four sampling sites; ozone only was measured at the
fourth site.
The purpose of the airborne 0,, monitoring program was to collect
supplementary air quality data; i.e., measurements of concentrations of
ozone above the ground within the geographic region of the fixed moni-
toring stations established under Contract 68-02-1077.
A solid phase 0 analyzer with support equipment was installed in
a C-45 aircraft supplied by the National Environmental Research Center -
Las Vegas. Data validation consisted basically of ground calibrations
performed with the equipment installed in the aircraft, and confirmed by
making low altitude passes at each of the monitoring sites when weather
conditions permitted. Data from airborne instruments were compared with
fixed station values for the respective times. Data were collected over
a flight path which covered the route from Raleigh, N.C., to Lewisburg, W.Va,
to Garrett County, Md., to Kane, Pa., to Coshocton, Ohio, and return to
Raleigh, N.C. Vertical descents in 2000 ft (610 m) increments were made
over each of the ground stations as weather conditions permitted.
3-3
-------
2.0 MEASUREMENT SYSTEM
2.1 Aircraft System
The measurement system consisted of a solid phase chemiluminescent
0 monitor installed in a C-45 aircraft. The aircraft and pilot were fur-
nished by the NERC-LV. The aircraft is shown in Figure 1.
The C-45 (Beechcraft) has a cruising speed of 150 knots, a minimum
speed of 95 knots, climb rate of 500 ft/min, turning radius of 1-1/4 mile,
useful load above pilot and fuel of 1600 Ibs and a range of 5-1/2 hrs or
880 miles (1416 km) at cruise speed. Navigation equipment included 2 VOR
units with DME, ADF, and Flux-gate Compass.
The air intake probe visible in Figure 1 was approximately 12 ft (3.6 m)
in length and coupled into an expansion cone inside the cabin, which effectively
slowed the air velocity by a factor of 10/1. The intake air was vented
directly into the cabin. The analyzer sample inlet line was located near
the center of the expansion cone at the outlet end.
The 0 meter was mounted on shockmounts and bolted to a shelf just
aft of the pilot's compartment on the port side of the aircraft. The
instrument is shown installed in Figure 2. Other equipment in view in
Figure 2 includes a strip chart recorder for data collection, mass flow
meter anct digital voltmeter for precise output voltage measurements.
Power for the instrumentation was supplied by an inverter (28 Vdc
to 115 V 60 Hz) which was connected to the aircraft's primary power bus.
Two inverters were available, one as a. standby in case of failure of
the other unit. These units were supplied with the aircraft. A block
diagram showing the configuration at the instrumentation is given in
Figure 3.
3-4
-------
Figure 1. C-45 aircraft used for flight
program showing sampling probe
-------
Figure 2. Instrumentation for airborne ozone
measurement program
3-6
-------
K C W
O -rt Q
P-, -^ O
PH CN FT1
H •—' 2
CO
w
CO
Q
O
era
H
Z;
W
co pi
W W O
MOM
Is!
w
K
CJ
N
a
O
O
vo
i-l &
O W
> H
cj S
Pi
£
Is
00
CNJ
CO
U CO
M PQ
4-J
05
c:
o
•H
c
cu
(-.
4-1
CO
•H
bC
•H
•d
o
o
m
bO
•H
3-7
-------
2.2 Ozone Analyzer
The 0^ analyzer utilizes the rhodamine B chemiluminescent disc as the
basic sensor and a photomultiplier tube for collecting and amplifying the
emitted light. As air containing 0« is passed over the disc, a reaction
between the 0., and the organic compound or dye imbedded on the disc takes
place and light is emitted. The amount of light emitted is directly propor-
tioned to the concentration of 0 . The design and operational characteri-
zation are described in Reference 1.
A functional diagram of the 0_ meter is shown in Figure 4. The system
has three modes of operation: measure, purge and calibrate. In the
measure mode, the inlet air is passed directly to the detection chamber
and the 0 detected; in the purge mode, the inlet air is first passed
through a clean up filter which removes water vapor and other contami-
nants and destroys any 0_ present before it passes through the detection
chamber; in the calibrate mode, the unit operates in the purge configura-
tion but a controlled amount of 0. is generated in the "clean air" as it
passes through the calibration unit by exposing the air stream to a UV
light source. A calibrated aperture control facilitates varying the 0.,
concentration and a front panel control and meter provides for lamp current
control and monitoring, respectively. The light emitted from the disc is
detected by a photo-multiplier tube, amplified and displayed on a strip
chart recorder.
