EPA-450/3-76-033
August 1976
FORMATION AND TRANSPORT
OF OXIDANTS
ALONG GULF COAST
AND IN NORTHERN U.S.
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
EPA-450/3-76-033
FORMATION AND TRANSPORT
OF OXIDANTS
ALONG GULF COAST
AND IN NORTHERN U.S.
by
C.E. Decker, L.A. Ripperton, J.J.B. Worth
and
P.M. Vukovich, W.D. Bach,'J.B. Tommerdahl, F. Smith, D.E. Wagoner
Research Triangle Institute
Research Triangle Park, N. C. 27709
Contract No. 68-02-2048
Program Element No. 2AC129
EPA Project Officer: Edwin L. Meyer, Jr.
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
August 1976
-------
This report is issued by the Environmental Protection Agency to report
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 - in limited quantities - from the
Library Services Office (MD35) , Research Triangle Park, North Carolina
27711; or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22161.
This report was furnished to the Environmental Protection Agency by
Research Triangle Institute, Research Triangle Park, N. C. 27709,
in fulfillment of Contract No. 68-02-2048. 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-450/3-76-033
11
-------
ACKNOWLEDGMENTS
This project was conducted by the Research Triangle Institute (RTI),
Research Triangle Park, North Carolina, under Contract No. 68-02-2048 for
the United States Environmental Protection Agency. The support of this
agency is gratefully acknowledged as is the advice and guidance of the
Project Officer, E. L. Meyer, Jr. and other staff members of the Office
of Air Quality Planning and Standards.
Special acknowledgments are given to two laboratories of the United
States Environmental Protection Agency—the Environmental Monitoring and
Support Laboratory (EMSL) of the National Environmental Research Center,
Research Triangle Park; and the Environmental Monitoring and Support
Laboratory of National Environmental Research Center, Las Vegas. The
EMSL-RTP participated in the quality assurance auditing program at cooper-
ating state and local agency urban stations and at RTI rural stations
and analyzed high volume particulate samples for sulfates, nitrates, and
ammonium ion. Special recognition is given of the technical support provided
by T. A. Hartlage, B. Martin, J. A. Eraser, and Dr. R. J. Thompson, as is
the advice and assistance provided by officials of the EMSL, Dr. T. A. Hauser,
and Laboratory Director, Dr. D. S. Shearer. The EMSL-LV provided RTI access
to their environmental test chamber for altitude testing of air quality
analyzers and their B-26 aircraft, pilots and instrument technicians for
joint airborne measurement flights in the gulf coast area in October 1975.
Work on this project was performed by staff members of the Systems
and Measurements and Energy and Environmental Research Divisions of RTI
under the general direction of Mr. J. J. B. Worth, Group III Vice President.
Mr. Worth was Laboratory Supervisor for this program. Mr. C. E. Decker
served as Project Leader and was responsible for the coordination and
conduct of the program. Staff members of RTI who contributed to the prep-
aration of this report are recognized and listed in alphabetical order:
Dr. W. D. Bach, Mr. C. E. Decker, Mr. R. B. Denyszyn, Mr. W. C. Eaton,
Dr. L. A. Ripperton, Mr. F. Smith, Mr. J. B. TommerdaH, Dr. F. M. Vukovich,
Dr. D. E. Wagoner, and Mr. J. J. B. Worth.
iii
-------
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ill
LIST OF FIGURES ix
LIST OF TABLES xvii
Section
1.0 EXECUTIVE SUMMARY 1
1.1 Introduction and Objectives 1
1.2 Design of Study/Field Measurement Program 1
1.3 Principal Findings 2
1.4 Conclusions 7
2.0 INTRODUCTION 9
2.1 Background 9
2.2 Research Objectives 10
3.0 DESIGN OF STUDY 11
3.1 Northern High Pressure Oxidant Study 12
3.1.1 Airborne Measurements 12
3.1.2 Ground Measurements 13
3.1.3 Ozonesoncle Measurements 13
3.2 Gulf Coast Oxidant Study 15
3.2.1 Airborne Measurements 15
3.2.2 Ground Measurements 23
3.2.3 Ozonesonde Measurements 23
3.3 Analysis Procedure 23
3.3.1 Northern High Pressure Oxidant Study 23
3.3.2 Gulf Coast Oxidant Study 24
4.0 FIELD MEASUREMENT PROGRAM 25
4.1 Ground Stations 25
4.1.1 Sampling Protocol 25
4.1.2 Siting Considerations and Description
of Monitoring Stations 26
4.1.3 Air Quality Measurements 32
4.1.3.1 Instrumentation 32
4.1.3.2 Instrument Calibration
and Maintenance 35
iv
-------
TABLE OF CONTENTS (con.)
Section Page
4.1.4 Data Acquisition and Data Processing 37
4.1.5 Data Validation and Quality Control 38
4.2 Airborne Measurements 39
4.2.1 Airborne Measurements System Description 39
4.2.1.1 Aircraft 39
4.2.1.2 Measurement System 42
4.2.1.3 Instrumentation 42
4.2.1.4 Data Acquisition System 47
4.2.2 Instrument Calibration and Pressure Effect
Tests 49
4.2.2.1 Instrument Calibration 49
4.2.2.2 Pressure Effects Test 50
4.2.3 Operations Procedures and Data Validation
Techniques 50
4.2.4 Data Reduction and Processing 55
4.2.5 Aircraft Flight Summary 63
4.3 Ozonesonde Measurement Program 63
4.3.1 Introduction 63
4.3.2 Instrumentation and Data Acquisition 63
4.4 Program Summary 72
5.0 QUALITY ASSURANCE PROGRAM 75
5.1 Quality Assurance Protocol 75
5.1.1 Qualitative Systems Audit 76
5.1.2 Performance Audit 77
5.2 Description of the Air Pollution Monitoring Network 77
5.3 Audit Procedures 81
5.3.1 RTI Ozone Auditing Procedures 81
5.3.2 EPA Ozone Auditing Procedures 82
5.3.3 EPA Oxides of Nitrogen Auditing Procedures 82
5.4 Analysis and Interpretation of Audit Data 83
5.4.1 Precision/Accuracy Estimated for Ozone
Measurement 84
v
-------
TABLE OF CONTENTS (con.)
Section
5.4.2 Precision/Accuracy Estimates for Nitric Oxide
and Nitrogen Dioxide Measurements 89
5.5 Summary of Audit Results 89
6.0 SUMMARY OF DATA AND STATISTICS 91
6.1 Summary Statistics 91
6.2 Diurnal Patterns 100
6.3 Summary of Climatic Conditions 113
7.0 INTERPRETATION OF RESULTS: NORTHERN HIGH PRESSURE
OXIDANT STUDY 121
7.1 Examination of Aircraft Ozone Measurements 121
7.1.1 Data Analysis Approach 121
7.1.2 Summary of Aircraft Ozone Measurements
and Meteorological Conditions 122
7.2 The Relationship Between the High Ozone in the Rural
Boundary Layer and High Pressure Systems 128
7.2.1 Introduction 128
7.2.2 Statistics on High Ozone Concentrations
Versus High Pressure Systems 129
7.2.3 Distribution of Ozone Relative to a Moving
High Pressure System 133
7.2.4 Source Regions and Residence Times of Air
Relative to a Moving High Pressure System 139
7.2.5 Some Aspects of the Mechanism Governing the
Chemistry of Ozone in High Pressure Systems 148
7.2.5.1 Theory 148
7.2.5.2 Variation of the Diurnal Cycle in a
Moving High Pressure System 158
7.2.5.3 Variation of Ozone Chemistry in a
High Pressure System in the Eastern
Portion of the United States 161
7.2.6 Summary of Northern High Pressure Analysis 169
7.3 Chemistry of Ozone Generation 173
7.3.1 Relationship of Ozone and Population Density 177
7.3.2 Ozonesonde Releases 191
vi
-------
TABLE OF CONTENTS (con.)
Section
7.3.2.1 Ozonesonde Releases at Huron, South
Dakota, September 5-7, 1975 192
7.3.2.2 Stratospheric-Tropospheric Ozone
Distribution 192
7.3.2.3 Aircraft and Ozonesonde Profiles 196
7.3.3 Hydrocarbons and Halocarbons 196
7.3.3.1 Variation of Selected Hydrocarbons 196
7.3.3.2 Hydrocarbons: High Pressure System
Flight of September 6-7, 1975 201
7.3.4 Particulates, Northern High Pressure Study 201
7.3.4.1 Ground Site Measurements 201
7.3.4.2 Aircraft Particulate Measurements,
Northern Study 205
7.3.5 Summary of Ozone Generation (Northern High
Pressure Oxidant Study) 207
8.0 INTERPRETATION OF RESULTS: GULF COAST OXIDANT STUDY 209
8.1 Examination of Ozone Measurements and Meteorological
Conditions 209
8.1.1 Data Analysis Approach 209
8.1.2 High- and Low-Ozone Days 211
8.1.3 Summary of Aircraft Ozone Measurements and
Meteorological Conditions 221
8.1.4 High Ozone Occurrences at Austin, Texas 255
8.1.5 Summary of Gulf Coast Aerial Survey 266
8.2 Chemistry of Ozone Generation 268
8.2.1 Bag Irradiation Experiments 280
8.2.2 Ozonesonde Releases 280
8.2.2.1 Introduction 280
8.2.2.2 Ozonesonde Releases at DeRidder,
Louisiana, October 24-27, 1975 284
8.2.2.3 Stratosphere-Troposphere Ozone
Distribution 284
8.2.2.4 Low Altitude Distribution of Ozone 287
8.2.3 Variation of Selected Hydrocarbons 289
vii
-------
TABLE OF CONTENTS (con.)
Section
8.2.3.1 Theoretical Considerations
8.2.3.2 Variation of Selected Hydrocarbons
8.2.4 Particulates
8.2.4.1 Total Suspended Particulates;
Sulfate, Nitrate, Ammonium
8.2.4.2 Gulf Coast Aircraft Particulate
Samples
8.2.5 Summary of Ozone Generation (Gulf Coast
Oxidant Study)
9.0 CONCLUSIONS
10.0 REFERENCES
Page
289
291
294
294
294
297
299
301
APPENDIXES
APPENDIX A.
APPENDIX B.
APPENDIX C.
APPENDIX D.
APPENDIX E.
APPENDIX F.
Calibration Systems/Procedures
Performance Characteristics and Operational
Summaries for Instruments
Airborne Platform Air Sampling System Design
Ozone and Oxides of Nitrogen Analyzer Evaluation
at Reduced Pressure
Summary Sheets for Aircraft Data and Selected
Flights
Background Data and Emission Study for Texas
Gulf Coastal Area
305
315
319
343
353
447
viii
-------
LIST OF FIGURES
Figure Page
1 Site locations for ground station network. 14
2 Sea breeze flight plan. 16
3 Downwind plume flight plan. 18
4 Areal survey flight plan. 19
5 North-south survey flight plan. 20
6 Land-sea survey flight plan. 21
7 Double-box flight plan. 22
8 Air monitoring station, Bradford Regional Airport. 28
9 Interior view cf air monitoring station. 28
10 Aerial view of Creston, Iowa (Creston Municipal
Airport) site. 29
11 Air monitoring station, Creston Municipal Airport. 29
12 Aerial view of Wolf Point, Montana
(Wolf Point International Airport) site. 30
13 Air monitoring station, Wolf Point International
Airport. 30
14 Aerial view of Lewisburg, West Virginia (Lewisburg
Airport) site. 31
15 Air monitoring station, Lewisburg Airport. 31
16 Aerial view of DeRidder, Louisiana (Beauregard
Parish Airport) site. 33
17 Air monitoring station, Beauregard Parish Airport. 33
18 Instrumented aircraft. 40
19 Functional layout of airborne sampling system. 41
20 Block diagram, aircraft air sampling system. 43
21 Diagram of physical layout. 44
22 Diagram of grab sampling system for hydrocarbon
sample collection. 46
23 Diagram of selective filter sampling system. 46
24 Sketch of airborne high volume sampler. 47
25 Aircraft vertical profile (typical flight track). 51
26 Example of ground measurement comparison. 53
27 Low pass pattern for aircraft/station comparison. 53
ix
-------
LIST OF FIGURES (con.)
Figure Page
28 Comparison data during low passes. 54
29 Comparison data during low passes. 55
30 Normalized response versus altitude for Bendix
ozone analyzer. 57
31 Normalized response versus altitude for Bendix
oxides of nitrogen analyzer. 57
32 Example of initial data plots. 58
33 Sample flight track. 61
34 Vertical profile; flight 079 flown on September 21,
1975. 62
35 Gulf coast sea-breeze flights (6/28/75-10/21/75). 64
36 Gulf coast: inland survey flights (6/25/75-10/31/75). 65
37 Northern high pressure flights (7/8/75-9/30/75). 66
38 Mean diurnal ozo:ie concentrations at Bradford,
Creston, and Wolf Point (June 27-September 30, 1975). 107
39 Mean diurnal ozone concentrations at Bradford,
Lewisburg, and DeRidder (June 27-September 30, 1975
at Bradford; June 27-October 31, 1975 at Lewisburg
and DeRidder). 108
40 Mean diurnal ozone concentrations at Pittsburgh,
Columbus, and Poynette (June-September 1975). 109
41 Mean diurnal ozor.e concentrations at Omaha, Cedar
Rapids, and Des Koines (June-September 1975). 110
42 Mean diurnal ozone concentrations at Nederland,
Austin, Houston, and DeRidder (June 1-September 30,
1975, except DeRidder—July-October 1975). HI
43 Mean diurnal ozone concentrations at Kane (1973),
DuBois (1974), and Bradford (1975) Pennsylvania. H2
44 Mean diurnal ozona concentrations at Columbus, Ohio
for 1974 and 1975, 114
45 Mean diurnal nitrogen dioxide concentrations at
Bradford, Creston, Wolf Point, and DeRidder
(June 27-October 31, 1975) 115
46 Tracks of high pressure center (July 1975). H6
47 Tracks of high pressure center (August 1975). 117
48 Tracks of high pressure center (September 1975). 118
x
-------
LIST OF FIGURES (con.)
Figure Page
49 High pressure system flight on July 24, 1975. 124
50 High pressure system flight on July 25, 1975. 125
51 High pressure system flight on July 26, 1975. 126
52 Area averaged value of the daily maximum ozone con-
centration (solid line) and the area averaged surface
pressure (dashed line) versus day of month for latter
part of the summer of 1973 and 1974. 130
53 Daily maximum ozone concentration at Wolf Point,
Montana; Creston, Iowa; and Bradford, Pennsylvania
and the daily average pressure obtained from the
nearest synoptic station for the summer of 1975. 131
54 Nine-point running average of the data presented in
figure 53 and of the Bradford ozone data and the
Pittsburgh pressure data presented in figure 53. 134
55 The temporal and spatial variation of the diurnal
maximum ozone concentration through a high pressure
system located in the east in the summers of 1973,
1974, and 1975. 135
56 The average temporal and spatial variation of the
diurnal maximum ozone concentration through a high
pressure system located in the east based on the
1973, 1974, and 1975 data. 137
57 The temporal and spatial variations of the diurnal
maximum ozone concentration through a high pressure
system based on the 1975 data at Wolf Point, Montana;
Creston, Iowa; and Bradford, Pennsylvania. 138
58 Hypothetical high pressure system for which residence
time were calculated. 141
59 The number of days air parcels, in various locations
in a high pressure system, will spend within that
system versus the speed of the system for a circular
symmetric high pressure system. 143
60 The number of days air parcels in high pressure sys-
tems have spent (residence time) within the system
versus the speed of the system for a circular
symmetric high pressure time.
xi
-------
LtST OF FIGURES (con.)
Figure Page
61 The number of days air parcels in various locations
in a high pressure system will spend within that
system versus differently shaped elliptical high
pressure systems. The system speed is 7.5 ms~l. 146
62 The number of days air parcels have spent (resi-
dence time) within a high pressure system versus
differently shaped elliptical systems. The system
speed is 7.5 ms"!. 147
63 The diurnal variation of the transport and syn-
thesis term in a rural boundary layer in North
Carolina. 153
64 The variation of 6 versus A. 153
65 The variation of the concentrations of NO, NO , and
ct-pinene versus 6 for various values of T. 154
66 The variation of g versus 8* and &**, 157
67 The variation of the ratio, 0., /0_ versus 6. 157
3o 3max
68 The average diurnal variation of ozone concentra-
tion at Wolf Poict, Creston, and Bradford based
on eight high pressure systems which consecutively
passed through these stations. The day with the
largest diurnal maximum ozone concentration, when
the high pressure; system was in the vicinity of the
station, was used to compute the average for all
eight systems. 159
69 The average diurnal variation of ozone for those
days when the diurnal maximum ozone concentration
exceeded the NAAQS (solid line) and when the diurnal
maximum was less than the NAAQS (dash-dot line)
based on the data for August 1973 (A), 1974 (B), and
1975 (C) and at Kane, DuBois, and Bradford, Penn-
sylvania, respectively. 162
70 The average diurnal variation of ozone at Lewisburg,
West Virginia for those days when the diurnal maximum
ozone concentration exceeded the NAAQS (solid line)
and when the diurnal maximum was less than the NAAQS
(dash-dot line) based on data for August 1973 (A) and
1975 (B). 166
71 Population density by counties: 1970. 178
72 General equal response curves (0.08 ppm ozone) for an
alkane-N02 system (solid line) and for an olefin-N02
system (broken ILne). The two diagonals with their
indicated slopes define the HC/N02 ratios which
correspond, to maximum ozone production for each type
of hydrocarbon. 182
xii
-------
LIST OF FIGURES (con.)
Figure Page
73 Ozone versus median NOX concentration range at
Bradford, Pennsylvania and Wolf Point, Montana
(July-September 1975). 190
74 Time-altitude cross section of ozone and potential
temperature. 194
75 Vertical profiles of ozone at Huron, South Dakota. 197
76 Vertical profiles of ozone at Huron, South Dakota. 198
77 Bar graph showing propane and acetylene concentra-
tions by day of week at Bradford, Pennsylvania,
Creston, Iowa, and Wolf Point, Montana (July-
September 1975). 199
78 Linearized plot of selected hydrocarbons from flight
on September 6-7, 1975. 202
79a Arriving air trajectories associated with lower
decile concentrations of maximum ozone at DeRidder,
Louisiana. 212
79b Arriving air trajectories associated with upper
decile concentrations of daily maximum ozone at
DeRidder, Louisiana. 213
79c Arriving air trajectories associated with lower
decile concentrations of daily maximum ozone at
Nederland, Texas. 214
79d Arriving air trajectories associated with upper
decile concentrations of daily maximum ozone at
Nederland, Texas. 215
79e Arriving air trajectories associated with lower
decile concentrations of daily maximum ozone at
Houston, Texas. 216
79f Arriving air trajectories associated with upper
decile concentrations of daily maximum ozone at
Houston, Texas. 217
79g Arriving air trajectories associated with lower
decile concentrations of daily maximum ozone at
Austin, Texas. 218
79h Arriving air trajectories associated with upper
decile concentrations of daily maximum ozone at
Austin, Texas. 219
xiii
-------
LIST OF FIGURES (con.)