A typical output signal is illustrated in Figure 5, as measured
during one of the flights. A calibration signal appears once per cycle
and is used as follows in determining the value of the measure signal:
. . . , measure value in mV
0~ concentration in ppm = ck —r— : ;
3 calibrate value in mV
where k is a correction factor which is a function of the ambient pres-
-------
6P
ca
4-J
0!
B
3-9
-------
o
o
t
1 f~ —
o /
ui ;
CO ,
s
• t )
cci
_i
<
o
UJ
CO
_l
<
o
o
10
ro
CJ
OJ
IO
OJ
Q
UJ
LU
ro
ro
CVJ
ro
CVJ
OJ
C
o
N
o
01
CD
to
J3
fX
T3
o
M
o
tfl
c
bO
•H
tn
tfl
o
•H
LO
0)
60
•H
3-10
-------
sure or altitude and c is the calibrate value in ppm determined by standard
calibration procedure.
As may be noted from Figure 5, the instrument is in the measure
mode 25% of the time. The value of each calibrate and measure signal
is obtained by using the amplitude of the signal minus the preceding
purge value; thus, the drift of the instrument is accounted for each
cycle. The response time of the instrument was essentially limited
by the response time of the strip chart recorder, which has a response
of 1/2 sec full scale, when the analyzer amplifier time constant control
is operated in its "normal" setting.
Comparisons and evaluations have been made previously between the
gas-phase chemiluminescent and the solid-phase chemiluminescent 0
analyzer under field conditions, see References 2, 3.
3-11
-------
2.3 Calibration
The analyzer was calibrated in the laboratory before installation
and again after it was removed from the aircraft. In addition, in-situ
calibrations were performed on the ground just prior to data flights
with the analyzer installed and operating off of a ground based power
supply. The calibration unit used was a portable 0 source which utilized
a UV lamp for generation of known concentrations of 0 , referenced to
the neutral-buffered KI procedure (as published in Federal Register, April
30, 1971, Edition). 0 output at the source manifold was controlled by
maintaining a constant mass flow of air through the quartz tube and a
constant UV lamp current. A flow rate greater than the sampling requirements
of the analyzer was maintained.
The 0 analyzer incorporated an internal calibration unit which
uses the radiation flux from a UV lamp for generation of 00. It has been
determined that if the UV radiation flux is held constant (i.e., main-
taining constant lamp current) the ratio of 0 to 0 is constant for a
given volumetric flow rate. Therefore, by maintaining a given volumetric
flow rate or knowing the flow rate, the 0 concentration in ppm may be
determined for any altitude or pressure.
Because of various factors, it was not feasible to maintain constant
volumetric flow at all altitudes, but constant mass flow was obtained
by using a mass flow meter in the gas line of the instrument and adjusting
for a pre-selected mass flow rate at each altitude. Tests were
conducted in flight to determine the volumetric flow rate at the various
altitudes with a fixed mass flow rate of 200 cc/min. These were checked
several times with a bubble flow meter to test the repeatability of the
method. It was not feasible to use a bubble flow meter on descents
3-12
-------
over a station during a data collection flight because the time period
at each altitude was too short. A couple of test flights were flown
to check out the operation at the lower altitudes in areas suitable to
this type of flight.
Tests were performed in the laboratory on the analyzer which had
been flown to determine the ozone concentration generated by the internal
calibrate unit for volumetric flow rates over the range of interest. This
was done by varying the flow rate in the 0 analyzer and measuring the
0 concentrations in the calibrate mode against a known concentration
established in an external manifold. This, then, yielded a value for the
internal calibration unit corresponding to each altitude or volumetric
flow rate of interest.
3-13
-------
3.0 DATA COLLECTION PROCEDURE
The normal flight procedure was to warm-up the equipment, run
the calibration checks and be ready for take-off from the Raleigh-
Durham airport around 0900. The equipment was turned off during take-
off due to power requirements of the aircraft and turned on immediately
after clearing the field. The equipment was allowed to stabilize for
approximately 30 minutes after take-off.