Figure Page
80 Ozone concentrations, air trajectories and sea level
pressure distribution for August 7, 1975 flight. 222
81 Ozone concentrations, air trajectories and sea level
pressure distribution for August 8, 1975 flight. 223
82 Ozone concentrations, air trajectories and sea level
pressure distribution for August 9, 1975 flight. 226
83 Ozone concentrations, air trajectories and sea level
pressure distribution for September 19, 1975 flight. 227
84 Ozone concentrations, air trajectories and sea level
pressure distribution for September 21, 1975 flight. 229
85 Ozone concentrations, air trajectories and sea level
pressure distribution for October 10, 1975 flight. 231
86 Ozone concentrations, air trajectories and sea level
pressure distribution for October 13, 1975 flight. 232
87 Ozone concentrations, air trajectories and sea level
pressure distribution for October 14, 1975 flight. 234
88 Time altitude cross section of potential temperature
(°K) at Lake Charles, Louisiana. Dates are indi-
cated at 0000 GMT. 236
89 Ozone concentrations, air trajectories and sea level
pressure distribution for October 19, 1975. 237
90 Ozone concentrations, air trajectories and sea level
pressure distribution for October 20, 1975 flight. 239
91 Ozone concentrations, air trajectories and sea level
pressure distribution for October 21, 1975 flight. 242
92 Ozone concentrations and temperature from vertical
profile flight of October 21, 1975 at DeRidder,
Louisiana. 244
93 Ozone concentrations, air trajectories and sea level
pressure distribution for October 22, 1975 flight. 245
94 Ozone concentrations, air trajectories and sea level
pressure distribution for October 24, 1975 flight. 247
95 Air trajectories and sea level pressure distribution
for October 30, 1975 flight. 249
96 Analysis of aerial ozone distribution as measured on
EPA and RTI flight paths at 305 m near Port Arthur,
Texas, October 30, 1975. 250
-------
LIST OF FIGURES (con.)
Figure Page
97 Vertical profile of ozone concentration, temperature,
and winds at Lake Charles, Louisiana, October 30,
1975. 252
98 Ozone concentrations, air trajectories and sea level
pressure distribution for October 31, 1975 flight. 253
99 Ozone concentrations, air trajectories and sea level
pressure distribution of the August 8, 1975 flight
and the trajectories of parcels arriving in Austin
on the morning (O) and the evening (A) of August 9,
1975. 258
100 Ozone concentrations, air trajectories and sea level
pressure distribution of the August 9, 1975 flight
and trajectories of parcels arriving in Austin on the
morning (O) and the evening (A) of August 10, 1975. 259
101 Temperature profiles at Victoria, Texas, August 9,
0000 GMT to August 10, 0000 GMT, Isothermal (C) are
skewed. Dashed lines are the dry adiabats for the
0900, 1200 and 1500 CST temperatures at Austin, Texas
on August 10, 1975. 260
102 Hourly average ozone concentrations, August 9, 10,
1975 at Austin, Texas and the mean hourly average
concentration for July 1 to October 31, 1975. 261
103 Air trajectories arriving in Austin, Texas on
September 3-4, 1975 (Q - morning, A - evening). 264
104 Hourly average ozone concentrations, September 3, 4,
1975 at Austin, Texas and the mean hourly average
concentration from July 1 to October 31, 1975. 265
105 Ozone concentrations observed on sea-breeze flight
of September 19, 1975. 270
106 Ozone concentrations observed on aerial survey
flight on June 27, 1975. 271
107 Ozone concentrations observed on box flight pattern
on October 19, 1975. 272
108 Ozone concentrations observed on aerial survey
flight on June 26, 1975. 273
109 Ozone concentrations, transitions flight on
August 5, 1975. 277
110 Ozone concentrations, transition flight on June 25,
1975. 278
xv
-------
LIST OF FIGURES (con.)
Figure Page
111 Ozone increase as a function of initial NOX concen-
tration; captured air irradiation experiments,
DeRidder, Louisiana (October 1-31, 1975). 282
112 Ozone increase as a function of initial ozone con-
centration; captured air irradiation experiments,
DeRidder, Louisiana (October 1-31, 1975). 283
113 Time-altitude cross section of ozone and potential
temperature, lower portion. 285
114 Time-altitude cross section of ozone and potential
temperature, upper portion. 286
115 Vertical profile of ozonesonde data (solid) and
best fit profile (dashed) 100 to 100 mb. 288
116 Vertical profile of ozone departures from best fit
profile and time-altitude cross section of poten-
tial temperature. 290
117 Bar graph showing acetylene and propane concentra-
tions by day of week at DeRidder, Louisiana (July-
October 1975). 292
xvi
-------
LIST OF TABLES
Table Page
1 Pollutants measured at rural monitoring stations 26
2 Calibration techniques 35
3 List of data channels 48
4 Flight data, sea-breeze flight 079 (9/21/75) 59
5 Flight summary, northern route high pressure system surveys 67
6 Flight summary, gulf coast area 68
7 List of vertical profile flights 69
8 Flight summary, joint EPA-RTI gulf coast flights 70
9 List of location identifiers 71
10 Program schedule 73
11 Summary of ozone and oxides of nitrogen audit data 85
12 Statistical summary of hourly ozone concentration
measurements by station 92
13 Summary of ozone data above NAAQS by station 93
14 Statistical summary of hourly oxides of nitrogen
concentration measurements - rural stations
(June 27 - September 30, 1975) 94
15 Statistical summary of selected hydrocarbon and
halocarbon analyses 95
16 Summary of mean 24-hour particulate concentrations:
(TSP, NH£, NOg, S02p for rural stations (July to
September 1975) 96
17 Summary of mean ozone and oxides of nitrogen
concentrations by month 97
18 Summary of mean hydrocarbon and halocarbon concentra-
tions by month at rural stations (July - September 1975) 98
19 Cumulative frequency distributions of hourly
concentrations of ozone - rural stations (June 27-
September 30, 1975) 101
20 Cumulative frequency distributions of hourly concentra-
tions of ozone - State/local stations (June-September 1975) 102
21 Cumulative frequency distribution of hourly concentra-
tions of oxides of nitrogen - rural stations (June 27-
September 30, 1975) 103
XVI I.
-------
LIST OF TABLES (con.)
Table
22 Means of hourly concentrations of ozone for each hour of
day - RTI rural stations (June 27-September 30, 1975) 104
2^ Means of hourly concentrations of ozone for each hour
of day - State/local stations (June-September 1975) 105
24 Means of hourly concentrations of oxides of nitrogen
for each hour of day - rural stations (June 27-
September 30, 1975) 106
25 Candidate missions by sector 123
26 Average ozone concentration at mission altitude 123
27 Average ozone concentration for each operational sector 127
28 The relationship between the number of hours a high
system is near a station and the number of hours of high
ozone (greater than the NAAQS) observed at that station
for the period 4 July to 3 September. In 1975, the station
used was Bradford; in 1974, DuBois; and in 1973, Kane. 132
29 The values of t , 0 «/0 ** (where 0 * is Wolf Point
max 33 3
ozone amplitude in all cases) , g (estimated from t and
letting t' = 1400 LDT) , a, diurnal ozone amplitude
(diurnal maximum, ozone minus diurnal minimum ozone) , and
the diurnal minimum ozone concentration for Wolf Point,
Creston, Iowa, and Bradford, Pennsylvania obtained from
the diurnal curves given in figure 68. 160
30 The values of t , 0 */0 ** (0 * is the amplitude of ozone
TJlclX j j j
for the high ozone case), £ (using t and letting t1
m&x ni
1400 LDT), and a for Kane, Pennsylvania obtained from the
1973 diurnal curves given in figure 69.
31 The values of t , 0 */0 ** (0 * is the amplitude of
U13,X J J J
ozone for the high ozone case) , 3 (using t. and letting
fllcl2C
t1 = 1400 LDT), and a for DuBois, Pennsylvania obtained
max ' ' }
from the 1974 diurnal curves in figure 69. 163
32 The values of t , 0 */0 ** (0 * is the amplitude of ozone
max 333
for the high ozone case), g (using t and letting t1 =
1400 LDT) , and a for Bradford, Pennsylvania obtained from
the 1975 diurnal curves in figure 69. 164
-------
LIST OF TABLES (con.)
Table
33 The values of t , 0 */0 ** (when 0 * was the amplitude
max 3 J J
of ozone in 1973 in all cases) , 3 (using t ^ and letting
THclX
t' = 1400 LOT), a, the diurnal amplitude of ozone (ozone
max / » »
maximum minus the ozone minimum) , and the diurnal minimum
ozone concentration for the high ozone cases in 1973, 1974,
and 1975 from figures 69A, B, and C. 165
34 The values of t , 0 */0 ** (where 0* is the amplitude
max 33 3
of ozone in the high ozone case) , 3 (using t and letting
t' = 1400 LDT) , and a for Lewisburg, West Virginia
max
obtained from the 1973 diurnal curves given in figure 70. 167
35 The value of t , 00*/0 ** (where 0 * is the amplitude
max J 3 3
of ozone in the high ozone case) , 3 (using t and assuming
H1&2C
t' = 1400 LDT), and a for Lewisburg, West Virginia
IQ3.2C
obtained from the 1975 diurnal curves given in figure 70. 167
36 The values of t , Q*/0** (where 03* was the amplitude
of ozone in 1973 in all cases) , 3 (using t and letting
t' = 1400 LDT), a, the diurnal amplitude of ozone (maximum
ozone minus minimum ozone) , and the diurnal minimum ozone
concentration for the high ozone cases in 1973 and 1975 from
figures 70A and 70S. 167
37 The average maximum concentration of N02 found between the
time of maximum ozone and midnight using only the data on
high ozone days for August 1974 at DuBois, Pennsylvania and
in 1975 at Bradford, Pennsylvania; and the computed NO
concentration at the time of maximum N0£ assuming a three-
gas system (NO, NO , and NO ) . 168
38 Population density for States west of Fargo, N.D. -> Dallas,
Texas line* 175
39 Population density for states between Fargo, N.D. -»- Dallas,
Texas line and east of Chicago, Illinois -»- St. Louis,
Missouri line* 176
40 Population density for states east of Chicago, Illinois •>
St. Louis, Missouri line* 176
41 Summary of ozone data for rural stations 179
42 Summary of aircraft ozone data for population density areas 179
-------
LIST OF TABLES (con.)
Table
43 Dark phase ozone half life
44 Results of computer simulation runs
45 Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges at Wolf Point,
Montana (July-September 1975) 183
46 Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges at Creston, Iowa
(July-September 1975) 184
47 Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges at Bradford,
Pennsylvania (July-September 1975) 185
48 Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges for aircraft
samples - Region 1 - west of Fargo, N.D. -*• Dallas,
Texas 186
49 Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges for aircraft
samples - Region 2 - east of Fargo, N.D. and Dallas,
Texas line arid west of Chicago, Illinois •> St.
Louis, Missouri line 187
50 Mean hydrocarbon and haLocarbon concentrations for
selected ozone concentration ranges for aircraft
samples - Region 3 - east of Chicago, Illinois ->
St. Louis, Missouri line 188
51 Summary of NO data 189
X
52 Comparison of ozone measuring techniques 193
53 Average daily total suspended particulate (TSP) by
month at three sites (1975) 203
54 Sulfate as a percentage of TSP by month 203
55 Selected particulate constituents as percentages of
gross suspended particulates (1966-1967)* 204
56 Nitrate as a percentage of TSP by month 205
57 Nitrate and sulfate: north high pressure flights 206
58 Snmirarv nf cort'Htrons accompanying upper decile ozone
concentrations at Austin, Texas, July 1 to October 31,
1975 256
59 Mean hydrocarbon and halocarbon concentrations for
ozone concentration ranges at DeRidder, Louisiana
(July-October 1975) 274
xx
-------
LIST OF TABLES (con.)
Table Page
60 Mean hydrocarbon and halocarbon concentrations for
ozone concentration ranges, gulf coast - flights
over land (July-October 1975) 275
61 Average hydrocarbon and halocarbon concentrations
for ozone concentration ranges, gulf coast - flights
over water (July-October 1975) 276
62 Population density for southeastern States* 279
63 DeRidder, Louisiana bag irradiation experiments, 1975,
experimental data from bag irradiation 281
64 Average daily total suspended particulate (TSP) by
month at DeRidder, Louisiana (1975) 287
65 Selected particulate constituents as percentages of
gross suspended particulates (1966-1967)* 295
66 Nitrate and sulfate concentrations for samples
collected during gulf coast flights 296
xxi
-------
1.0 EXECUTIVE SUMMARY
1.1 Introduction and Objectives
During the summer of 1975, Research Triangle Institute (RTI) conducted
a two-part field measurement program designed (1) to determine the change in
the concentration of ozone in the center of a high pressure system, as the
system moves from an area of low population density to an area of high
population density; and (2) to determine the areal extent of high ozone
concentrations in the northern gulf coast region of Texas.
In the Northern High Pressure System Oxidant Study, the objective was
to determine the change in the concentration of ozone near the center of
high pressure systems, as these systems traverse the northwestern, north-
central, and northeastern areas of the United States. During their passage
over the United States, these systems traverse, first, regions of low
population density and little industrial activity—that is, regions having
small emission densities of the hydrocarbon and nitrogen oxide precursors
necessary for the production of the photochemical oxidants. During this
initial period, low ozone concentrations were anticipated. As the systems
move eastward, however, population density, industrial activity, and,
consequently, emissions of oxidant precursors increase. It was anticipated
that ozone levels would increase.
The objective of the Gulf Coast Oxidant Study was to document the areal
extent of high ozone concentrations in the northern gulf coast region of
Texas. Primary emphasis was on the roles and/or contribution of land-sea
breeze circulations, of local emissions of ozone precursors, and of transport
of ozone and ozone precursors within and downwind of the study areas.
1.2 Design of Study/Field Measurement Program
To accomplish the objectives described above at a minimum cost, two
independent studies were designed that could be conducted concurrently and
that employed similar measurement systems.
Based on the above considerations, two modes of measurement were
employed: (1) a network of five, fixed, rural sites operated by RTI plus
supplementary rural and urban ground-level sites operated by State/local/other
agencies, and (2) an instrumented aircraft flying specified patterns. The
field measurement program included continuous ozone and nitrogen dioxide
measurements, collection of grab samples for selected hydrocarbon and
-------
halocarbon analyses, and 24-hour, total suspended particulate (TSP) samples
at four manned, rural stations located in Bradford, Pennsylvania; Creston,
Iowa; Wolf Point, Montana; and DeRidder, Louisiana. Ozone measurements
were also made at an unmanned station at Lewisburg, West Virginia, and at
supplementary stations located in the general study areas, which were
operated by State and local governments or by private industry. Ozone and
oxides of nitrogen were measured and grab samples collected during aircraft
flights designed to accomplish specific objectives for each study area.
Supplemental vertical ozone data were obtained from ozonesonde releases at
Huron, South Dakota, and DeRidder, Louisiana. A joint RTI-EPA quality
assurance program was designed and implemented to assure that high quality
data were obtained.
The data acquisition program began at all stations before July 1, 1975.
Data were collected and quality assurance performance audits were performed
at specified intervals at each of the stations included in the monitoring
network through September 30, 1975, at northern stations and through
October 31, 1975, in the South.
Since only one aircraft was utilized for both studies, the aircraft was
based in DeRidder, and flights were conducted in the gulf coast area until
an appropriate high pressure system developed in the northern study area.
A total of 111 individual aircraft flights were flown during the combined
studies. These flights were flown under varying meteorological conditions
and included sea breeze flights, coastal areal survey slights, downwind
plume flights, vertical profile flights, double-box patterns around Nederland,
Texas, calibration and instrument checkout flights, and northern high pressure
system flights. During the month of October, several joint RTI-EPA/Las Vegas
flights were flown in the gulf coast area.
1.3 Principal Findings
The data obtained during the field measurement program were summarized
statistically and segmented into four general subject areas for analysis
and interpretation. These areas are: (1) Northern High Pressure Oxidant
Study, (2) Gulf Coast Oxidant Study, (3) Chemistry of Ozone Generation in
Rural Areas, and (4) Quality Assurance Program. Data were analyzed and
interpreted according to the objectives for each study and have been
incorporated into a comprehensive section that combined both a chemical and
-------
meteorological interpretation of the results. Principal findings are
presented separately below. Results of the Quality Assurance Program and
an abbreviated statistical summary of the data are also included.
A. Quality Assurance Program
Based on the precision and bias estimates obtained from the audit
data, it is concluded that the quality of the ozone and nitrogen oxides
measurements was comparable to the quality of similar measurements made in
other well managed networks and was sufficient to satisfy the summer study
requirements. Estimates of the relative bias, the coefficient of variation
and the 90 percent confidence interval for the error (deviation of the
measured value from the audit value) in the ()„, NO, and NO measurement
data are presented in section 5.5 of the report.
B. Statistical Summary of Data
1. The mean hourly concentrations of ozone at rural stations
ranges from a low of 58 yg/m3 at Wolf Point to 81 yg/m3 at
Bradford. For urban stations the range for mean hourly
concentrations was from 44 to 73 yg/m3. The standard devia-
tions for all stations, both rural and urban, were similar
in magnitude. The overall ozone mean (i.e. , ozone burden)
was, in general, higher for rural stations than for urban
stations.
2. Maximum hourly average ozone concentrations for rural
stations ranged from a low of 128 yg/m3 at Wolf Point to
a high of 256 yg/m3 at DeRidder. Maximum hourly averages
at urban stations ranged from a low of 180 yg/m3 to
629 yg/m3.