The flight path was, in general, as shown in Figure 6. A constant
altitude was maintained between sampling sites, and cruising speed was
150 knots. Standard airway routes were flown with some deviation under
VFR conditions. The use of the VOR radials and the DME allowed for
fairly precise position location in addition to visual checks over
identifiable points.
Upon reaching a sampling station, descent in 2000 feet (610 m) incre-
ments were made; at each altitude, the aircraft was leveled off and a cir-
cular or race-track pattern was flown for approximately 5 minutes or until
at least two data points were obtained. A 50 ft (15.2 m) pass over the run-
way was made at the Greenbrier Valley Airport and the Garrett County Airport
to obtain a data point as near as practical to the sampling station. The
descent pattern was flown (approximately 8 km) to the NW of the Greenbrier
Valley Airport and around the Garrett County Airport. At the Kane, Pa.
sampling site, the lowest altitude was approximately 600 ft. (183 m) above
the terrain and the flight pattern was around the perimeter of the town.
The circular flight patterns were such that a full 360° was usually
completed in approximately 5 minutes.
3-14
-------
BUFFALO
CLEVELAND
AKRON
CANTON
COSHOCTON
354 in
-------
The flight pattern over the Coshocton, Ohio site consisted of fly-
ing a VOR vector so that a straight ground track was flown for this site
Straight and level flight at each altitude was approximately 5 minutes
with the sampling station near the center of the ground track. Descent
over the station was limited to approximately 600 ft above the terrain under
VFR (visual flight rules) conditions.
Refueling stops were made at Bradford, Pennsylvania just after the
descent over the Kane, Pennsylvania sampling site. The total flying
time required for one full circuit was approximately 7-1/4 hours. The
program flying time required a total of 26 hours which included test
flights for equipment procedure check-out and the data acquisition flights.
3-16
-------
4.0 DATA
Data are presented in the following formats as available for the
data acquisition flights of August 8 and September 11-12, 1973.
a) Ground track plot - showing relative locations of cities and
sampling stations. Ozone concentrations and time are super-
imposed on the ground track for the flight. Altitude for
each leg of the flight is also shown.
b) Elevation plot - showing relative elevation of the ground track
above MSL in the lower plot and ozone concentration with respect
to time and location in the upper plot.
c) Vertical sounding - showing the ozone concentration versus alti-
tude over the sampling sites. Ground elevation and data points
for the three 15 min averages which bracket the time when the
closest approach data point were obtained.
d) Sampling site data - for period of 0700 to 1800 for the flight
day, showing the time that the aircraft was over the station.
e) Strip chart data - examples of rapidly varying ozone concentrations.
The flight altitudes and ground elevation above mean sea level (MSL)
are presented in feet and meters since the aircraft altimeter, the assigned
flight altitudes and navigation chart data are given in feet. The altitudes
given are nominal values since there is some error in the altimeter and some
variations in flight altitude over short periods of time. All of the
ozone contration data are given in Ug/m at standard conditions.
The altitudes at which the measurements were made may be obtained from the
respective figures.
A.I Data Acquisition Flight for August 8, 1973
The flight log was as follows:
a) Flight was started under VFR conditions, increasing cloud
cover required subsequent increases in altitude; unable to descend over
Greenbrier Valley Airport because of low visibility; climbed to 11,900
ft to get into clear and file IFR flight plan, unable to maintain flow
3-17
-------
rate at this altitude, therefore, no valid data; rest of flight was made
under IFR conditions.
b) Flew over Kane, Pa. sampling site, approximately 2000 ft
above terrain, data checked quite closely with data observed by calibration
crew at the sampling station.
c) Flight to Coshocton, Ohio was aborted near Youngstown, Ohio
due to engine failure; emergency landing was made at Akron-Canton Regional
Airport.
Data for the flight are presented in Figures 7, 8, and 9. Ozone
concentration for the fixed sites for the same date are given in Figure 10.
Two illustrations of changing ozone concentrations are given in the strip
chart tracings in Figures 11 and 12.
3-18
-------
W«SHINS TOM
AIRBORNE OZONE
MEASUREMENTS
FLIGHT DATE e/e/73
-' *-CQURS£
RAl EtGH - -* BRADfORD
BRADFORD -*- AKRON-CANTON
TIME-(EOT J-SHOWN TO RIGHT
OF FLIGHT PATH
Oj CONCENTRATION (/-"J /ml)
SHOWN TO LEFT OF FLIGHT
PATH
FLIGHT ALTITUDE SHOWN TO
RIGHT OF FLIGHT PATH
Figure 7. September 8, 1973 data acquisition flight.
3-19
-------
60
o
•H
4-1
•H
CO
•H
3
cr
o
rt
03
oo
OJ
4-)
cx
a)
oo
01
3
60
O ID c\J
NOIlVdlN30NOO JNOZO
NOI1WA313
QNOOdO
3-20
-------
OZONE
UJ
a:
o
m
a:
**•
C/)
h-
UJ
f^
UJ
cr
ID
<
UJ
*!>
§
a.
._
O
r-
LJ
UJ
.
8/8/73
i
UJ
I-
o
X
o
ll
Z
ON-CANTO
cr
rf
1
g
o
u.
a
-i
M
•H
lf>
(fiUJ/ t>T<)
NOIlVdlN30NOO 3NOZO
NOI1VA313
3-21
-------
o
oo
CM
o
CO
OJ
o
-
o
CM
o
o
o
o
o
o
o
CD
o
o
o
CO
o
o
g
o
CO
e
o
•H
4-1
Cfl
J-l
CO
c
3
o
•1-1
so
o
Hi
ex
in
0)
OD
en
D
CD
V-i
o
o
•rl
a
CJ
CJ
c:
o
o
c
o
O
i-l
a)
60
•H
£uu/ 6TT
3-22
-------
CO
<
LJ
w
Z co
0 H
N Z
O w
§
w H
3 ;zj
o co
pq <
03 W
M S
<
n
r~
a-,
i— i
B
00 ^|
oo
n o>
<
!-i
cfl
1)
C
1
C
0
•H
4-1
rt
a
o
^J
CJ
CVJ
ro
c\J
a
LU
LU
c
o
•H
4-J
n)
l-l
J-l
c
01
o
c
o
o
-------
w
53 V*
O H
Nl S3
O W
S
W W
e
>^
•H
a
to
c
o
•H
ro
o
en
3
00
t
to
CN
60
•H
3-24
-------
4.2 Data Acquisition Flight for September 11, 1973
The flight log was as follows:
a) Roundtrip flight to Greenbrier Valley Airport with a descent over
the station, afternoon flight.
b) The flight was made under VFR conditions, a landing was made at
the Greenbrier Valley Airport.
c) Flow rate tests were conducted in flight, which is the reason for
the different altitudes flown for different legs of the flight path
and the breaks in the data.
Data for this flight are given in Figures 13, 14 and 15. Vertical
measurement data is presented in Figure 16 and fixed site measurements
are given in Figure 17.
3-25
-------
CHARLESTON
BECKLEY
PULASKI
AIRBORNE OZONE
MEASUREMENTS
( FLIGHT PATH )
FLIGHT DATE - 9/11/73
RALEIGH
COURSE
—» GREENBRIER
TIME-(EOT)-SHOWN TO RIGHT
OF FLIGHT PATH
03 CONCENTRATION-(/ig /m3)
SHOWN TO LEFT OF
FLIGHT PATH.
FLIGHT ALTITUDE SHOWN TO
RIGHT OF FLIGHT PATH
BB2 GREENBRIER
f?
GREENSBORO
Figure 13. September 11, 1973, data acquisition flight.
3-26
-------
CHARLESTON
GREENBRIER
BECKLEY
PULASKI
AIRBORNE OZONE
MEASUREMENTS
( FLIGHT PATH )
FLIGHT DATE - 9/11/73
-——^COURSE
GREENBRIER »• RALEIGH
TIME-(EDT)-SHOWN TO RIGHT
OF FLIGHT PATH
03 CONCENTRATION (/tig /m3)
SHOWN TO LEFT OF
FLIGHT PATH
FLIGHT ALTITUDE SHOWN TO
LEFT OF FLIGHT PATH
1527
Figure 14. September 11, 1973, data acquisition, flight.