3
3. The NAAQS for photochemical oxidants (160 yg/m ) was
exceeded approximately 4, 3, 1, 1, and 0 percent of the
hours at Bradford, Lewisburg, Creston, DeRidder, and Wolf
Point, respectively and from less than 1 to 8 percent of the
hours at the urban stations.
4. Based on the percentage of days exceeding the NAAQS and
hours above the standard a west-to-east gradient in
ozone concentration was observed using northern rural
stations data as follows:
-------
Percent of
Hours above Days above days above
Site standard standard standard
Wolf Point 0 0 0.0
Creston 17 7 8.0
Lewisburg 59 11 11.1
Bradford 100 18 18.5
5. Mean hourly concentrations of nitric oxide and nitrogen
dioxide measured at the four rural stations (Bradford,
Creston, Wolf Point, DeRidder) were in the noise or below ~
the detectability level of the measurement method (<_ 10 yg/m ).
6. An increase in the percentage of sulfates and nitrates in
the suspended particulate matter from west-to-east stations
was observed. A definite trend was not observed in the
hydrocarbon and halocarbon concentrations observed at the
rural stations.
C. Northern Oxidant Study
Analyses of rural O2;one concentrations were made for (1) four
locations (Wolf Point, Creston, Bradford, and Lewisburg) and (2) three
locations in northwestern Pennsylvania during comparable dates of 1973
(Kane), 1974 (DuBois) , and 1975 (Bradford). The analyses were aided by
simplified models of atmospheric synthesis and destruction of ozone and
of the airflow near the ground in a transient anticyclone.
1. The analyses stiow that high ozone occurrences during the
summer months are associated with high pressure systems.
In the midwest and eastern United States, the lowest
ozone concentrations are found in the leading portion of
the system and the highest ozone concentrations in the
trailing portion of the system. The airflow model shows
that air parcels in the leading portion of the system have
the shortest residence time in the system and those parcels
in the. trailing portion, the longest residence time. Ozone
concentrations measured at Wolf Point show little vari-
ability as high pressure systems passed through that area.
2. Ozone—as indicated by the daily maximum concentration or the
number of hours the NAAQS was exceeded—increased when the high
pressure system was in the east compared to when it was in
the west. The increase in ozone was apparently due to a
west to east increase in the diurnal minimum rather than
a substantial west to east increase in ozone production.
3. The data suggested that the west to east increase in the
diurnal minimum can be attributed to lower concentration of
ozone destructive agents in the west. This allowed a residual
amount of ozone to remain after each diurnal cycle as the
parcel drifted eastward.
-------
4. The evidence indicates that there was a reversal of the role of
synthesis and destruction of ozone in high pressure systems
located in the east from 1973 to 1974 and 1975. Lower concentra-
tions of ozone destructive agents and smaller amplitude of
synthesis were found in 1973 compared to 1974 and 1975.
5. The summer of 1973 and the largest total number of runs of high
pressure and high ozone compared to 1974 and 1975. The large
number of hours of high ozone was summarized to be a result of
the relatively stagnant condition allowing air parcels in the
high pressure systems to have large residence times and to
experience many diurnal cycles depleting the concentration of
ozone precursors and destructive agents and increasing the level
of diurnal minimum.
6. In 1975 fewer hours of high pressure, fewer hours of con-
currant high ozone, and lower average ozone concentrations
during high pressure were observed than in 1974 or 1973 in
northwestern Pennsylvania, Macroscale high pressure systems
were not as persistent in 1975 as in the two previous years.
7. Ozone data from all aircraft flights show an increase of
afternoon (1300-1900 LDT) average concentrations from west
to midwest to east sectors of the study area as follows:
Sector Mean Standard deviation
3 3
West 60 yg/m., 4 yg/m~
Midwest 96 yg/m3 11 yg/m::
East 106 yg/m 16 yg/m
8. Anthropogenic pollutants (acetylene and halocarbons) are
present in all grab samples collected at northern rural ground
stations and during aircraft flights. Mean hourly average
oxides of nitrogen concentrations even though in the noise
level of the measurement technique are higher on a relative
basis at Bradford and at Creston than at Wolf Point. This
observation is based on over two thousand samples at each
site.
9. Analysis of a series of ozonesonde releases at Huron, South
Dakota, does not indicate ozone intrusion from the stratosphere.
10. Ozone concentrations measured at ground stations and during
aircraft flights and the sulfate and nitrate composition of
total suspended particulates increase from west to east. This
pattern is generally consistent with population density
patterns.
D. Gulf Coast Oxidant Study
1. Air moving slowly over areas of large hydrocarbon emissions
was associated with upper decile ozone concentrations at
-------
urban and rural locations in the gulf coast area. In most
cases, trajectory analysis showed air with high ozone con-
centrations arrived from nonprevailing directions.
2. Air that moved rapidly, showed weak anticyclonic curvature,
and had long overwater fetches was associated with lower
decile ozone concentrations at all of the ground station
locations.
3. In most cases, trajectory analysis attributed the highest
ozone concentrations to principal cities or areas of high
precursor emissions located in the gulf coast study area.
These observations suggest that ozone plumes commonly
develop downwind of large precursor emission areas.
4. Aircraft ozone measurements clearly demonstrate an ozone
plume (280 yg/m-^, maximum) upon a low (< 100 yg/m^) back-
ground downwind of the petrochemical complex at Port Arthur,
Texas. During the period observed, ground level ozone
concentrations at a continuously operating ground station
near the emission area were less than 100
5. Intercity urban plume transport of ozone or ozone precursor
materials was evident. This was shown as a potential cause
of some violations of the NAAQS at Austin.
6. In the survey flights, the mean ozone concentrations over
water were usually found to be less than those over the land,
regardless of the level of ozone encountered. When elevated
ozone concentrations were measured over the water, the tra-
jectory analyses usually showed the air parcel had a recent
(< 24 hr) history over continental areas , usually over high
precursor emission areas .
7 . When areawide ozone concentrations exceeded the NAAQS ,
vertical mixing was usually restricted by a stable layer
below 2 km.
8. Analysis of a series of ozonesonde releases at DeRidder,
Louisiana, does not indicate ozone intrusion from the
stratosphere. Mid- to-upper tropospheric ozone concentra-
tions changed by 50 percent or more during a day, but they
did not contribute to ozone changes at ground level.
9. Anthropogenic pollutants (acetylene and selected halocarbons
were present in all grab samples collected at the DeRidder
station and during aircraft flights. Examination of the data
indicate that hydrocarbon and ozone concentrations are not
linearly related at DeRidder.
-------
10. Four distinct areal distributions of ozone were identified from
aircraft flights:
3
a) area-wide low concentrations (-70 yg/m ),
b) localized plumes downwind of precursor areas,
c) elevated ozone concentrations, some exceeding NAAQS
for ozone, usually increasing from west to east,
d) area-wide (North Carolina to Louisiana) ozone con-
centrations exceeding the NAAQS for ozone.
1.4 Conclusions
The following conclusions derived from the data are listed separately
for each of the two study areas. Section numbers are provided to refer to
the section of the report that pertains to each conclusion.
A. Conclusions: Northern High Pressure Oxidant Study
3
1. In the summer, high concentrations of ozone (i.e., >_ 160 yg/m )
in the rural boundary layer and in the eastern portions of
the United States are most often found within high pressure
systems. Sustained periods of high ozone are associated with
macroscale high pressure systems that persist for more than
20 days. (Section 7.2.2)
2. Highest concentrations of ozone were found in the back side
of a high pressure system. A relative minimum is observed
in the front side or near the center. (Section 7.2.3)
3. Locations of maximum and minimum ozone concentrations in a
moving high pressure system correlate with the location of
air having maximum and minimum residence time in that
system. (Section 7.2.3)
4. The air initially in the northeastern quadrant of an east-
ward moving high pressure system has the longest residence
time in that system. (Section 7.2.4)
5. Oxides of nitrogen concentrations in rural areas in the
western section of the study area were apparently too low
to promote the generation of ozone concentration equal to
or greater than the NAAQS. (Section 7.3.1)
6. High ozone concentrations and the frequency of exceeding the
NAAQS for photochemical oxidants are associated with increased
population density (i.e., both increased from west to east).
(Section 7.3.1)
-------
B. Conclusions: Gulf Coast Oxidant Study
1. Ozone concentrations over the Gulf of Mexico usually were
less than those over land. High ozone concentrations (i.e.,
>_ 160 ug/rn-^) that were measured over water or in air flowing
off the Gulf of Mexico were associated with air that had
previously passed over continental sources of pollution.
(Section 8.1.3)
2. Changes in the vertical structure of ozone concentrations
below 3 km are primarily controlled by boundary layer
processes. (Section 8.1.2)
3
3. Elevated ozone concentrations (i.e., >_ 160 yg/m ) are
frequently measured in plumes downwind of potential
ground sources of precursors, i.e., cities, major refineries,
and petrochemical installations. (Section 8.1.3)
4. Upper decile concentrations of ozone are associated with
slow moving air that had passed over high precursor emis-
sion areas and arrived from a nonprevailing wind direction;
lower decile concentrations are associated with faster
moving air, havLng a long over-water fetch with a weak
anticyclonic trajectory. (Sections 8.1.3 and 8.1.4)
-------
2.0 INTRODUCTION
2.1 Background
Since the mid-1960's, surface concentrations of ozone greater than
3
the National Ambient Air Quality Standard (NAAQS) (160 yg/m hourly
average not to be exceeded more than once a year)—'' have been reported
2-9 /
at nonurban sites in many areas of the United States.—— Prior to
this time the range of surface ozone concentrations was considered to
be 40-120 yg/m .— High concentrations of ozone are now being measured
in areas which many consider remote, rural, or "clean" (i.e., devoid
of anthropogenic pollution).
In 1970, in the Mt. Storm area— of West Virginia, high ozone
concentrations were found to occur and persist for several days at a
time. In 1972, it was predicted that the phenomenon was probably wide-
3/ 4/
spread.— In 1973,— it was determined that the phenomenon was wide-
spread over at least a four-State area and was not confined to the
vicinity of Mt. Storm. Stations located in southern West Virginia,
western Maryland, central Ohio, and northwestern Pennsylvania all showed
numerous simultaneous onsets of periods of high ozone. In 1974,- a
study was conducted in Ohio to investigate the relationship between high
rural ozidant levels and urban hydrocarbon control strategies. All
data obtained in the 1974 study— showed strong evidence for the involve-
ment of anthropogenic precursors and urban effluvia in the generation
of the high ozone concentrations in rural areas. As a resuJt of these
investigations, it was postulated that the high concentrations of ozone
(produced by photochemical processes) found in nonurban portions of the
area studied are primarily an air mass characteristic and will occur
when a slow-moving, high atmospheric pressure system passes over the
region.
A series of studies—J conducted by the Air Control Board of
the State of Texas have shown occurrences of concentrations of ozone in
3
excess of 160 yg/m in nonurban areas. Due to the short duration of
the periods of measurement and the sequential mode of measurement at
the several locations reported in these studies, the areal extent of
the region of high ozone concentrations was not determined. Correspond-
ingly, the source of the high concentrations of ozone was not determined.
-------
Faced with the problem of devising strategies to reduce urban
oxidants to concentrations conforming to the NAAQS, the Texas Air Control
Board found that nonurban air entering the cities had ozone concentrations
above the NAAQS. The board, therefore, concluded that strategies applied
to the urban areas would not achieve oxidant levels at or below the NAAQS
and that "In some areas of Texas, it appears likely that the photochemical
oxidant standard would be exceeded even if emissions from human activity
were reduced to zero."—
2•2 Research Objectives
In response to the Environmental Protection Agency's interest in
these areas, RTI planned and conducted a two-part study program: (1)
to determine the change in the concentration of ozone in the center of
a high pressure system as the system moves from an area of low popu-
lation density to an area of high population density; and, (2) to in-
vestigate the areal extent of regions of high ozone concentrations in
the northern gulf coast area of Texas, with particular attention being
given to the determination of whether the high ozone concentrations
extend over several contiguous States.
10
-------
3.0 DESIGN OF STUDY
In order to accomplish the objectives set forth in section 2.2, it
was necessary to design two independent studies that had different objec-
tives but could be conducted concurrently and employed similar measure-
ment systems.
The Northern High Pressure Oxidant Study focused on tracking the
movement of high pressure systems originating in the Alberta-Saskatchewan
region of Canada as the system moves southward over the northern Plains
States, then eastward over the northern Midwest States and the northern
mid-Atlantic States or New England States. During this slow passage
over the United States, the air mass first traverses regions of low
population density and little industrial activity—that is, regions
having small emission densities of hydrocarbon and nitrogen oxide pre-
cursors necessary for the production of the photochemical oxidants. Dur-
ing this initial period, low ozone concentrations in high pressure systems
are anticipated. As the air mass moves eastward, however, population
density, industrial activity, and, consequently, emissions of oxidant
precursors increase.
Primary emphasis for the Gulf Coast Oxidant Study was focused on
the roles and/or contribution of land-sea breeze circulations, the local
emissions of ozone precursors, and transport of ozone and ozone precursors
to concentrations of ozone measured within and downwind of the study areas.
Based on the above considerations, two modes of measurement were
employed to measure the concentration of ozone (0~) , oxides of nitrogen
(NO ), hydrocarbons (HC) and halocarbons (HCX) , and particulates (TSP,
=x +
SO., NH,, NO ): a network of five fixed, rural, ground-level stations
and an instrumented aircraft flying specified patterns. Measurements at
each ground-level station within the network were the same for both
studies; however, the flight patterns were quite different. These are
discussed separately in the following sections. In addition, vertical
profiles of ozone were obtained from ozonesonde releases at Huron, South
Dakota, and at DeRidder, Louisiana.
11
-------
3.1 Northern High Pressure Oxidant Study
3.1.1 Airborne Measurements
Upon a forecast of a high pressure system moving out of the Canadian
prairie provinces into the northern United States and having the potential
to move in a desired manner across the eastern United States, the air-
craft was dispatched from the gulf coast area to Sidney, Montana, arriving
within 24 hours of notification. Beginning there, the aircraft flew a
daily mission with a pattern based on weather information transmitted to
RTI by the weather forecasting service of Murray and Trettel, Inc. ,
Northfield, Illinois. Using this information, RTI then planned the flight
to maximize the data collection in relation to the aircraft and crew
capabilities.
In general, each flight was flown at altitudes between the morning
inversion height and the afternoon mixing height. Normally, this alti-
tude was about 1,520 m (5,000 ft) MSL or about 912 m (3,000 ft) above
ground. Insofar as possible, the flight was flown under visual flight
rules (VFR) to maintain maximum flexibility of flight plans. Depending
upon conditions, takeoff was scheduled to occur after 1030 hours Local
Daylight Time (LDT) and to terminate before 1830 LOT. The aircraft has
a range of about 1,609 km (1,000 mi) or 5.1 hours of flight at 198 mph,
before reaching a minimum of 45 minutes fuel reserve to reach an alternate
destination.
Two modes of flight were used to acquire data in the vertical and the
horizontal. Each day at approximately 1300 LDT, the aircraft was'to be
near the center of the high pressure system. At that time a vertical
profile of ozone concentration was obtained. The aircraft altitude was
then altered to 1,216 m (4,000 ft) MSL and sampled for 3 minutes at that
altitude. The aircraft then ascended to 1,830, 2,135, 2,440, 2,745, and
3,050 m (6,000, 7,000, 8,000, 9,000 and 10,000 ft), sampled for 3 minutes
at each altitude with ascent rates of 152 m (500 ft) per minute between
levels. The aircraft then descended and sampled at these same altitudes.
The flight continued at an altitude of 1525 m (5,000 ft) about 1 hour
after the vertical profile began.
The horizontal transects flown across the high pressure system were
determined by a combination of factors such as the location of the high
12
-------
pressure center relative to the aircraft's overnight position, the track
the system was expected to follow, and the time required to be in position
for the 1300 LDT sounding near the high pressure center. Based on ex-
perience gained through simulated flights in historical weather systems
the following objectives were desired and achieved in most situations:
(1) cross the high as it moves along the population gradient, (2) make
a low pass tie-in with a ground sampling location, (3) make a tie-in
with the ozonesonde location, (4) cross a major portion of the high
pressure system, and (5) be in a good location to do one of the above
on the following day (including becoming airborne in sufficient time to
the vertical profile).
3.1.2 Ground Measurements
To support the airplane data-gathering program, four instrumented
ground stations were established at airports near Wolf Point, Montana;
Creston, Iowa; Bradford, Pennsylvania; and Lewisburg, West Virginia.
These locations are depicted in figure 1. In addition, a 15-station
monitoring network utilizing existing state and local agency stations
was used to provide additional supporting data. These locations are
also shown in figure 1.
Aerometric data, as listed in table 1 in section 4.0, were obtained
at these ground stations. Day-by-day and diurnal behavior of the
pertinent variables were observed. Details regarding the measurement
parameters and sampling schedule are presented in section 4.0.
3.1.3 Ozonesonde Measurements
As an addition to the northern high pressure study, a program of
serial ozonesonde soundings across high pressure systems which passed
through the northern Great Plains was conducted. High pressure systems,
which RTI identified for flight analysis, were investigated with re-
leases from Huron, South Dakota. The objective of the program was to
document the vertical ozone distribution and its changes, especially
these changes that could be attributable to intrusion of stratospheric
ozone into the troposphere across discontinuities in the tropopause.
Secondarily, the influence of stable layers in the planetary boundary
layer upon the diurnal ozone changes at the ground was sought.
13
-------
o
s
QJ
c
c
o
•rl
4->
CO
4J
CO
§
O
S-l
bO
CO
c
o
rt
u
o
cu
4J
•rl
00
•H
14
-------
3.2 Gulf Coast Oxidant Study
3.2.1 Airborne Measurements
The airborne operations in the gulf coast area were designed to
document the areal extent of high ozone concentrations in the gulf coast
area of Texas and to investigate the possible influence of four factors
that could affect ground level ozone concentrations. These are: (1) a
sea breeze, (2) a petrochemical complex, (3) a coastal areal ozone dis-
tribution, and (4) a vertical ozone distribution. Flight protocol re-
quired that crew and aircraft safety take precedence at all times. All
flight patterns in the gulf coastal area were subordinate to aircraft
operations associated with the northern high pressure oxidant study de-
scribed in section 3.1. Area survey flight plans were normally scheduled
for two consecutive days of operation. The flight patterns designed to
be flown in the gulf coast area and the type of information sought from
each type of flight are discussed below.
(1) Sea Breeze Flight
A flight pattern was designed to investigate the role that
the land-sea breeze circulation may play in contributing
to high ozone concentrations during the time of onshore
winds. The occurrence of high ozone concentrations near
the sea surface as compared to concentrations at higher
altitudes and the change in concentration across the sea
breeze front were of primary interest. The flight pattern
shown in figure 2 was designed to be flown twice daily
under the appropriate conditions of high ozone concentra-
tions (i.e., > 160 yg/m^ hourly average at DeRidder,
Louisiana, or Nederland, Texas, on the previous day) and
when an identifiable sea breeze circulation could be ex-
pected. The flight track in the morning extended 321 km
(200 mi) to sea at an altitude of 1,220 m (4,000 ft) MSL,
well above the onshore flow. The return flight was
at 244 m (800 ft) MSL, within the onshore component of the
sea breeze. The afternoon flight was flown 321 km
(200 mi) to sea at an altitude of 610 m (2,000 ft) MSL
with an overwater vertical profile to 3,050 m (10,000 ft)
as shown in figure 2.
(2) Petrochemical Complex Downwind Plume Flight
The extensive petrochemical complex along the Texas gulf
coast provides a major source of ozone precursor materials,
specifically hydrocarbons and oxides of nitrogen. The
downwind plume flight pattern was designed to investigate
15
-------
a
4J
•a
•rl
0)
N
QJ
0)
CO
0)
00
•H
16
-------
and document the changing ozone and ozone precursor con-
centrations downwind of the complex.
The flight pattern, shaped like a parallelogram, is
shown schematically in figure 3 and was flown downwind
of the urban-industrial complex of the Freeport to
Beaumont area. The flight plan was scheduled to be flown
on two successive days; however, the orientation varied
with the mean wind. The altitude for this pattern was
610 to 915 m (2,000 to 3,000 ft) MSL under VFR conditions.
(3) Coastal Areal Survey Flights
The areal extent of ozone concentration across the States
bordering the Gulf of Mexico is largely unknown, especially
in interior (noncoastal) areas. A prime objective of this
study was to document the areal distribution of ozone and
its precursors. Survey flights were designed to investi-
gate the possibility of long-range transport of ozone and
its precursors into the study area from Southeastern States,
where high concentration of nonurban ozone have been docu-
mented, or from the Florida peninsula where occasional
high ozone episodes have been reported. These flight tracks
are shown in figures 4, 5, and 6.
Several missions were flown in conjunction with an EPA
operated aircraft in October. This was done in order to
extend the distance of the survey within the same time frame
and also served as a comparative check of the aircraft mea-
surement system. Included among the flights were a north-
south survey, as shown in figure 5, a land-sea survey as
shown in figure 6, and a superimposed set of "double-box"
patterns about the Nederland, Texas, area to assess the in-
fluence of the petrochemical complex located there. This
flight is diagrammed in figure 7.
(4) Vertical Profile Flights
The possibility of high ozone concentrations resulting
from intrusion of stratospheric ozone into the troposphere
and being transported to the ground was investigated using
a vertical profile flight pattern.
The vertical distribution of ozbne was examined by a
series of flights with measurements made at prespecified
altitudes to 6,100 m (20,000 ft). The airplane ascent rate
was 152 m (500 ft) per minute. Ozone measurements were
recorded at 305 m (1,000 ft) increments from the ground
to 1,830 m (6,000 ft) and in 610 m (2,000 ft) increments
from 1,830 m (6,000 ft) to 6,100 m (20,000 ft).
17
-------
18
-------
•a
•H
g
CO
•it
01
19
-------
20
-------
8
rH
ft
n)
0)
co
9
VD
0)
I
•H
21
-------
— r*-* — RTI Navajo
EPA B-26
•100 mi
Figure 7. Double-box flight plan.
22
-------
Each profile flight required about 2 hours for completion.
Because of restrictions on altitudes above 3,660 m (12,000 ft)
in the DeRidder area, some profiles were flown near the Jasper,
Texas, area about 31 km (50 mi) west.
In addition, several vertical ozone distribution flights were
carried out in support of the ozonesonde releases being con-
ducted in the gulf coast area.
3.2.2 Ground Measurements
A single ground station was located at the DeRidder, Louisiana,
airport for the gulf coast study. This station provided a continuity
of aerometric information which, with the data obtained from the air-
craft, was expected to provide insight into the areal extent of high
ozone concentrations in the area. Day-by-day and diurnal behavior of
the measured pollutants were observed. Details regarding the measure-
ment parameters and sampling schedule are presented in section 4.0. In
addition, ozone data from several urban stations operated by the Texas
Air Control Board in the general study area were used by RTI in this
investigation. These data are further described in sections 4.0 and
5.0.
3.2.3 Ozonesonde Measurements
An ozonesonde program was conducted to provide serial soundings from
the DeRidder, Louisiana station in October 1975, This program was de-
signed to expand the vertical dimension of the survey through the tropo-
sphere to the stratosphere and to provide another comparison of in-
situ ozone measurements to altitudes of 6 km (~20,000 ft).
3.3 Analysis Procedure
The data analysis undertaken for the two studies described in this
report have many elements in common; however, because of the vast differ-
ence in objectives, the analysis procedures for the two studies are
discussed separately.
3.3.1 Northern High Pressure Oxidant Study
The Northern High Pressure Oxidant Study was designed to examine the
change in ozone concentration within a high pressure system as that high
pressure system moves from relatively large regions of low-density emis-
sions of precursor materials into regions of successively greater den-
sities of emissions of precursors.
23
-------
The measurements made on successive days during airplane flights
through the high pressure system should provide a data set in a quasi-
Lagrangian coordinate system. Time-space sections of these measurements,
in particular the ascents near the location of the day-to-day highest
pressure, were analyzed to determine a measure of the change in the
system's ozone concentration. Since ozone production within an air mass
requires time, it may be that high concentrations can result without
a progressive increase in the injected precursors. Accordingly, high
pressure systems that remained over a region of low density of emissions
of precursors were also analyzed.
Data from horizontal flights through the high pressure system were
analyzed for discontinuities in systematic changes of, and symmetry of,
ozone and ozone precursor concentrations to define the areal extent of
regions of high ozone concentration. In addition, patterns suggested
by these data were expected to offer clues to the processes taking place
within the high pressure system that lead to the large areal coverage of
high ozone concentrations.
3.3.2 Gulf Coast Oxidant Study
Concentrations of ozone and ozone precursor ratios along the flight
paths and from the airplane ascents were examined for gradients and dis-
continuities. An attempt was made to relate departures from systematic
changes to large-scale topographical features of the underlying surface
of the earth, and/or to changes in the meterological conditions. Par-
ticular attention was given to nonsystematic changes evident in data from
corresponding portions of both outbound and inbound (relative to the base
station) flights or in corresponding regions (defined by altitude or
by thermal structure) of ascents at the two extremities of the flight
path.
24
-------
4.0 FIELD MEASUREMENT PROGRAM
This section gives a description of the protocol for the establish-
ment and operation of the ground station network, aircraft measurement
program, and gas chromatographic analyses of selected hydrocarbons, acet-
ylene, and halogenated compounds. Site locations, instrumentation, cal-
ibration, maintenance, data acquisition, and data processing procedures
employed in each of the above-mentioned areas are described.
4.1 Ground Stations
4.1.1 Sampling Protocol
The field measurement program was designed to measure continuously
the ambient concentration of ozone (0_) and nitrogen dioxide (NO ) at
four manned stations located in rural areas in the States of Pennsylvania,
Iowa, Montana, and Louisiana and ozone at an unmanned station in West
Virginia. In addition to the continuous measurements for 0 and NO ,
discrete grab samples were collected in Tedlar bags and on solid adsorp-
tion traps at specified intervals at four of the five rural sites and
during aircraft flights. The samples were shipped to RTI for analysis of
selected hydrocarbons, acetylene, and halogenated compounds by gas chro-
matography. Total suspended particulate (TSP) samples were also collected
by the high volume sampler method at four sites on a daily basis for sub-
sequent EPA analysis of sulfate, nitrate, and ammonium ions. Measurement
parameters for each rural station are summarized in table 1.
Ozone was continuously measured at approximately 15 monitoring sta-
tions operated by State, local, or private industry. These data were
provided to RTI by the Environmental Protection Agency. These State/
local stations were located at: Colstrip, Montana; Indianapolis, Indiana;
Columbus, Ohio; Pittsburgh, Pennsylvania; Northwest Houston (Aldine),
Nederland, Austin, Corpus Christi, and Port O'Connor, Texas; Pensacola,
Florida; Omaha, Nebraska; Des Moines and Cedar Rapids, Iowa; Kansas City,
Missouri; and Poynette, Wisconsin. Figure 1 of chapter 3 showed the
selected locations of fixed ground stations for the study (i.e., both
rural and urban stations). Valid data from those sites that were re~
ceived as of November 30, 1975, were employed in the present analysis.
25
-------
Table 1. Pollutants measured at rural monitoring stations
Station Pollutants*
Bradford, Pennsylvania 03, N02, HC and HCX by GC, so|~, N03~, NH4+
Creston, Iowa 0}, N02, HC and HCX by GC, S0^~, N03~, NH4+
Wolf Point, Montana 03, N02, HC and HCX by GC, S0^~, N03~, NH4+
DeRidder, Louisiana 03, N02, HC and HCX by GC, S0^~, N03~, NH4+
Lewisburg, West Virginia 0_
_ __ _ _
Hi Vol filters returned to EPA for analysis of SO, , N03 , and NH, .
4.1.2 Siting Considerations and Description of Monitoring Stations
The principal criteria for the selection of the rural monitoring
station locations were: (1) that the location be free of natural and
manmade obstructions to air movement, (2) that it be removed from local
sources of ozone and ozone precursors, and (3) that it be readily acces-
sible by aircraft in order to facilitate timely maintenance and routine
calibration schedules. The locations selected meet these criteria.
Four rural ground stations were utilized in the ozone study in the
north-central and north-east regions of the United States. These stations
were situated approximately 1,200 km (750 mi) apart and are frequently
located in the path of high pressure systems that enter the United States
from Canada, move across the Plains towards the Ohio Valley and New York
State. One rural ground station was located in the southern part of
Louisiana and served as the base of operations in the Gulf Coast area.
A. Bradford, Pennsylvania, Monitoring Station
The monitoring station in Pennsylvania was located at the Bradford
Regional Airport situated approximately 26 km (16 mi) south of Bradford.
At an elevation of 653 m (2,143 ft) above mean sea level, the airport was
well exposed to air flow frora all compass points. The 0 and NO analyzer?
and associated equipment were housed in an environmentally controlled
2.4 m x 4.8 m (8 ft x 16 ft) Cortez van. The site is illustrated in
26
-------
figure 8. An Interior view of the stations showing the analyzers and
associated equipment is presented in figure 9. The interiors of the
other three RTI rural stations were similar to the Bradford station.
B. Creston, Iowa, Monitoring Station
The Creston Municipal Airport, located approximately 5 km (3 mi)
south of Creston, Iowa, was the site of a second rural monitoring station.
The location was 394 m (1,293 ft) above mean sea level and was well ex-
posed. Air quality analyzers and associated equipment were located in an
environmentally controlled 2.4 m x 4.8 m (8 ft x 16 ft) Avion trailer
which was located approximately 185 m (1,600 ft) away from the runway.
An aerial and ground level view of the site are shown in figures 10 and 11.
C. Wolf Point. Montana. Monitoring Station
The northernmost monitoring station was located at the Wolf Point
International Airport which is situated 3.2 km (2 mi) southeast of Wolf
Point, Montana. The airport is located approximately 605 m (1,985 ft)
above mean sea level. Exposure is excellent from all directions. Instru-
mentation and associated equipment were housed in an environmentally con-
trolled 2.4 m x 4.8 m (8 ft x 16 ft) Avion trailer located 92 m (100 yd)
southeast of a small terminal building. Aerial and ground-level views of
the site are shown in figures 12 and 13. Although the airport is well"-
situated for an air monitoring site, the facilities are only marginally
adequate for the aircraft. Therefore, all flight operations in the area
were flown out of the Sidney-Richland Airport which is located 64 km (40 mi)
southeast of Wolf Point, Montana.
D. Lewisburg, West Virginia, Monitoring Station
The fourth monitoring site for the study in the north-central and
north-east regions of the United States was a cooperative station provided
by Bendix Corporation at the Greenbriar Valley Airport, near Lewisburg,
West Virginia. Situated at an elevation of 702 m (2,301 ft) above mean
sea level, the Lewisburg site is well exposed and provided ozone data to
the south of the anticipated path of high pressure systems. Aerial and
ground-level views of the site are shown in figures 14 and 15.
27
-------
Figure 8. Air monitoring station, Bradford Regional Airport.
Figure 9. Interior view of air monitoring station.
28
-------
Figure 10. Aerial view of Creston, Iowa (Crest on Mu.iicj
Airport) site.
Flgiirp II. Air monitoring station, Creston Muni, i f..-j I 'irport.
29
-------
Figure 12. Aerial view of Wolf Point, Montana
(Wolf Point International Airport) site.
Figure 13. Air monitoring station, Wolf Point
International airport.
30
-------
Figure 14. Aerial view of Lewisburg, West Virginia
(Lewisburg Airport) site.
Figure 15. Air monitoring station, Lewisburg airport.
31
-------
E. DeRidder, Louisiana, Monitoring Station
The Beauregard Parish Airport, situated 8 km (5 mi) southwest of
DeRidder, Louisiana, served as the base station for gulf coast operations
during the summer study. The elevation of the airport is 62 m (203 ft)
above mean sea level and exposure is excellent from all directions. In-
strumentation and associated equipment were housed in the RTI Semi-
Mobile Environmental Monitoring Laboratory. The leased aircraft used in
this program for aerial measurements was based here. Aerial and ground-
i.
level views of the site are shown in figures 16 and 17. Beauregard Parish
Airport is near a large air pollution source (a paper mill), which is
situated approximately 3.2 km (2 mi) to the northwest. The prevailing
wind direction during the summer months is from the southwest with flow
out of the northwest occurring less than 5 percent of the time. Wind
speed and direction were monitored continuously to identify periods of
time when the sampling site might be under the influence of the paper
mill plumes. No other airport sites examined in the Texas^Louisiana
area were acceptable from the standpoint of remoteness from oil refineries
(sources of hydrocarbons) or logistics [i.e., within 64 km (40 mi) of the
Gulf Coast] .
4.1.3 Air Quality Measurements
4.1.3.1 Instrumentation
Ambient ozone concentrations were measured at all rural ground sta-
tions using the Bendix Model 8002 Chemiluminescent Ozone Analyzer or an
equivalent instrument. The principle of operation of this instrument is
based on the gas-phase Chemiluminescent reaction between ethylene and
1
ozone. The reliability, stability, specificity, and precision of ozone
measurements by this technique have been adequately demonstrated and de-
scribed in the literature.—
Nitric oxide and nitrogen dioxide concentrations were measured at
four of the five rural stations using the Bendix Model 8101-B NO-NO-NO
£ X
Analyzer. The principle of operation of this instrument is based on the
gas-phase Chemiluminescent reaction between NO and 0 . Measurement of
the NO- concentration by this method requires that NO be reduced to NO,
32
-------
Figure 16. Aerial view of DeRidder, Louisiana
(Beauregard Parish Airport) site.
Figure 17. Air monitoring station, Beauregard
Parish Airport.
33
-------
which then reacts with 0,.. The sum of the initial NO measurement plus
the NO produced by the reduction of NO is the nitrogen oxides (NO )
*• X
measurement. Electronic subtraction of the NO measurement from the NO
x
measurement gives the N0« concentration.
Detailed hydrocarbon analyses of discrete grab samples collected
during aircraft flights and at the four manned ground stations were per-
formed by use of a modified Perkin-Elmer Model 900 gas chromatograph
coupled to a Hewlett-Packard Model 2100A computer programmed for peak
area analysis and quantification. Twelve nonmethane hydrocarbons were
routinely analyzed and are listed below:
(1) ethane and ethylene
(2) acetylene
(3) propane
(4) propylene
(5) isobutane
(6) n-butane
(7) 1-butene
(8) isopentane
(9) cyclopentane
(10) n-pentane
(11) toluene
(12) o-xylene
The first ten compounds were separated on a 1.5 m x 0.15 cm i.d. Durapak
n-Octane column operated at room temperature. The two aromatic compounds
were separated on a 1.5 m x 0.15 cm i.d. stainless steel column packed
with GP 5% SP-1200/5% Bentone 34 on 100/120 Supelcoport. Samples for
these analyses were collected in the field in Tedlar bags and transported
to the Research Triangle Institute campus for analysis.
Supplemental acetylene samples were collected at four ground stations
and during aircraft flights using special traps containing silanized mo-
lecular sieve type 5A. Acetylene is quantitatively collected on the
molecular sieve solid adsorbent, desorbed by heat, and separated from
other hydrocarbons on an alumina-packed column at 175° C. Analyses were
performed on a modified Beckman 6800 Air Quality Chromatograph. This
technique was developed for the Environmental Protection Agency by Beck-
18/
man Instruments, Inc., and is described in EPA Report 650/2-74-056.—
34
-------
Concentrations of selected halocarbons were determined from bag
samples collected at four ground stations and during aircraft flights
f O
using a Perkin-Elmer Model 900 gas chromatograph equipped with a Ni
electron-capture detector. The following halogenated compounds were
analyzed: Freon-11, carbon tetrachloride, 1,1,1,-trichloroethane, and
tetrachloro-ethylene. All four of these compounds are heavily used in
industry and can be definitely labeled as anthropogenic pollutants,
having no known natural sources. Separation of these compounds was made
on a 1.5 m, 2 mm i.d. glass column packed with Chromosorb-W and coated
with 10% DC-200.
4.1.3.2 Instrument Calibration and Maintenance
Dynamic calibration techniques were used to calibrate each analyzer
at 3-week intervals during the approximately 100-day period of field op-
erations. These techniques are outlined in table 2. Data obtained from
these calibrations were used to provide transfer equations for converting
3
analyzer voltage outputs to pollutant concentration (yg/m ).
A brief description of the calibration techniques listed in table 2
is given below.