3-27
-------
r
UJ
0
UJ
z
o
CD
cc
^5
en
i —
UJ
UJ
a:
UJ
4^
£
O_
Z
o
UJ
1
UJ
m
V.
01
1
Ul
h-
X
_1
U.
IE
Ul
IE
CO
GREEN
I
I
«
Ul
_i
*
3
Ul
IE
t
IE
Ul
IE
CD
2
Ul
Ul
IE
(9
Ul
U
on
0
CO
Ul
INDICAl
z
STATIC
IE
Ul
g
1
^
'f.
1
CO
z
10
Ul
INDICAl
1
— A
CHECK
I-
z
Ul
oj
800
01
CM *
CO ^1
C\J
K)
o
io
Q
UJ
O
-------
iu
Q
3000 r
2700
2400
2100
1800
1500
JI 1200
900
600
300
g /m° EOT
88.2 -r 1319
92.! - - 1325
149.0 - - 1327
168.6 -- 1329
183.0
178.4
1333
1335
A 701 m
6REENBRIER VALLEY
STATION DATA
ug /m3 HOUR
183.0
190.1
192.1
1315
1330
1345
AIRBORNE OZONE
MEASUREMENTS
(VERTICAL SOUNDINGS)
FLIGHT DATE-9/11/73
-274m ABOVE TERRAIN
k- GROUND ELEVATION
Figure 16. Vertical Measurements for September 11, 1973.
3-29
-------
oc
LU
o:
o
£ 2
< o
CD Sr
O
O
00
o
o
o
o
CO
o
o
in
o
o
o
o
o
o
o
o
o
05
o
o
o
00
o
UJ
O UJ
O S
cvj —
o
CM
CM
o
o
CVJ
o
00
o
(O
o
CVJ
o o o
O 00 U>
o
cvj
1
w
c
o
•H
T)
fi
3
O
^
60
4-1
a
a)
P.
CO
0.1
i-t
a,
0)
CO
CD
S
O
•H
a
co
a
a
o
o
0)
a
o
N
o
Ml
•H
6n uj
3-30
-------
4.3 Data Acquisition Flight for September 12, 1973
The flight log was as follows:
a) Flight originated at RDU and was flown under VFR conditions.
b) All legs of flight path from RDU to BFD were flown at 8500 feet;
and from BFD to RDU an altitude of 9000 feet was maintained.
c) Descents in 2000 ft increments were made over each of the sampling
sites.
d) Meteorological observations are tabulated for this flight.
Data are presented in Figures 18, 19 and 20 showing ozone concentration
versus position or location. Vertical measurements over each of the fixed
sites is given in Figure 21 along with ground station data for the three
15-minute periods which bracket the time of closest approach to the station.
In-flight observations for the flight are given in Table 1. The fixed
site data for the day of the data flight is given in Figure 22. An example
of the strip chart recording for a rapidly changing concentration condition
is shown in Figure 23.
3-31
-------
AIRBORNE OZONE
MEASUREMENTS
FLIGHT DATE 9/12/73
»• COURSE
Rfl!_E!GH ^-=V BFADFORD
BRADFORD —1JV RALEIGH
r
Figure 18. September 12, 1973, data acquisition flight.
3-32
-------
o
QD
o:
EVATIO
s is
60
•H
C
O
•H
4J
•H
CO
•H
cr
o
cfl
0)
^
0)
0)
en
M
NOI1WW1N33NOD 3NOZO
NOI1VA313
ONnoao
3-33
-------
o
^g
Li-l ?H
•<=: O-
0=
-------
UJ £/)
o^
8^
UJ UJ
ztr
tr ^
ocn
m <
cr uj
< ^
C
O
•H
4-J
•H
CO
•H
3
a1
o
n)
cfl
-a
2
or o •> o
3 « » g
«- E S°~2ie
S § S §
o
o
o
o -
o t
O Cft
(O 00
OD OD
(-1
0)
•i
01
4-)
a,
0)
LO
fc
u
o
o
o
o
o
o
o
o
o
CD
o
o
o
o
o
o
o
CO
o
o
ro
) aannnv
o
M-I
5 §
C
0)
B
a)
M
3
en
CO
01
s
cfl
o
4-1
M
QJ
RIER
OJ
M
00
•H
3-35
-------
INFLIGHT OBSERVATIONS - 9/12/73
The in-flight observations in Table 1 were made for the flight
on 9/12/73. The location of the observations may be determined by locating
the time on the ground track on the appropriate figure. The outside
temperature was measured using the aircraft outside air temperature
sensor. In general, climb and descents were made at a rate of 500 ft/min,
the response time of the air temperature sensor was not known; the readings
were made in approximately 500 ft increments, where practical.