Ozone-Ultraviolet ozone generator referenced to the neutral-buffered
potassium iodide (KI) method
Dynamic calibration of the ozone analyzers was accomplished by use of
an ultraviolet ozone generator. The oz'one generator consists of a 20 cm
(8 in) length mercury vapor lamp which irradiates a 16 mm (5/8 in) diameter
Table 2. Calibration techniques
Pollutant Calibration Technique
Ozone Ultraviolet ozone generator referenced
to the neutral buffered potassium
iodide method (Federal Reference Method)
Nitric oxide/nitrogen dioxide Gas phase titration technique
Selected hydrocarbons/freons/ Mixtures prepared from pure hydrocarbons
acetylene and dilution of certified mixtures
35
-------
quartz tube through which clean (compressed) air flows at 5 &/min.
Variable ozone concentrations over the measurement range are generated by
variable shielding of the mercury vapor lamp. Although the ultraviolet
ozone generator has been shown to be quite stable and reproducible, the
neutral-buffered KI method was used as the reference method. The ozone
concentration of each calibration point was verified by the neutral-buf-
hod.—
oxide /nitrogen dioxide — gas phase titration technique
fered KI method.—
19 /
The gas phase titration technique developed by Hodgeson et al —
of the Environmental Protection Agency was used for dynamic calibration
of the chemiluminescent NO-NO -NO analyzers. The technique is based on
X 2.
the application of the rapid gas-phase reaction between nitric oxide and
ozone to produce a stoichilometric quantity of nitrogen dioxide. A cer-
tified tank of nitric oxide in nitrogen (of an approximate concentration
of 50 ppm by volume) was diluted with zero air to provide NO concentra-
tions in the range of 20 to 940 yg/m . Nitrogen dioxide concentrations
are produced by the reaction of NO with ozone. Primary calibration of
the NO in nitrogen concentration was accomplished by use of the gas
19 /
phase titration technique. — The NO concentration of each cylinder was
determined prior to going to the field, once per month in the field, and
at the conclusion of the study.
Selected hiidrooca>bons3 freons, and acetylene
Hydrocarbon standards used to calibrate the Perkin-Elmer flame
ionization detector gas chromatograph were obtained from Scott Research
Laboratories as mixtures of hydrocarbons in hydrocarbon-free air. The
concentration of these mixtures was certified to be accurate within ± 2
percent. Based on these concentrations, response factors were calculated
by the computer system for subsequent use in the analysis routine. Com-
parison of the peak heig'it, retention times, and peak area for hydrocar-
bons in standard mixtures were used to identify and quantify the various
hydrocarbons collected in grab samples. A Perkin-Elmer electron-capture
detector gas chromatograph was used for analysis of halocarbons. Halo-
carbon standards for calibration of the gas chromatograph. were prepared
36
-------
by use of permeation tube devices and/or by quantitative dilution of pure
liquids by injection of aliquots of liquid into Tedlar bags containing a
known volume of hydrocarbon-free air.
A detailed discussion of all the calibration methods and procedures
utilized during this program is given in appendix A.
Routine maintenance (excluding emergency trips for instrument fail-
ure) was performed by the station operator as required by each individual
analyzer. Routine maintenance was performed on analyzers at the unmanned
Lewisburg site during the regular dynamic calibration period. When fail-
ures occurred, an analyzer was brought back on-line in the most expedient
manner by substitution or repair. A complete record of operational
status (i.e., operational, repairs, maintenance, calibration, inoperative,
etc.) was maintained for each analyzer throughout the duration of the
project. This information was used in the data validation process.
4.1.4 Data Acquisition and Data Processing
Ozone, NO, and NO concentration data at each station were recorded
in analog form on strip chart recorders and in digital form on magnetic
tape recorders. The strip chart recorders were used for backup in case
of failure of the magnetic tape unit and, also, as a real-time record for
assessment of the operational status of each analyzer. Primary data
storage was on magnetic tape.
Since five identical data acquisition systems were not available,
two different systems were used to record data at the five rural sta-
tions. A Metrodata DL-630 data logger was used at the DeRidder station,
and Westinghouse Pulse-0-Matic recording systems were used at Bradford,
Creston, Wolf Point, and Lewisburg. A brief description of each data
acquisition system is presented in the following paragraphs.
The Metrodata Model DL-630 data system is a complete data-acquisi-
tion system capable of recording up to 40 channels of analog data plus a
time code, station identification, and manual data entry on magnetic tape.
Commands are locally selectable via front-panel, lighted push buttons.
Channel selection allows the observation of the current signal of any
channel as an electronic three-digit display on the front panel in real*-
37
-------
time. The scan rate is 40 channels/second, and data are recorded on
magnetic tape at 5-minute intervals.
The Westinghouse Pulse-0-Matic magnetic tape data acquisition system
integrates the signal from t:he air quality analyzer for a 15-minute inter-
val, thus providing a true 15-minute average. A. battery backup unit is
included in the recorder to preserve the time information on magnetic
tape in the event, of a powe~: failure.
Magnetic tapes containing field data from the Metrodata and Westing-
house data acquisition systems were returned to Research Triangle Institute
for processing on a weekly basis. The data manipulation required to re-
trieve the data stored on a magnetic tape consists basically of two
phases :
(1) Translation of the tape to a form compatible with available
data-processing equipment.
(2) Processing of the data on a computer to obtain pollutant
concentrations in units of
The end result was a hard-copy printout, which was then made avail-
able for inspection and validation. This printout consisted of the date,
station identification, hourly average of 5- or 15-minute readings, and
3
24-hour average concentrations in ug/m for each pollutant.
In order to obtain a printout of data that contains all the informa-
tion, certain supplementary data must be supplied to the computer. These
data include a listing of times when the instruments were inoperative or
not functioning properly and linear best-fit equations relating the volt-~
age output of the instrument to the concentration of the pollutant being
measured. The times for instruments being inoperative were obtained from
operator logs, calibration log sheets, quality control charts, examination
of preliminary computer runs, and strip chart records. Linear best-fit
equations were derived from data obtained during calibration — the known
input gas concentrations and the resulting voltage output from the instru-
ment. A regression analysis was performed on these points to obtain a
best-fit equation characterizing the instrument's response.
4.1.5 Data Validation and Quality Control
In order to achieve and maintain a high level of confidence, in air
38
-------
quality data, it was essential to routinely monitor critical instrument
parameters and to maintain appropriate records. Quality control for the
summer oxidant study included procedures for verification of calibration
procedures, standards, and operating procedures, performance of dynamic
calibrations at specified intervals, and maintenance of adequate records
describing instrument performance as well as thorough training of field
operators.
Calibration data, as well as daily zero and span information were
examined for excessive zero and span drift. When zero drift exceeds ± 1
percent of full scale per 24-hour period, the data of the preceding 24-
hour period was considered to be of questionable validity and was invali-
dated. Span drift was determined on a daily basis (ozone instruments)
and from multipoint calibration data every 2 to 3 weeks. Span drift
exceeding ± 1 percent per 24-hour period (ozone instruments only) or ± 3
percent per 2 weeks constituted grounds for invalidation of data. To
verify accuracy of the data acquisition system, a constant voltage data
input standard was recorded every 5 minutes on the Metrodata DL-630 data
system in conjunction with the air quality data.
At the completion of each computer program set, the processed data
were compared with strip-chart data for randomly selected periods. A
typical check included comparison of computer-processed data and strip-
chart data for six selected 1-hour averages per day. The recorded data
on strip charts were also edited for signs of equipment malfunctions,
excessive pollutant levels, or unusual diurnal patterns.
4.2 Airborne Measurements
4.2.1 Airborne Measurements System Description
4.2.1.1 Aircraft
A twin-engine light aircraft was instrumented to measure ozone,
oxides of nitrogen, condensation nuclei, temperature, dew point, and
pressure, and was also equipped to collect: grab samples for EC analysis,
selective filter samples for acetylene analysis, and high-volume filter
samples for sulfate and nitrate analysis. The aircraft used was a Piper
PA-31-350 Navajo Chieftain and is shown in figure 18 with the sampling
39
-------
Instrumented aircraft.
probe am-'1 -.;!;" te sampler in place. The aircraft was operated at a
typical ;..'<-«?,! of 180 nph (280 km/hr) and with an operational time
of 3.5 h-i . ,'i.tii 45 minutes of fuel reserve, a nominal range of 1,000
km was pra* ( i. •. . '!'!,,.=> operational altitude was 22,000 feet (b,700 in) with
a nominal <. i i r!> r,a-- of 1,000 ft/min (305 m/min). It was equipped with
instrumentation aid sur\ival gear for night and over-water operations.
Communication -H'J navigation equipment included Dual VOR and VHF Communi-
cation, UME, AD!7, Radar Altimeter, Weather Radar and Transponder.
The air sample intake system consisted basically of a 2.5-cm I.D.
Teflon tube- extending from approximately 60 cm in front of the nose of
the aircraft ;o a stagnation type sample manifold on board the aircraft
40
-------
I I Forward
' J Bulkheads
Data System
and
Sample Intake Tube
2.5 ca 1.1). Teflon
Sample Probe
0, Analyser
Exhaust
Sample Manifold
Figure 19. Functional layout of airborne sampling system.
as shown in figure 19. The Teflon tube was inserted in a 3-cm I.D. steel
tube mounted to the aircraft frame. A Pitot tube, a total temperature
probe, and a high-volume particulate sampler were mounted to the sample
probe.
The sample manifold was constructed of aluminum and coated internally
with heat-cured Teflon. The manifold was designed with an inlet diverg-
ing diffuser section to allow deceleration of flow to a more controllable
velocity, typically in the range of 1-2 m/sec.
Analyzer sample lines, temperature and dew point probes are situated
along the main body of the manifold, an adequate distance from the dif-
fuser, to allow unobstructed, isentroplc flow of the air sample. Manifold
velocity, monitored with a hot wire anemometer, was adjusted with an exit
damper for variations in cabin pressure, altitude, and aircraft speed. A
minimum flow velocity of 1 m/sec was maintained to insure a relatively
rapid air sample exchange. In addition, the exhaust end of the manifold
was designed to minimize the possibility of exhaust contamination from
41
-------
the aircraft. A detailed description of the sample probe and manifold
design is presented in appendix C.
4.2.1.2 Measurement Sy s t em
A block diagram of the aircraft air quality measurement system is
shown in figure 20. Parameters continuously measured (directly or in-
directly) were ozone, nitric oxide, nitrogen oxides, condensation nuclei,
ambient air temperature, manifold temperature, dew point, ambient and
manifold pressure, altitude, air speed and time. Samples were collected
for laboratory analysis of hydrocarbons, halocarbons, acetylene, sulfates,
and nitrates.
The physical position of the instruments and supporting equipment
inside the aircraft is shown in figure 21. The instruments and equipment
racks were mounted with Aeroflex steel rope shock absorbers bolted in
place to the aircraft main frame.
The power supply system consisted of 28 Vdc to 115 Vac inverters
that operated off of the aircraft Vdc power source. Two circuit breakers
rated at 40 amps each were mounted in the pilot compartment along with an
ammeter. One of the inverters was a surge-type which was used for the
larger pumps. The system was wired such that it automatically switched
over from aircraft power to ground power when external 115 Vac was con-
nected. In addition, two 42 ampere-hour lead acid batteries provided
power up to 2 hours when external power was not immediately available
or impractical.
4.2.1.3 Instrumentat ion
Ozone was measured with a Bendix Model 8002 gas phase chemilumine-
scent ozone analy2:er, operated continuously on the 0.2 ppm range. C. P.
grade ethylene support gas for the analyzer was supplied from a size 3A
gas cylinder. The instrument exhaust was routed through plastic tubing
and dumped overboard through a bulkhead panel, to the rear and underneath
the aircraft.
Oxides of nitrogen were initially measured with a Bendix gas phase
chemiluminescent NO-NO-NO analyzer, Model 8101-B. The instrument was
2. X
operated in a cyclic mode with both the NO and N0? ranges set to 0.5 ppm
42
-------
L 1
&
0
U]
C
a
w
J
— '
i
^
j.
S
r
t
o
u
w
H
0
a-
l
a
•«
JMantfol
Total press.
transducer
v ]
A
S
a
r
2
r-4
4
1
-J U
*
u
•as3
•H ej A
vt S.1-
to
u
II
g S
s^r
0 g
M
u i
OJ
j
tlM
rdh-
Diff.
press . L_
transducer
1]
0
14
a
u
u
fl
4J
n
^
»
•H
i.
ErH
§
-rt o
| •
5 M
4J -H
j a
i
o
p
u p
i-t V
ti N
44 «H
0 g
CD O
at
•a U
x
0
Is
i/1 O
(-1 \O
^
,S
at
id
01
*j
u
1
-H
1
!4Vdc o ev
lattery backup ~^° —
Urcraft 28Vdc
On a
- 8-
,«
•0 0
*> -I!
n >
L
•5-S.3 M
U J 0 .H
« u u rt
1
S
CO
U t,
s-s-s .H
25 g n)
cn o y
2 *-»
cd
O
-^ M J^
•H U T3 *H
M <0 M M
W O CJ
2
s.
M
60
•H
4rf
O
0
rH
PQ
O
CM
01
M
a
fn
43
-------
£
Sarple _
Intake
0,, NO
3' x —
Analyzers
'-lot uo-Filot
Sample
Manifold
Strip Chart *
V-ecorJeio
Instrument
and
Equipment
Observer
I
X
Gas
- __^O
CylinJer (~\ O
R?.cU ^-^ V^
k
Grab
Simple
Storage
~ata Recording
— System
"agnetic Tape
Seek
Instrument
Operator
Povei
— Inverters
Sack-up Batteries
Figure 21. Diagram of physical layout.
full scale. Oxygen support gas for the analyzer was supplied from a size
3A gas cylinder. Near the midpoint of the program, the Bendix NO ana-
lyzer was replaced with TECO Model 14-B chemiluminescent NO-NO-NO
£ X
analyzer. This unit had the advantage of a more sensitive operating
range, 0-0.1 ppm. The TECO instrument was operated continuously in the
NO mode.
Condensation nuclei were measured with an Environment One, Model
Rich 100, C.N. Counter. According to manufacturer specifications, the
44
-------
unit is capable of counting particles 0.0016 microns and larger in diam-
eter, with a maximum concentration of 300 x 10 particles/cc and with a
repeatability of ± 3 percent of full scale on all linear ranges.
Environmental characterization of the air being sampled included
temperature and dew point measurements inside the sample manifold and
measurements of the adiabatic stagnation temperature of the air relative
to the moving aircraft. Manifold air temperature and dew point were
monitored with a Yellow Springs Instrument Company Model 91 hygrometer.
The temperature sensor, a 0.635-cm diameter bead thermistor, and the dew
point sensor, a 0.9-cm diameter x 5.5-cm length lithium chloride probe,
were mounted inside the sample manifold, downstream from the analyzer's
sampling lines.
Ambient temperature measurements were made with a YSI bead thermis-
tor, Type 44202, mounted in a total temperature probe, positioned in
front of the aircraft, mounted to the sample probe. Design considerations
for the total temperature probe are given in appendix C.
The Pitot tube was mounted on the sample probe a sufficient distance
in front of the aircraft in order to eliminate erroneous measurements
caused by aerodynamic distortion by the aircraft. The probe provided
continuous measurements of the ambient static pressure, altitude, indi^
cated air speed and Mach number.
A total pressure sensor was also used to continuously monitor mani-
fold pressure, providing an additional measurement of the sample environ-
ment at the analyzer intake.
Bag samples for subsequent hydrocarbon analysis were collected with
the apparatus shown in figure 22. Sample air was pumped from the mani-
fold, through 0.635 mm I.D. Teflon tubing, through a manganese dioxide
catalytic converter (for the purpose of converting ozone present in the
sample to oxygen), into a 5 liter Tedlar bag. The bags were mated to the
sampling system with stainless steel quick disconnect fittings.
A similar system was used to selectively trap acetylene from the
sample air (fig. 23). A trap containing silanized molecular sieve was
installed with stainless steel quick disconnect fittings between a metal
bellows pump and the manifold. Flow rate was set for each filter by means
45
-------
From Sample
Manifold
3.175 ran I.T\
Teflon Tubing
Quick Disconnect
Connectors
Mn02/Glass Wool
Packed in S.S. Tube
5 Liter
Tedlar Bag
Metal Bellows
M.B. 41 with
Teflon Coated Bellows
Figure 22. Diagram of grab sampling system for
hydrocarbon sample collection.
Magnesium
Perchlorate
Scrubber
Glass Rotameter
with Glass Float
Float
Stanple
Manifold
3.175 mm I.D.
Teflon Tubing
SELECTIVE FILTER
Silanized
Molecular Sieves
with Quick Disconnect
Connectors
Exhaust to cabin
Metal Bellows
M.B. 41
Needle
Value
Figure 23. Diagram of selective filter sampling system.
46
-------
1/4" Copper Tubing
to Differential Pressure
Sensor
Intake
Filter
Support
(rear view)
Figure 24. Sketch of airborne high volume sampler.
of a rotameter and needle value in series with the filter. A scrubber
containing magnesium perchlorate removed moisture from the air sample
prior to filter exposure. The filters were sealed prior to and immediate-
ly after exposure.
A high-volume filter type of sampler was designed to collect air-
borne particulates for sulfate and nitrate analysis. A probe was de-
signed, which utilized aerodynamic pressure to cause a flow of approxi-
3
mately 35 cfm (1 m /min) through a standard high-volume glass filter
which was trimmed to fit a circular holder approximately 18 cm (7 in) in
diameter. A shutter was incorporated into the vent so that the flow
through the filter was allowed only while in flight at altitude, not dur-^
ing takeoffs, landings or climbs or descents. A sketch of the unit is
shown in figure 24. Design details are given in appendix C.
4.2.1.4 Data Acquisition System
The data acquisition system, shown in the overall system diagram,
figure 20, consisted of a Monitor Labs Data Logger, System 9400, a Cipher,
47
-------
Table 3. List of data channels
Channel
No.
Time
2 x 10
digit words
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
Instrument
Channel
No.
Instrument
Day:Hour:Min:Sec
Manual data entry
Ozone 15
NO 16
N02 17
NOX 18
Condensation nuclei 19
Manifold temperature 20
Dew point 21
Manifold flow rate 22
Spare 23
Spare 24
Total temperature 25
Manifold pressure 26
Static pressure 27
Differential presure 28
Spare 29
Spare
Spare
Selective filter event
Grab sample event
Sulfate sampler timer
Bridge volt (ambient temp)
Spare
Spare
Spare
Spare
Spare
Spare
Short
10 mV (recorder span)
1.0 V (recorder span)
Model 85H magnetic tape recorder, a signal coupler and three Hewlett-
Packard Model 680, strip chart recorders. Analog signals from the instru-
ments were coupled to the data system through the signal coupler (junction
box) that also housed all the necessary bridge circuits, scaling networks
and voltage reference source. The analog signals were also available at a
patch panel on the front of the signal coupler for purposes of maintenance
and strip chart record selection. The panel included a simultaneous event
marker and zero/span voltages for the three recorders and provided convenient
access when checking instrument status with an external digital voltmeter.