Table 1
Flight Observation for September 12, 1973
Time
(EDT)
0946
0950
0955
0959
1007
1013
1018
1037
1045
1047
1050
1112
1128
1146
1153
1203
1210
1250
1253
1300
1427
1432
1434
Altitude
(ft)
6000
7500
8500
8500
it
ii
M
ii
8500
6500
5500
4500
8500
M
6500
4500
8500
M
6500
4500
2500
5000
7500
8500
Outside
Air Temp
(°c)
12
9
11.5
10
8.5
9
10
12
16
15
-
14
12
10
8
15
8
8
3
9
14
' 6
0
6
Aircraft
Attitude
(C-climb)
(L-level)
(D-descending)
C
C
L
L
L
L
L
L
D
L
D
L
L
D
L
L
L
L
L
L
C
C
L
Observation
Top of Haze
Thin cloud layer
around 700 ft
Over Greenbrier
Valley
Top of Haze Jevel
Thin scattered 6000 f 1.
Cloud base 5500 (>
Over Kane, Pa. site
u
u
Cloud base 6500 ft
3-36
-------
Table 1 (Continued)
Meteorological Observations - 9/12/74
1450
1502
1514
1516
1525
1537
1542
1548
1552
1603
1604
\
1610
1644
1653
1710
1720
1733
i
f
1742
Altitude
ft)
9000
II
Tl
ft
II
6500
4500
2500
2000
5000
6000
6500
7000
7500
8000
8500
9000
9000
11
It
It
9000
8000
7500
7000
6500
6000
5500
5000
Outside
Air Temp
Aircraft
Attitude Observation
(°C) (C-climb)
(L-level)
(D-descending)
5
6
7
8
9
7
12 *
17
19
9
8
7
6.5
8
8
8
10
10
10
10
11
12
13
14
15
15
14
15
17
L
L
L
L
L
L
L
L
L
C
C
C
C
C
C
C
L
L
L
L
L
D
D
D
D
D
D
D
D
Over Coshocton, Ohio site
Haze level
Cloud Tops
Haze level
1750
1500
26
Near RDU
3-37
-------
or
LJ
tr
o
c.
LU
CD
o
o
OJ
o
00
o
OJ
o
o
o
00
o
CVJ
o
o
GO
o
o
o
o
to
o
o
m
o
o
o
o
to
Q
LJ
_ uj
o 5
o _
CVJ t-
O
O
O
O
o
o
o
05
o
o
o
00
o
o
o
T3
c
JJ
>
•H
Ol
O.
CO
0)
Oi
,-3
B
c
(U
c
o
d
G
N
M
p
SC
Brf
3-38
-------
0)
o
01
r^
W en
IZ CO rH
OH g
N Z
O W OJ rH
^ ' — ' O~>
W U u-1
?5 2 r- rsj
PS t> cu
O co .n
CQ •< E 4-)
Pi W 0) rH
Mi* 11 ^-r1
^^ 4-J
<1J
^
O
c
tO
o
&
M
0)
>
O
1
c
o
•H
4-J
tO
CJ
O
>J
CD
_l
<
O
O
CJ
o
OJ
o
c
o
•rH
4-1
a)
M
4-1
e
0)
CJ
C
o
o
0)
c
o
N
o
60
C
•H
60
d
tfl
j;
u
o
•rH
P.
03
o
OJ
OJ
o
ro
OJ
O
C
O
•H
4J
cfl
V-i
ro
CN
60
•H
3-39
-------
5.0 SUMMARY AND CONCLUSIONS
The purpose of the program was to obtain some measurements of tht'
ozone concentration in the troposphere in the region bounded by the
fixed monitoring sites operated under Contract 68-02-1077.