The Monitor Labs data system included internal clock/control, a digital
18-column printer, a 20-digit manual data entry, and a. 30-channel analog
signal input capability. Channel number and utilization are listed in
table 3.
The manual input data entry was used to indicate the operational status
of the analyzers and instruments.
48
-------
The Cipher tape transport system produced a computer-compatible, 1/2
inch, 7-track magnetic tape with a data storage density of 556 BPI.
4.2.2 Instrument Calibration and Pressure Effect Tests
4.2.2.1 Instrument Calibration
Dynamic calibration techniques were used to calibrate the ozone
analyzer and the nitric oxide/nitrogen dioxide analyzer before and after
each major survey conducted from the Raleigh-Durham area. Dynamic multi-
point calibrations were also conducted during periods when the aircraft was
stationed at DeRidder, Louisiana, where calibration equipment was maintained
for the RTI field station. An ultraviolet ozone generator, referenced to the
neutral-buffered potassium iodide (Federal Reference Method) was used to
calibrate the ozone analyzer.— The gas phase titration technique used to
calibrate the nitric oxide/nitrogen dioxide analyzer was the tentative method
19 /
reported in the Federal Register, Vol. 38, No. 110.—
The condensation nuclei counter was calibrated at the factory by comparing
instrument response to that of a Pollak Counter simultaneously sampling the
same air source. The Pollak Counter was considered to be a suitable standard,
and the Environment One instrument was adjusted to give comparable readout.
The thermistor temperature sensors used to measure ambient and manifold
temperature were calibrated at the beginning and throughout the program by
submersing the sensors in a water bath maintained over a range of temperatures
and referenced to a laboratory-type mercury thermometer. The reference
thermometer was calibrated in the laboratory against a Hewlett-Packard
Quartz thermometer, Model 2801A.
Periodically, standard sling psychrometer readings were made and com-
pared with the dew point sensor readings.
The pilot static pressure measurements were calibrated at the begin-
ning of the program and involved repeated low passes over a runway of
known length during a time when the meteorological conditions were reasonably
stable. Several low passes were made from different directions, varying the
aircraft speed over a maximum safe range. Each pass was timed with a stop-
watch. Using airport temperature and barometric pressure readings, and time
and distance measurements, the differential and total pressure sensor outputs
were scaled to indicate ambient static pressure and true airspeed.
49
-------
4.2.2.2 Pressure Effects Tests
The effects of changing altitude on instrument response have been pre-
viously investigated and reported in "EPA Contract No. 68-02-1286 Task 4,
Final Report."— Prior to and at the conclusion of this program, additional
tests were performed on the ozone and oxide of nitrogen analyzers used in
this program. The environmental test chamber facility at NERC, Las Vegas,
was used for these tests. The results of these tests produced a predictable
decline in instrument response to a given concentration with increasing
altitude (decreasing pressure). A detailed description of the chamber facility,
the apparatus used, and the test results are presented in appendix D.
4.2.3 Operations Procedures and Data Validation Techniques
Specific operational and data validation procedures were routinely per-
formed in order to insure uniform instrument operation and provide confidence
in collected data beyond that established by the analyzer calibrations already
described.
Preflight procedures
Analyzers, instruments, and data recording equipment were checked for
proper operating modes. Time and altimeter readouts were synchronized with
those reported by airport ground control. The strip chart recorders were
zeroed and spanned with a known voltage source and adequately identified with
time, flight notation, and parameter to be recorded. The data system time
and manual input codes were preset, and the data system printout was iden-
tified with flight number and description.
Analyzer flow rates were-, checked with a soap bubble flow meter, and
the ozone analyzer span/zero response was verified using a single point
ozone generator and clean air source. The exceptions to the later procedure
were flights preceded with a major analyzer calibration.
Inflight procedures
A flight log, which was maintained by the copilot throughout each flight,
included time, course, airspeed, altitude, climb/descent, position informa-
tion, and observed local weather observations. An instrument/systems log
was maintained by the instrument technician, which included equipment per-
formance or irregularities and inflight maintenance procedures. Additional
logs were kept for grab sample and filter identification.
50
-------
Inflight data validation procedures included span/zero checks for the
ozone analyzer with the single point ozone generator, zero check for the
nitric oxide/nitrogen dioxide analyzer, analyzer flowrate measurements, mani-
fold flow measurements, wet and dry bulb temperature measurements inside the
manifold and low pass air sampling at airports where RTI field stations were
in operation.
Vertical flight profiles were made routinely and in general the flight
pattern illustrated in figure 25 was followed.
Poet flight oheoks
With the conclusion of each flight and after aircraft refueling and
maintenance, the entire instrumentation system (with the exception of filter
18000
16000
2000
1000
Predominant wind
direction
Figure 25. Aircraft vertical profile (typical flight track)
51
-------
and grab sample pumps) was switched to a 115 Vac ground power source. The
Instruments aboard the aircraft were then repaired or adjusted as indicated
by operational checks.
The data for the preceding flight was spot-checked and documented and
flight patterns were mapped with the appropriate time/position information.
Data validation techniques
In an effort to verify aircraft data other than by analyzer calibration
procedures, comparative sampling in the vicinity of an RTI field station
was conducted while the aircraft was on the ground and situated immediately
next to the field station, and during inflight low-pass maneuvers.
An example of comparative; ozone data from the aircraft and from the
DeRidder station during simultaneous overnight sampling is shown in figure
26. During these sampling periods, the sample intake for the aircraft ozone
analyzer was removed from the manifold and with Teflon tubing positioned
approximately 1 meter above the top of the aircraft cabin. Sample intakes
for the two systems were then separated by a distance of approximately 4
meters vertically and 2.0 meters horizontally.
Low-pass comparison of aircraft and field station data were conducted
routinely during the program when the aircraft was based at DeRidder or
when aerial surveys were in the vicinity of other RTI field stations and the
weather permitted low-level flying. A typical low pass pattern is shown in
figure 27 and consisted of the aircraft being flown approximately 50 ft (16 m)
above the runway, maintaining a straight and level flight path as long as
practical (generally approximately 30 seconds). Analyzer outputs were re-
corded near the end of the pass, which allowed the maximum stabilization
time at the low altitude.
A plot of aircraft low-pass ozone data versus ground-station ozone data
is presented in figure 28. Although these data were collected simultaneously
during low-pass samplings, considerable scatter is evident. A review of the
base station ozone strip-chart records frequently revealed a rapidly changing
ozone concentration near the sampling times for many of the outlying data
points on the graph. These departures from a homogeneous ozone concentration
are not unexpected and often result from incomplete mixing during stagnation
52
-------
100
90
80
"I 70
M
a
g 60
•H
«J
*
£ 50
V
<3 40
N
=> 30
20
10
\ ~
0100 0300 0500
0700 0900
TIME (GMT)
1100
1300 1500
Figure 26. Example of ground measurement comparison.
Down Wind Coume
SOI,"
Ground Station
Simple Port
Figure 27. Low pass pattern for aircraft/station comparison.
53
-------
180 r
160
140
120
100
80
< 60
20
45° line
40
hO
80
100
120
140 160
Field Station O (ug/m )
Figure 28. Comparison data during low passes.,
periods or from rapid changes in concentration as a function of time, as
evidenced by the ground-station data recordings.
During the latter part of the program the EMSL-LV aircraft, an instru-
mented B-26, was flown in support of the basic data acquisition program.
Several comparison flights between the B-26 and the RTI Navajo were flown
for purposes of obtaining inflight comparison of data. These were, in
general, on flights between Lake Charles and DeRidder in wing-tip to wing-tip
formation, typically at 305 m (1,000 ft) MSL. In addition, several low
passes were made at DeRidder by sequencing both airplanes across the field
54
-------
with approximately 30 seconds time-spacing. These low pass comparison data
are shown in figure 29.
4.2.4 Data Reduction and Processing
General reduction formulas
The following formulations were used in reducing signal output voltages
for the ozone, nitric oxide/nitrogen dioxide analyzers, and the condensation
nuclei counter.
Ozone and nitric oxide/nitrogen dioxide data are expressed in micrograms-
3
per-cubic meter (yg/m ) as follows:
ioU
p 120
n
o
4_)
^ 80
o
•H
40
s
/
X
0 /
/
/
•o . x
/«° o
»
*
/ °
/ °
/ • EPA Aircraft
x o RTI Aircraft
X
/
£ i —i 1 »
80
120
160
Field Station 0 (yg/nT)
Figure 29. Comparison data during low passes.
55
-------
m V + b
CO C
c _—
a
where
o
c » concentration Ln yg/m , corrected to standard
pressure;
V «• analyzer signal output in mV;
m , b « slope and constant respectively, representing
analyzer calibration curve as determined during
primary calibration; and
K - altitude correction.
a
Altitude correction factors (K ) for the ozone and oxides of nitrogen
Si
analyzers were determined from analyzer response tests conducted in an
environmental chamber at varying altitudes. These tests, described in
detail in appendix D, provided the response versus altitude curves shown in
figure 30 for the ozone analyzer and figure 31 for the oxides of nitrogen
analyzer.
Condensation nuclei (CN) are reported as number-per-unit volume and
are corrected for pressure as follows:
r = r1 —
C C P
m
where
C = number of CN per-unit-volume (CN/yl) corrected
to standard pressure at sea level,
C' = number of CN as measured,
P = standard pressure at mean sea level, and
P = ambient static pressure.
m
Initial data plots
In order to provide a quick look at primary data and aid subsequent
flight planning, flight tracks were initially traced to World Aeronautical
Chart scales (1:1,000,000) (see example, fig. 32). The tracks include
departure, ozone concentration at 2-minute intervals, and location of
selective filter and grab samples.
56
-------
i.o r
o
» .8
1000
5000 9000 13000 17000
ALTITUDE (Feet above HSL)
21000 25000
Figure 30.
Normalized response versus altitude for
Bendix ozone analyzer.
.9
2. -7
.6
3 -5
3
U
O
* .4
.3
3000 7000 11000 15000 19000
Altitude (.Feet above MSL)
Figure 31. Normalized response versus altitude for
Bendix oxides of nitrogen analyzer.
57
-------
_J . /Ol 10
Ul CH. lit If 131 131 121. ""t^Xll 16 i
Figure 32. Example of initial data plots.
Final data format
A sample of the final data format is shown in table A. The processed
data were tabulated in a manner to facilitate future automatic flight plot-
ting by computer. A ground track for the flight data presented in the
tabulated data is shown in figure 33, and a graphic illustration of the
ozone and ambient temperature data, measured during a vertical profile con-
ducted during the flight, is shown in figure 34. Data for all the flights
conducted during this program have been archived in the above manner and
are available upon request.
58
-------
Table 4. Flight data, sea-breeze flight 079 (9/21/75)
TIME
IGMT) POSITION
19:18 DRI
20
?2
24
25
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
20:00
02
04 94°W; 29°N
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
38
42
46
50
54
58
HEADING
TAKEOFF
193"
il
II
M
II
II
II
tl
II
H
II
II
M
II
II
II
II
It
II
II
II
II
090°
II
II
II
II
II
II
M
tl
It
II
II
H
VERTICAL
II
H
II
It
II
II
II
II
ALTITUDE
(ft)
203
750
_
2C51
,
1954
.
2004
_
2001
-
2126
-
2021
_
2011
_
_
_
2060
-
2042
_
.
.
1046
_
2050
.
2051
-
2050
-
2043
.
2043
_
3083
4108
5169
6127
8361
10409
8158
6017
03
(uS/m3)
92
93
98
87
87
87
87
93
87
109
104
115
131
125
169
125
_
142 •
115
131
120
131
-
136
125
125
109
142
169
136
153
136
158
158
158
158
97
90
88
93
119
118
104
99
TCMP.
(°C)
24.8
_
20.5
_
20.0
fc
20.1
_
20.6
-
21.0
.
20.8
_
21.2
_
_
.
22,0
-
22.1
-
-
21.5
-
—
21.5
-
20.7
-
20.7
-
20.5
-
20.5
_
19.5
18.8
17.5
15.4
12.7
9.6
9.7
14.3
DE'.iPT.
(°C)
11.6
_
9.3
_
10.8
^
12.1
-
12.7
-
14.2
-
15.7
-
15.4
_
_
_
15.1
-
14.9
.
-
17.3
-
^
17.6
_
18.0
-
18.3
-
17.7
- -
17.8
_
16.4
13.3
13.3
10.9
6.3
0.8
12.7
10.5
MANIFOLD
TEMP.
(ec)
23.3
_
25.4
_
25.7
.
25.3
-
25.3
-
25.7
_
25.8
_
25.9
_
_
_
26.7
-
26.9
_
-
2614
-
_
26.7
-
26.0
-
26.0
_
26.0
-
26.0
^
25.0
24.4
23.3
21.7
19.6
17.5
16.9
20.3
TRUE
AIR SPEED
(r.iph)
151
„
177
-
185
184
181
181
184
181
_
^
-
181
-
183
_
_
183
-
.
180
-
181
184
184
_
183
_
183
175
175
180
177
174
198
194
59
-------
Table 4 (con.)- Flight data, sea-breeze flight 079 (9/21/75)
TIME
GMT)
21:02
06
10
14
16
18
20
22
24
26
28
30
32
34
36
37
22:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
45
POSITION HEADING
it
11
II
M
92°W; 298N 360°
II
II
1 "
It
II
II
tl
I)
H
II
LFT LANDING
LFT TAKEOFF
350°
II
II
II
M
It
II
II
II
92°W; 31°N 262°
It
II
II
II
II
II
II
II
II
II
II
DRI LANDING
ALTITUDE
(ft)
5059
4032
3008
1935
2018
-
2001 '
-
2022
-
2055
_
1289
_
633
.
2128
_
2080
_
2116
-
2098
-
2101
-
2125
-
2102
«.
2081
-
2098
-
1029
-
203
03
(ug/in3)
88
90
103
164
164
164
164
169
185
196
223
174
153
147
131
••
.
104
109
115
109
109
125
120
115
109
104
109
104
104
98
98
98
98
98
98
104
104
-
TLMP.
cc)
15.9
16.7
18.4
20.0
21.3
-
21.6
-
21.9
-
22.2
^
23.0
-
25.0
™
-
20.4
-
19.7
_
19.6
-
19.6
-
19.2
-
19.1
.
19.3
„
19.0
-
18.7
-
19.9
_
-
DEWPT.
(°C)
14.7
16.3
17.7
18.2
17.2
-
17.2
_
17.4
-
17.2
„
17.2
.
17.1
*
.
13.7
-
13.7
_
11.4
-
10.3
-
11.0
-
9.4
-
9.2
—
10.5
-
10.4
-
11.9
-
-
MANIFOLD
TEMP.
Cc)*
21.7
22.3
23.9
25.2
25.0
-
26.5
.
26.9
-
27.2
_
28.2
_
28.7
••
-
26.6
-
26.0
—
25.3
-
25.4
24.7
-
24.6
-
24.6
_
24.2
-
23.9
-
24.8
-
-
TRUE
AIR SPEED
(mph)
188
188
138
189
184
-
186
_
184
_
181
_
187
_
122
~
.
ISO
_
179
_
181
_
181
-
179
-
183:
_
183.
—
181
_
179
-
173
„
-
60
-------
DeRidder
A2245Z
(LP-1)
D1920Z
TEXAS
2220Z
92°00'W
20042-^-
94°00'W V
29°00'N
GULF OF MEXICO
central
time zone
2116Z
92°00'W
29°00'N
SEA-BREEZE (079). 9/21/75
SCALE: 1 in =• 35 mi
TIME: GMT
WEATHER: 200*15
Figure 33. Sample flight track.
61
-------
H
O
§
o -4 n
W
P
§
w
o
S3
s
I
VI
a)
1
4-1
ft
-------
4.2.5 Aircraft Flight Summary
During the program, a total of 111 missions were flown for a total
flight time of 292 hours, ranging from 1.5 to 4 hours C2.6 hour average) per
flight. The type of flights included 34 gulf coast surveys Cfigs. 35 and 36),
28 high pressure surveys (fig. 37), plus transition and test flights.
Summaries of the northern route high pressure surveys and the Gulf
Coast surveys are presented in tables 5 and 6, respectively. Table 7 lists
the verticals conducted during the program and includes the locations where
the vertical was performed and the maximum altitude surveyed. Table 8 lists
joint EPA-RTI flights conducted in the Gulf Coast area during the latter
part of October. A listing of location identifiers is given in table 9.
4.3 Ozonesonde Measurement Program
4.3.1 Introduction
To assess the role of the vertical distribution of ozone upon the
ozone measurements near the ground, a special program of serial ozonesonde
releases was initiated. These releases documented the vertical distribu-
tion of ozone and its changes. Special attention was given to any evidence
of the intrusion of stratospheric ozone into the troposphere. The behavior
of ozone in the planetary boundary layer was also investigated. Twenty-four
ozonesondes were released during three program periods from Huron, South
Dakota, and one period from DeRidder, Louisiana. These soundings were made
sequentially, three times a day—near sunrise, in the early afternoon, and
after sunset—showing the greatest contrasts of ground^level ozone during
the day. Ozone data were taken during vertical profile flights, and a
quality assurance program was conducted by the EPA.
4.3.2 Instrumentation and Data Acquisition
The ozonesonde consists of a regular radiosonde unit with a separate
ozone sensor package attached. The radiosonde unit telemeters data from
onboard temperature and humidity sensors at calibrated intervals of pressure
changes. If the balloon carrying the package is accurately tracked, winds
can be calculated. The ozone sensor package used in this study was the
20/
electrochemical concentration cell (ECC) developed by Komhyr and Harris.—
The ozone (oxidant) is measured by amount of current generated as ambient
air is pumped through KI reagent in the sensing cell. The ozone partial
63
-------
Lafayette
Figure 35. Gulf coast sea-breeze flights (6/28/75-10/21/75),
64
-------
m
r^
i-i
m
o
iH
m
r~-
m
CM
to
4-1
•a
•H
1
4-1
01
«
O
u
U-l
iH
O
vO
J-l
a
•H
65
-------
u-i
O
oo
in
00
r~
^/
CO
4J
•a
(0
CO
•a
0)
1-1
O
52
ro
cu
M
a
•H
66
-------
Table 5. Flight summary, northern route high pressure system surveys
Date
1975
7/8
7/8
7/10
7/11
7/12
7/13
7/14
7/22
7/23
7/24
7/25
7/26
7/27
7/27
8/12
8/13
8/14
8/16
9/4
9/5
9/6
9/7
9/8
9/11
9/12
9/15
9/27
9/28
9/29
9/30
No.