A solid phase chemiluminescent ozone meter with internal calibration
unit was installed and flown in a C-45 aircraft. Calibrations were
performed on the ground and the flow rate was monitored in-flight. In
addition, tests were conducted in-flight and on the ground to obtain and
verify the altitude correction factor. The equipment functioned well in tht;
aircraft; vibration was not a problem; the power sources were adequate
and stable.
Flights were made both under VFR and 1FR conditions. During IFR
conditions, it was not feasible to make descents over the sampling sites
and the altitude requested was contingent on approval of the flight
plan. Operationally, Lt was more practical to n:ake descents to the
selected altitudes, (i.e., acquire the desired data at one altitude
and then descend to the next altitude) over the station than it was to
climb sequentially to successive altitudes.
inspection of the data on the strip chart records indicated that the
ozone concentration was stable over the period of the sample,
which represented 30 sec in time and approximately 1.44 miles spatially
at the normal cruising speed. Rapid changes in ozone concentrations were
observed in some regions. Verification of the aircraft data by flying
3-40
-------
as close to the fixed monitoring sites as feasible proved to be practical
in most instances. The dynamic checks made in this manner showed good
correlation between the ground station and the airborne ozone measurements,
thus confirming the satisfactory operation of the airborne units (See Table 2)
Table 2. COMPARISONS GROUND AND AIRCRAFT OZONE DATA
Greenbrier Valley
Garrett Co.
Kane, Pa.
Coshocton, Ohio
Groundreading
(15 min. avg.)
129.4 yg/m3
145.0 yg/m3
90.2 yg/m3
119.6 yg/m3
Time
1100
1145
1300
1600
Aircraft
127.4 yg/m
160.7 yg/m3
94.1 yg/m3
129.4 yg/m3
Time
1057
1154
1301
1554
Height
Above
Ground
100 ft
100 ft
600 ft
600 ft
The isopleths drawn in Figure 24, which shows 0 concentrations during
the flight of 12 September 1973, are speculative, but tend to show that
the 0 behavior is regional and not local in nature. The horizontal
gradients are small and much interpolation between data points has been
used.
There is a temptation to ascribe various features of lower and higher
0 values to nearby urban areas, but the data are too few. Also, inter-
pretation would require a more detailed past history of the air as well
as a knowledge of the chemical input from the various cities. Both increases
and decreases in 0 values appear downwind of urban areas. There is
a corridor of low values of about 59 iig/m downwind of Charleston, West
3
Virginia and an area of higher values (137 to nearly 157 Mg/m ) downwind
of the cluster of cities, Cleveland, Akron, and Canton.
3-41
-------
BRADFORD
*"*7/<, ^CLEVELAND
AKRON YOUNGSTOWN
980
1372
1176
COSHOCTON/
^COLUMBUS
PARKERSBURG
WASHINGTON
CHARLESTON/
58 8
BECKLEY
OZONE CONCENTRATION
CONSTANT ALTITUDE FLIGHT
SEPT 12, 1973
(O955-I755 EOT)
GRLtNSHORO
\ DURHAM
Figure 24. Ozone concentration, constant altitude flight, September 12, 1973.
(Observed concentration values have been reduced to standard conditions)
3-42
-------
Extrapolation of temperature sonde data from pertinent weather
stations on September 12 (confirmed by temperature measurements taken from
the aircraft) indicated the presence of an inversion located somewhere
between 5000 and 8500 feet aloft over the four sampling sites.
The vertical distribution of ozone data indicates that the relation-
ships between ground values and data aloft are not obvious. The Oo
vertical profiles shown over the Greenbrier Valley Airport (Lewisburg)
present a picture of a low value of 0~ above the inversion and 0., generated
from ground-sourced precursors below the inversion level. Ozone measured
at and near the ground at the Greenbrier Valley Airport was generated in
the lower atmosphere and was not transported from the stratosphere by
vertical turbulence.
Another implication of the vertical data is that the 0 or 0
precursors had moved into the ground sampling area at a level below the
altitude of the aircraft, but that at times surface influence on Oo
concentration can extend several thousand feet.