013
014
015
016
017
018
Oi9
012, 022
023, 024
025, 026
027
028
029, 030
031
042, 043
044, 045
046, 047
048, 049
055, 056, 057
058, 059
060, 061, 062
063, 064
065, 066
067, 068
069, 070
071, 072
084, 085
086, 087
088, 089
090, 091
*
Type
Transition, RDU-FSD
FSD-OLF-SDY
SDY-OLF-HARVE-OLF-SDY
SDY-OLF-STANLEY-OLF-SDY
SDY-AIA-DEN
DEN-FRI-JEF
Transition, JEF-RDU
Transition, RDU-TCL-LCH
Transition, LCH-ICT-SDY
SDY-LNK
LNK-HUF
HUF-AGC
AGC-ACK
Transition, ACK-RDU
Transition, DRI-SDY
SDY-OLF-SDY
Transition, SDY-DEN
Transition, DEN-RDU
Transition, RDU-BIS
BIS-OLF-HON
HON-IND
IND-PWM
Transition, PWM-KDU
Transition, RDU-CAP-PIR
PIR-SGF
Transition, SGF-BNA-RDU
DRI-DAY
DAY-AGC
AGC-EWB
Transition, EWB-RDU
Identifiers listed in table 9.
-------
Table 6. Flight summary, gulf coast area
Group
1
2
3
4
Date
1975
6/25
6/26
6/27
6/28
6/29
6/29
6/30
8/5
8/6
8/6
8/7
8/7
8/8
8/9
8/11
9/18
9/19
9/19
9/20
9/21
9/21
10/9
10/10
10/10
10/13
10/14
10/19
10/20
10/21
10/22
10/24
10/25
10/30
10/31
11/1
No.
003, 004
OOf.
006
007
008
009
010, Oil
033, 034
035
036
037
038
039
040
041
073, 074
075
076
077
078
079
092, 093
094
095
096
097, 098
099, 100
101, 102
103
104
105
106
109
110
111
*
Type
Transition, RDU-DRI
Downwind Plume
Downwind Plume
Sea Breeze
Vertical
Vertical
Transition, LCH-RDU
Transition, RDU-DRI
Vertical
Vertical
Vertical
Figure Eight
Figure Eight
Figure Eight
Downwind Plume
Transition, RDU-DRI
Sea Breeze
Sea Breeze
Sea Breeze
Sea Breeze
Sea Breeze
Transition, RDU-DRI
Sea Breeze
Sea Breeze
Sea Breeze
Box Pattern
Box Pattern
Box Pattern
Sea Breeze
RTI/EPA Comparison
RTI/EPA Comparison
Vertical
Box Pattern
RTI/EPA Comparison, Box Pattern
Transition, DRI-RDU
luentifiers listed in table 9.
68
-------
Table 7. List of vertical profile flights
Jjlfj Flight No.
(Northern Route
7/9 014
7/10 015
7/11 016.
7/11 016J;
7/12 017
7/13 018
7/24 025
7/25 027
7/26 028
7/27 029
8/13 044
8/14 046
9/5 059.
9/5 059?
9/6 060
9/6 061
9/7 063
9/12 070
9/27 085
9/28 086
9/29 089
9/30 090
Location
Verticals)
Wolf Point, Mont.
Wolf Point, Mont.
Wolf Point, Mont.
Stanley
Alliance
Fort Riley
Huron
Peoria
Chatham, Canada
Atlantic City, N. J.
Wolf Point, Mont.
Wolf Point, Mont.
Dickenson
Huron
Huron
i Peoria
Bradford
Concordia
Evansville
Parkersburg
Middletown, N. Y.
69°W, 43 °N
Max. Altitude Sampled
(feet)
10,000
10,000
12,000
12,000
18,000
10,000
12,000
12,000
10,000
10,000
20,000
12,000
10,000
12,000
10,000
10,000
10,000
12,000
12,000
10,000
12,000
10,000
(Gulf Coast Verticals)
6/29 008
6/29 009
8/6 035
8/6 036
8/7 037
8/11 041
9/19 076
9/21 079
10/10 095
10/21 103
10/22 104
10/25 106
10/30 109
Jasper
Jasper
Jasper
DeRidder, La.
DeRidder, La.
Newton
92°W, 28°N
92°W, 29°N
92°W, 28°N
DeRidder, La.
DeRidder, La.
DeRidder, La.
S. W. Beaumont
10,000
10,000
20,000
20,000
20,000
12,000
10,000
10,000
10,000
10,000
3,000
22,000
5,000
69
-------
Table 8. Flight summary, joint EPA-RTI gulf coast flights
Date
1975
10/20
10/27
10/27
10/24
10/29
10/30
10/31
10/31
11/1
No.
01
02
03
04
05
06
07
08
09
Type
Transition, LVN-LCH
RTI/EPA Comparison
Coastal Survey
RTI/EPA Comparison
RTI/EPA Comparison
Box Pattern
RTI/EPA Comparison
Box Pattern
Transition, LCH-LVN
pressure is proportional to the product of current generated, the sonde box
temperature, and the time required to pump a given sample volume into the
cell. The current output of the cell generates a signal, which is trans-
mitted three times a minute by the rawinsonde transmitter when a pump-driven
commutator interrupts the radiosonde data signal. A reference zero and span
current are transmitted on alternate minute intervals. A detailed descrip-
20 /
tion of the sensor package has been given by Komhyr and Harris.— The
system has compared favorably with other ozonesonde systems. Ozonesonde
releases and data reduction were performed by Western Scientific Services,
Inc., Fort Collins, Colorado, under subcontract to Research Triangle
Institute.
The nominal ascent rate of 1,000 ft/min for the rawinsonde package
was decreased to about 800 ft/min for this study to give better data
resolution from the ground to 3 km (-700 mb, 10,000 ft). At that rate,
an ozone reading was available each 90 m (-270 ft) during ascent, which
is sufficient to resolve significant ozone strata in the boundary layer.
The ozonesonde data—pressure, temperature, ozone partial pressure
3
and concentration (vig/m )—at significant levels of the sounding were
reduced from strip chart records and tabulated. Audits of several of the
soundings from the original strip charts and data work sheets showed less
3
than five percent differences. All ozone data are reported in yg/m at
standard conditions (i.e., 25°C, 760 mmHg).
70
-------
Table 9. List of location identifiers
Identifier
ABR
ACK
AGO
ACY
AEX
AGC
AGS
AIA
ALB
AND
ANW
ARB
BFO
BHM
BIS
BNA
BPT
BRG
BRL
BTR
CCT
CLT
CPR
CSQ
CTF
CYS
CZI
DAY
DEC
DEN
DGW
OIK
DPR
DRI
DYR
EVV
EWB
FM
FOD
FRI
FRR
FSD
GAD
GBG
GGG
GGW
GLD
GRW
GSO
GSP
HCH
HCM
HEZ
HKY
HLC
HMV
HON
HSI
HUF
City/ Airport
Aberdeen, TX
Nan tucket, MA
Akron, OH
Atlantic City, NJ
Alexander, VA
Pittsburgh, PA
Augusta, GA
Alliance, NE
Albany, NY
Anderson, SC
Ainsworth, NE
Walnut Ridge, AR
Bradford, PA
Birmingham, AL
Bismarck, ND
Nashville, TN
Beaumont, TX
Whitesburg, KY
Burlington, IA
Baton Rouge, LA
Central City, KY
Charlotte, NC
Casper, WY
C res ton, IA
iChesterfield, SC
Cheyenne, WY
Crazy Woman, WY
Dayton, OH
Decatur, IL
Denver, CO
Douglas, WY
Dickinson, ND
Dupree, SO
DeRidder, LA
Dyersburg, TN
Evans vi lie, IN
New Bedford, MA
Fort Mill, SC
Fort Dodge, I A
Fort Riley, KS
Front Royal, VA
Sioux Falls, SD
Gads den, AL
Galesburg, IL
Longview, TX
Glascow, MT
Goodland, KS
Greenwood, MS
Greensboro, NC
Greenville, SC
Hinch Mountain, VA
Harcum, VA
Natchez, MS
Hickory, NC
Hill City, KS
Holston Mountain, TN
Huron, SD
Hastings, NE
Terre Haute, IN
Identifier
ICT
IND
IRK
JAN
JAS
JEF
JST
LCH
LEX
LFT
LIB
LMN
LNK
LOU
LOZ
LRP
LUL
LWM
MCB
MED
MEI
MHE
MKC
MLC
MLS
MOB
MOT
OKM
OLF
OTM
PER
PFN
PIA
PIR
PKB
POU
PUD
PWM
RDU
RMG
SBI
SOY
SGF
SLN
SPA
SPI
SUX
TCL
TKO
TNU
TOC
TYS
UIN
VPS
YNG
City/ Airport
'Wichita, KS
Indianapolis , IN
Kirkville, MO
Jackson , MS
Jasper, TX
Jefferson City, MO
Johnstown, PA
Lake Charles, LA
Lexington, KY
Lafayette, LA
Liberty. NC
Lamoni , IA
Lincoln, NE
Louisville. KY
London, KY
Lancaster, PA
Lawrenceville, VA
Lawrence, MA
McComb, MS
Mansfield, OH
Meridian, MS
Mitchell, SD
Kansas City, KS
McAl ester, OK
Miles City, MT
Mobile, AL
Minot, ND
Oklahoma City, OK
Wolf Point, MT
Ottumwa, IA
Poner City, OK
Panama City, FL
Peoria, IL
Pierre, SD
Parkersburg, WV
Poughkeepsie, NY
Providence, RI
Portland, ME
Raleigh, NC
Rome, GA
Sabine Pass, LA
Sydney, MT
Springfield, MO
Salina, KS
Spartanburg, SC
Springfield, IL
Sioux City, IA
Tuscaloosa, AL
Mankato, KS
Newton, IA
Toccoa, SC
Knoxville, TN
Quincy, IL
Valparaiso, FL
Youngstown, OH
71
-------
4.4 Program Summary
The program schedule and the sequence of events that occurred during
the period of performance for field measurement activities (i.e., May 8^
October 31, 1975) are presented in this section. The overall program
schedule as originally projected at the beginning of the project and pre-
sented in the "Work Plan" for Contract 68-02-2048 is shown in table 10.
Subsequent discussions related to the events that occurred during the
ground-station and aircraft measurement phases of the project and the
quality assurance activities are presented in the following paragraphs.
The schedule presented in table 10 for each task was met with respect
to acquisition/checkout/preparation of equipment for ground stations; site
selection; preparation/installation/checkout of equipment in the aircraft;
installation of stations, analyzers, and equipment at the designated sites;
and calibration of all analyzers. Each of the five rural stations was
set up, calibrated, and brought on-line in a consecutive, orderly manner.
The 15 selected stations operated by State, local, and private groups were
operational, on-line stations. The field measurement program began at all
stations on or before June 30, 1975. Since one aircraft was being utilized
for both studies, e.g., northern high pressure and gulf coast oxidant
studies, a decision was made to base the aircraft in DeRidder and conduct
flights in the gulf coast area until an appropriate high, pressure system
developed in the northern study area. The aircraft was flown to DeRidder
on schedule and subsequently prepared for routine aircraft flights, as
dictated by meteorological conditions and the program plan. During the
ensuing field measurement program, dynamic calibrations were performed
on each analyzer on a monthly basis. Travel delays, instrument failures,
and quality control/assurance procedures altered the schedule somewhat.
Six quality assurance performance audits were conducted at each State/
local agency station audited by RT1 during the study. Monitoring and
data acquisition continued through September 30, 1975, at the northern
stations (Bradford, Creston, Wolf Point) and through October 31. 1975. in
the gulf coast area (DeRidder).
During the 120-day field measurement program, a total of 111 individual
aircraft flights were flown in support of ground-station measurements for
the combined studies. These flights were flown under varying meteorologi-
cal conditions and included sea breeze flights, coast areal survey flights,
72
-------
Table 10. Program schedule
Date Task
May 8, 1975 Start program/begin equipment acquisition.
9 Begin preparation of comprehensive program plan.
8-16 Conduct chamber tests at NERC, Las Vegas (03 and N02
analyzers).
10+ Receive government furnished equipment
12-16 Select sites for ground stations.
15+ Begin checkout of equipment and training of
operators/prepare equipment for installation in
aircraft/develop flight protocol/begin background
study.
31 Submit monthly technical progress narrative (MTPN).
June 1-15 Organize and develop operational procedures/continue
training of operators.
10-12 Submit comprehensive program plan.
10-20 Install equipment in aircraft/begin test flights.
18 Transport equipment from RTI to field locations.
21-25 Install and check out equipment at monitoring sites.
23-30 Calibrate analyzers/begin field measurement program.
23-25 Fly aircraft to base station/check out aircraft
system/begin aircraft measurement program.
30 Submit monthly technical progress narrative (MTPN).
July 7-15 Conduct first audit of state/local stations.
15 Begin data reduction and processing.
16-27 Calibrate analyzers at each RTI station.
21-29 Conduct second audit of state/local stations.
31 Submit monthly technical progress narrative (MTPN).
August 1-31 Continue data acquisition and processing.
4-12 Conduct third audit of state/local stations.
16- Calibrate analyzers at each RTI station.
18-26 Conduct fourth audit of state/local stations.
30 Submit monthly technical progress narrative (MTPN).
73
-------
Table 10 (con.)- Program schedule
Date
September
October
November
1-31
6-14
15-26
20-28
30
30
1-31
1-5
15-20
31
31
1-5
1-15
15-31
31
Task
Continue data acquisition and processing.
Conduct fifth audit of state/local stations.
Calibrate analyzers at each RTI station.
Conduct sixth audit of state/local stations.
Terminate field measurement program at northern
stations.
Submit monthly technical progress narrative (MTPN) .
Continue measurements in Gulf Coast area.
Disassemble equipment at northern monitoring
stations /transport equipment to RTI.
Calibrate analyzers at DeRidder station.
Terminate measurement in Gulf Coast area.
Submit monthly technical progress narrative (MTPN) .
Disassemble equipment/transport equipment to RTi.
Compleste data processing.
Return GFE to EPA.
Submit monthly technical progress narrative (MTPN) .
dowwind plume flights, vertical profile flights, double-box pattern
around Nederland, calibration and instrument checkout flights, and northern
high pressure system flights. A listing of all flights was presented in
section 4.2.5.
74
-------
5.0 QUALITY ASSURANCE PROGRAM
5.1 Quality Assurance Protocol
A network of 19 ground stations was used to provide much of the data
for the summer study. RTI was responsible for maintaining a quality assur-
ance program for the ozone measurements made by 11 of the 19 stations. EPA
was responsible for maintaining a similar quality assurance program for the
ozone measurements made by the remaining eight stations, four of which were
operated by the RTI. EPA also conducted performance audits of the NO-NO--
NO analyzers in the four RTI-operated stations.
X
The RTI quality assurance program consisted of a preliminary on-site
systems audit of the equipment, facilities, procedures, and personnel to
evaluate the capability of that station or agency to produce data of accept-
able quality. A series of six performance audits were conducted at each
station to assess and document the quality of the ozone measurements being
made. The ozone measurement network for which RTI's quality assurance pro-
gram was designed consisted of 11 ground stations operated by State or local
agencies or private industries. Station locations and operating agencies or
indus t ries are:
Lewisburg, West Virginia (Bendix Corporation)
Pittsburgh, Pennsylvania (Local Agency)
Columbus, Ohio (State Agency)
Indianapolis, Indiana (State Agency)
Colstrip, Montana (State Agency)
Pensacola, Florida (State Agency)
Port O'Connor, Texas (E. I. Du Pont de Nemours)
Houston, Texas (State Agency)
Nederland, Texas (State Agency)
Austin, Texas (State Agency)
Corpus Christi, Texas (State Agency)
The network of eight stations for which EPA maintained a quality assur-
ance program included four stations operated by the RTI plus four stations
operated by State or city agencies as follows:
75
-------
Bradford, Pennsylvania (Research Triangle Institute)
Creston, Iowa (Research Triangle Institute)
Wolf Point, Montana (Research Triangle Institute)
DeRidder, Louisiana (Research Triangle Institute)
Omaha, Nebraska (City Agency)
Cedar Rapids, Iowa (City Agency)
Des Koines (Bondurant), Iowa (City Agency)
Poynette, Wisconsin (State Agency)
The qualitative systems audit and the quantitative performance audit
are discussed in subsections 5.1.1 and 5.1.2, respectively.
5.1.1 Qualitative Systems Audit
The objective of the on-site/off-site qualitative systems audit was
to assess the potential of that station and agency to generate ozone data
of acceptable quality throughout the duration of the summer study.
The systems audit was conducted in two phases. Agencies operating the
respective stations were contacted prior to the start of the summer program
by telephone. Operation, calibration, and data validation procedures em-
ployed by the agency were discussed. Also, facilities and equipment in the
station or used by the agency were reviewed. In preliminary discussions, no
station or agency was judged to be deficient enough to constitute exclusion
from the air monitoring network.
In concurrence with the first quantitative audit, an on-site systems
audit was performed. This a.udit included:
1) Verification of the procedures used in calibration either through
a review of their written procedures or through actual observation
of a calibration or both.
2) A check, when applicable, of the station zero air supply by a vis-
ual inspection and by comparing analyzer responses to the auditor's
clean air supply and the station's zero gas.
3) A check of the agency's capability to perform dynamic calibrations
at suitable intervals.
4) A check of the agency's recordkeeping practices.
5) An inspection of the station, including sampling probe, manifold,
inlet filter, and instrumentation.
76
-------
5.1.2 Performance Audit
The objective of this task was to perform a series of systematic on-
site quantitative performance audits to collect information on the precision
and accuracy of summer study ozone measurements. During the summer study,
EPA and RTI teams provided their own 0- calibration systems and generated
known 0_ concentrations on-site for audit purposes. EPA, in addition, pro-
vided a calibration system for auditing RTI's NO and NO measurements.