Measurements made from an aircraft have proved extremely interesting
and informative. The flights made as part of this project suggest that
a greater sampling frequency would yield more than a one for one increase
in useful information.
-------
6.0 REFERENCES
1. "Ozone Chemiluminescent Study," Final Report, Under NAPCA
Contract No. CPA-22-69-7, by Research Triangle Institute,
December 1969.
2. "Field Evaluation of New Air Monitoring Systems," Final
Report under EPA Contract No. CPA-70-101, by Research
Triangle Institute, May 1972.
3. Hodgeson, J. A., et al., "Laboratory Evaluation of Alternate
Chemiluminescent Approaches for the Detection of Atmospheric
Ozone," presented September 1970 ACS Meeting.
3-44
-------
TECHNICAL REPORT DATA
(I'lcasc read Instructions on the reverse before completing)
1 RLPORT NO
EPA-450/3-74-034
a riTLE AND suuHTLE Investigation of Ozone and Ozone Precur-
sor Concentrations at Nonurban Locations in the Eastern
United States: Field Measurements; Quality Assurance
Program; Airborne Ozone Monitqring_^£ogram
7 AUTHOR(S)
Research Triangle Institute
9 PLflFORMING ORG \NIZATION NAME AND ADDRESS
Research Triangle Institute
Research Triangle Park, N.C. 27709
12 SPONSORING AGLNCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Office of Air and Waste Management
Research Triangle Park, N.C. 27711
3 RECIPIENT'S ACCESSI ON-NO.
5 REPORT DATE
._ May 1974
6. PERFORMING ORGANIZATION CODE
8 PERFORMING ORGANIZATION REPORT NO,
10. PROGRAM ELEMENT NO.
1HA326
11 CONTRACT/GRANT NO.
68-02-1077
68-02-1343
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15 SUPPLEMENTARY NOTES
16. ABSTRACT
Ozone concentrations were measured continuously at ground-level at McHenry,
Maryland; Kane, Pennsylvania; Coshocton, Ohio; and Lewisburg, West Virginia between
June 26 and September 30, 1973. Nitrogen dioxide and nonmethane hydrocarbon
concentrations were determined at all stations during this period with the exception
of McHenry, Maryland. The measurement period extended through October at Kane,
Pennsylvania. Hourly ozone concentrations exceeded the National Ambient Air Quality
Standard for photochemical oxidants 37, 30, 20, and 15 percent of the hours for which
data are available at McHenry, Kane, Coshocton, and Lewisburg, respectively.
Quality assurance studies indicated that the average relative bias for the ozone
concentration measurements was +10 percent. Nitrogen dioxide concentrations were
at or near background levels throughout the study. Between-station linear corre-
lation coefficients for hourly ozone concentration comparisons varied from .468
to .678 for simultaneous data. It was concluded that the occurrence of high ozone
concentrations at nonurban locations is widespread, affecting a large area in
eastern United States. A C-45 aircraft equipped with a solid-phase chemiluminescent
ozone meter was used to obtain ozone concentration measurements aloft. The
aircraft data support the contention that the high ozone concentrations observed at
the surface were generated in the lower tropoBphere.
KE Y WORDS AND DOCUMENT ANALYSIS
Rural
Ozone
Nitrogen dioxide
Hydrocarbon
Measurements
DESCRIPTORS
Photochemical
Oxidants
Airborne
Quality assurance
Nonurban
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Held/Group
13B
'I DISTRIBUTION STATEMENT
Release unlimited
19 SECURITY CLASS (This Report)
N/A
21 NO OF PAGES
20 SECURITY CLASS (This page)
N/A
22 PRICE
EPA Form 2220-1 (9-73)
3-45
-------
D
O
TJ
T)
O
» o
-too
^ a> o>
c
w
Z
T)
|—
O
O
*V
G
td
M
h-H
n
>
H
i—i
O
^
Z
O
m
^
>
i
4x
^i
O
OJ
I
-~J
*h
6
^>
4X
N O 3J
=6.2
o
O
o rr
Q.
O)
Q) O>
Cl • •
in O
O)
O)
a.
(D
D
.
_ n
<~> CD
OJ -
n •—
zr 3"
n
>
°
« O
5 c :
9 5 -
(A
Ifl
>0
Cl
f«
Z
o
-------
|