Each respective analyzer audited by EPA and RTI for ozone or oxides
of nitrogen was challenged, in most cases, at four upscale points. Stations
within the EPA-RTI monitoring network were audited a minimum of two times
and up to a maximum of six times. The responsible agency was notified by
telephone any time an analyzer was found to differ more than ± 20 percent
from the audit value. If the analyzer was within ± 20 percent of the audit
value, the results of the audit were mailed to the agency after completing
audit checks of all the stations.
5.2 Description of the Air Pollution Monitoring Network
Each station in the 19-station network is discussed hefte from a quality
assurance point of view. Information given in each description includes geo-
graphical location, site characteristics, and station description.
The first 11 stations described form the 11-station monitoring network
audited by RTI. The last eight stations, four of which were operated by RTI,
were audited by EPA.
A. Lewisburg, West Virginia, Monitoring Station
The Lewisburg, West Virginia, station was located at the Greenbrier Val-
ley Airport, approximately 160 km southwest of Garrett County, Maryland. The
airport elevation is approximately 705 m MSL. The airport serves both pri-
vate and commercial aviation. The valley surrounding the airport is rolling
pastureland with some wooded areas. Analyzers and associated equipment were
housed in an air-conditioned 2.5 * 9.2 m mobile laboratory owned by the
Bendix Corporation, Lewisburg, West Virginia. A 1.25-cm diameter Teflon
tube located 2 m above the laboratory served as the sample inlet line.
B. Pittsburgh, Pennsylvania, Monitoring Station
The Pittsburgh station was located in Penn Hills at 12245 Frankstown
77
-------
Road. The analyzer, owned and operated by the Allegheny County Air Pollu-
tion Control Agency, was housed in a large air-conditioned room in the Penn
Hills Municipal Building. A 0.6-cm OD Teflon line approximately 7.5m long
served as the sample inlet line with the sample inlet well below the build-
ing' s roofline. The probe inlet had only about 270 degrees of exposure.
C. Columbus, Ohio, Monitoring Station
The Columbus station was located in the NE section of the city at the
intersection of highway 161 arid Maple Canyon Drive. The analyzer was housed
in the bay of a fire station with no temperature control. A 0.6-cm OD Teflon
line approximately 11 m long served as the sample inlet line. The probe
inlet was above the roof line and had 360° of exposure.
D. Indianapolis, Indiaaa, Monitoring Station
The Indianapolis station was located in the NE section of the city at
71st and Tacona Avenue. The analyzer was located in a temperature-controlled
(aluminum) shelter. A 0.6-cm OD heated Teflon line approximately 18 ft in
length was mounted 1 m above the building and served as the inlet line.
The sample inlet was an inverted glass funnel and had 360° exposure.
E. Colstrip, Montana, Monitoring Station
The Colstrip station was located approximately 3 km southeast of Col-
strip, Montana (Burlington, Northern Site). The analyzer was housed in a
temperature-controlled camper trailer. A 0.6-cm OD Teflon line was con-
nected to a 2.54-cm glass manifold. The sample probe extended above the
trailer and had 360° of exposure. The trailer was parked on a mesa approx-
iajately 30 m above the surrounding ground level.
F. Pensacola, Florida, Monitoring Station
The Pensacola station was located on Ellysen Naval Air Station. The
analyzer was housed in a temperature-controlled trailer. Samples were taken
from a glass manifold. The sample inlet line was wrapped with heat tape to
prevent condensation in the inlet sample line.
G. Port O'Connor, Texas, Monitoring Station
The Port O'Connor station was located on the beach at the corner of
Washington Boulevard and Harrison Street and is considered a rural station.
78
-------
The analyzer, owned and operated by the Du Pont Company, was housed in a
storage shed (2.4 x 2.4 x 2.4 m) under a beachhouse. The sample was ob-
tained through a 9- to 10-m 0.6-cm OD Teflon line that extended 1 m above
the roof with 360° exposure. There was no temperature control, and high
humidity was experienced on all audit trips.
H. Texas Air Control Board Monitoring Stations
The Texas Air Control Board (TACB) stations are self-contained labora-
tories with the capability of performing over long periods of time. All
stations were custom designed after the crew-shelters used by oil companies
on the north slope in Alaska and in the Middle East. The windowless units
are constructed of 7.6-cm pressure-bonded laminate and include 3.8 cm of
styrofoam Insulation. The stations are 3mx 3mx 7.3m, with the inte-
rior divided into a 3 m x 6 m monitoring laboratory (main room) and a
3 m x 1.2 m utility room. The central air conditioning/heating unit pro-
vides a stable temperature (±1° C) and safety ventilation. The sample
manifolds are glass and are heated by incandescent bulbs. Sample air and
zero and span gases pass through the same filter before entering the ozone
analyzer. The ambient sample intakes are inverted glass funnels approxi-
mately 3.6 m above the roof with 360° exposure.
Each of the Texas monitoring sites is described below.
H.I. Connie* 8: Aldine
Connie station 8 is located in Aldine, approximately 3 km north of the
Houston city limits and about 20 km northwest of the Houston Ship Channel.
Data reported from Aldine will be considered as data for Houston in this
report.
The industries of the area emit primarily sulfur dioxide, carbon monox-
ide, nitrogen oxides, and various hydrocarbon compounds. Because of the
prevailing winds (wind direction south to southeast 38 percent of the time),
these emissions and photochemical oxidants are the principle pollutants
measured at the monitoring site.
*For convenience, the TACB has given the continuous monitoring program
the nickname of Project Connie and the individual sites are called Connie
stations.
79
-------
H.2. Connie 2; Nederland
Connie station 2 is located on the southeast corner of the public park-
ing lot at the Jefferson County Airport in Nederland, Texas. Elevation of
the site location is 5 m MSL. The Be.aumont-Port Arthur-Orange area is
moderately populated and heavily industrialized. Due to the type of indus-
try in the area, there should be significant quantities of hydrocarbons and
sulfur dioxide. The Jefferson County Airport is also the location of previ-
ous studies of photochemical oxidants. The wind direction is from the south
approximately 15 percent of the time.
H.3. Connie 3; Austin
Connie station 3 is located in the south parking lot of the Texas Air
Control Board, 8520 Shoal Creek Boulevard, Austin, with an elevation of 198 m
MSL. The city of Austin has no heavy industry, chemical industry, or petrole-
um-related gaseous emission sources. The city has two electric generating
plants. One plant is located in the center of the city on the Colorado River.
Bergstrom Air Force Base lies approximately 1.6 km southeast from downtown.
The University of Texas complex lies immediately to the north of the downtown
areas. Emissions from these sources, as well as those of the downtown area,
are carried to the Austin Connie station by southerly and southeasterly winds
prevalent in the Austin area. (37 percent of the time).
H.4. Connie 4; Corpus Christi
Connie station 4 is located in the southwest corner of the school
grounds at the Robert Driscoll Junior High School in the 200 block of Old
Robstown Road, Corpus Christi, Texas, and has an elevation of 12 m MSL.
The area within a 1.2-km radius of the site is primarily residential, but
schools, shopping centers, and light industry are also found within this
area. Outside this radius and to the north are 12 major heavy industrial
complexes, including chemical, petrochemical, and petrorefining facilities,
nonferrous smelters, and cement plants. Grain storage and shipping areas
and power-generating stations are also located to the north. Due to pre-
vailing winds from the northwest 24 percent of the time, pollutants emitted
from such industries reach the station.
80
-------
I. Rural Monitoring Stations Operated by RTI
The RTI operated rural stations were located in: Bradford, Pennsyl-
vania; Creston, Iowa; Wolf Point, Montana; and DeRidder, Louisiana. Des-
criptions of these stations were given in section 4.0 and are not repeated
here.
J. Poynette, Wisconsin, Monitoring Station
This station is located at the edge of Poynette and is considered a
rural station. The station has a heating system but no air conditioning.
The sample line is a 0.9-cm ID Teflon line approximately 7.6 m long with
a probe inlet height of about 9 m above ground.
K. Cedar Rapids, Iowa, Monitoring Station
This station was located about 3.2 km northeast (~20°) of the center
of Cedar Rapids in a residential area. The station was air conditioned.
The sample intake system consisted of a glass probe and manifold with a
Teflon line from the manifold to the analyzer.
L. Des Moines (Bondurant), Iowa, Monitoring Station
The station was located in the center of Bondurant about 16 km north-
west of Des Moines. A 9-m Teflon line 0.3 cm ID served as the sample in-
let line. The station had a heating system but no air conditioning.
5.3 Audit Procedures
The equipment and procedures used by the RTI and EPA audit teams are
discussed in the following subsections.
5.3.1 RTI Ozone Auditing Procedures
An ozone calibration system as described in appendix A of this report
was used by the RTI audit team to generate test atmospheres of known ozone
concentrations for challenging the ozone analyzers.
The ozone generator was calibrated in the laboratory at RTI using the
1% NBKI procedure prior to the field audits. When performing an audit the
generator output as read from its laboratory developed calibration curve
was taken as the reference 03 value with 1% NBKI analyses performed in the
station to verify that the calibration had not changed significantly during
shipment and handling.
81
-------
The NBKI sampling train used by RTI was the same as that in the Federal
Register.— The sampling train flow rate was controlled at about 500 cm /min
with a critical orifice. A bubble flow meter (traceable to NBS) was used to
determine the exact flow rate at each station. Ten-minute sampling periods
(timed with a stopwatch) were used. An aneroid barometer and a mercury-in-
glass thermometer were used to measure station pressure and temperature,
respectively.
5.3.2 EPA Ozone Auditing Procedures
The calibration system employed by EPA for audit purposes was based
upon reverse gas phase titration (RGPT) utilizing an NBS certified tank of
nitrogen oxide. This audit technique was based upon the rapid gas phase
reaction between 0. and NO ir. accordance with the following equation:
NO + 03 -> NO + 02, K = 1.0 x IQ7 1 mole"1 sec""1
An 0_ concentration of unknown magnitude was generated and sufficient NO of
known concentration was added in a. dynamic calibration system to decrease
the 0., concentration by 90-9.5 percent of its original value as measured on
the station chemilutninescent 0, analyzer. The assay for the NO cylinder
used in this technique was based on the gas phase titration (GPT) of NO
with 0, in which the 0. concentrations were determined iodometrically using
the 1 percent buffered potassium iodide procedure. The NO cylinder served
as a transfer standard traceable to the primary standard used in the cali-
bration procedure for the measurement of photochemical oxidants as specified
in the Federal Register. The flow conditions in the dynamic calibration
system were optimized to ensure the quantitative reaction of NO with 0- and
to minimize the reaction of NO with 0~ which could cause errors in the
audit procedure. Erroneous results will also result from this technique
if the analyzer response is non-linear.
The ozone analyzers were challenged at two upscale points. These
points were 40 percent and 80 percent of full scale.
5.3.3 EPA Oxides of Nitrogen Auditing Procedures
The EPA audit team utilized an NO cylinder with dilution to audit the
82
-------
NO analyzers. Gas phase titratlon was used to generate known N02 samples.
The gas phase titration system is illustrated in figure A-3 of appendix A.
A detailed description of the technique is given in appendix A and is not
repeated here.
Audit checks were generally made at 40 percent and 80 percent full
scale for the NO., analyzers.
5.4 Analysis and Interpretation of Audit Data
The primary objective of the on-site performance audits was to allow
estimates of the precision and accuracy of the 0.,. N0_ and NO measurements
made by the monitoring network and subsequently used in the summer study.
General data analysis procedures used to estimate measurement precision
and accuracy are discussed here. Quantitative information is given by pollu-
tant in the following subsections.
The variable used in the analysis of audit data is the percent differ-
ence in the reference sample concentration and the analyzer response calcu-
lated by the relationship
d = 100 (C - C )/C
m a a
where d = difference in the analyzer response or
measured value C and the reference or
m
audit concentration C , percent
Q.
C = analyzer response or measured value,
yg/m
C = reference or audit sample concentration,
The difference obtained at the it station on the jth audit is represented
by d . . . The bias in the pollutant measurements for the network of stations
over the audit period is estimated by
4
nk
83
-------
and the standard deviation of the differences, s,, is calculated by
The above calculated standard deviation s, is the standard deviation
d
of the difference of two measurements; that is, for example, the difference
in the 0,, concentration as determined by the auditor and the 0~ concentra-
tion as measured by the analysier. To prevent overestimating the quality of
the measurements, the "worst case condition" is assumed here. That is, it
is assumed that all the variability in the differences (d..'s) is due to
the measurement system. The audit value is assumed to be the true value.
Under these assumptions, s,
the measurement data, i.e.,
Under these assumptions, s, is an estimate of coefficient of variation of
s. = CV.
d
The coefficient of variation is used as a measure of the precision of
the air pollution measurements. Two parameters are ordinarily needed to
characterize the accuracy of a measurement process, one indicating its pre-
cision and the other its bias; therefore T and CV must be viewed together
(in a confidence interval, for example) as a measure of accuracy.
Precision and accuracy of. Q and oxides of nitrogen, i.e., NO and NO^,
are discussed in the following subsections.
5.4.1 Precision/Accuracy Estimated for Ozone Measurement
The ozone audit results are given by station, by concentration level,
and by the date of the audit in table 11. The results are given as the
relative difference between tie measured and the audit (input) values. The
station bias, d., is given in the right hand column of table 11.
It is noted here that the Pensacola, Florida, and Corpus Christi,
Texas, stations were audited by RTI; however, the. ozone data from these
two stations were invalidated or received too late for inclusion in the
84
-------
cfl
4J
rt
1
g
00
o
M
M-l
O
CO
0)
-O
•H
X
O
T3
§
0)
C
O
N
O
<4-l
O
Cfl
•9
H
JJ
•H
0)
0
*M
h
P*
•e
\e
m
-*
«n
(M
T
4
JJ
1
<
t
j
J
a
t
c
c
1
^
4.
fl
4.
V
.
/-s
"
V
•e
V
4J
2
XN
\f
*e
V
4J
Q
^
«
s
•«
4J
I
K
f
•o
4J
9
«
e>
•o
AJ
a
M
•o
£
(0
4
J
0
•< m
1 3
3
!
i
a
i
•
*
;
tfi
H
a
u
i
1
O
M
U
p» o in« m «in -.*** *^.*K."sfc**»*^
0s eft ON » OHOOOO
H t-t H H
h
iH%p oo ov m r*»>p tnrnn O ON m>H HO « cs O
.H? +?? . T . i . 7 7« + , |a 7<1,7
m in in m
min inm m r*. r* r^. r*
•**."x. "N. *««, •***•* in m m
in« MJ r^ eOHf-JHrH
^.*^. "«^'1^ "^OOOO
0* O% 019 QlH*HHH
O (N ro -ff r* o« r-co-e in ON O p^ O HO o -e H *o \ptnm
1 III HH III H++HHWH+-(- +
mm mm m in m m m
in %o r*. co ovHooiO
(N fM N -»nr«*r«*
COO* -N.O HOV-X.HO
HH O*«M (MfMHnm
coco coco co r- oo r- r*.
B "
CM oo cso\o rn H mmf) co tn co VD in Hen o ja
H4J*
(SCMHHHH+ I+H | +1 II || BJKH
1 1 + + + + + 4J ffl
"> ^
m in m m m
inr^. r^in inr*»r^r>.
*-*O H**. -^*OP«»CO
CT\H H kl r^HHH
r-.r^ r^co eor^r^r^
in o\ m o» co m ON mON «o r>. moo m co moo in ON GO
JS .
5 "£
MB eg
^ > C 9)
•H .H CO cd
> >• -H K
to *O 09 a)
» g S § S H
oj oi o jj m K i
S CM ^H * C » 10 0) U
X B O Q K H O
** O-H 32K9€ MCOHO
nn si w-Hwh
•H 4J 9*H (DJJ03V4J
S U H *O H O 3 -« M
$*H OC 03OVO
^ &- U -< U < X Z Cu
85
-------
c
o
a
•u
cfl
13
•H
T)
c
0)
60
O
•H
C
03
01
13
o
TJ
8
0)
C
O
•3
H
*J
*H
•o
1
o
*+•(
M
«
Pw
t
m
-w»
M
CN
-H
4J
•H
•e
^
iH
)
s
%c
•H
•o
V
4-»
01
O
S
"a
•a
v
4J
S
M
<•
«H
•a
0)
±j
5
N
n
^H
•o
«)
u
Rt
O
s-*
M
CM
•H
•o
.
in ^>
r-* a>
•**- IM
rH ^-
•^ o
O 0
??
»n
r^.
•-•*
N
i-t
o\
o co tn
in o o
1 iH 1 1 1 1
«n
r-»
\Ti "*—
r^ co
-1-. fM
d S
O>> rH
yp m O co
m rv . r^-
O% 0
fH CM
CO O\
o m o o o
n er* o o ft
-HI lit +
m m
m r^ f-"
r^ -^. •i^
•^ m o
iH tN CS
-^ — -*^
oo r^. co
•o
O O O O 0)
o co o m 4J
fn in ro m u
-* *n o
o o o T-N. o \o &
o o o o
m \o in m u
H in M "» o
z
Creston, Iowa
DeRidder , Louis Iana
Bradford, Pennsylvania
86
-------
c
o
o
cd
4-1
c
0)
60
o
CO
(II
•o
•H
c
CO
CD
C
O
N
O
&
cd
•s
H
4J
•H
?
V
g
?
£
fc
1
m
•*
m
(N
*H
4J
•3
3
W
v£
•o
0)
.u
W
O
s
IT
•o
01
JJ
td
o
x
-a
•o
4J
at
O
/~n
X
f*
•o
u
a
t-s
Cs
•o
0)
4J
«
^
•o
JJ
a
g
4J
ooo^ »ArH ON inm nnm ino >^rH
O^o r^-r11* oo Ov O»
€M ««• -^ O in OOOv
?? ?7 7 ' 9 t"
m r>. r-^ r^. r>.
r* "^. "^ "*^ v*.
"^. %o m r>. r4
Ot i-H tH i— 1 i-(
O* O* Ov W OS
T-l iH M C4 | | + ,H iH
m m m
in r*. r^ r^. m
•*-. oo o» vo -^.
CsJ r** CM CM .-t
O^ CO CO OO O\
-.^rcom^^^^om^ 00-,
1 1 iH ( i—t 1 OJ CS rH T— 1 I IfrH^H
1 1 + -f 1 1 II
mmmmiAm mm
oooo*fn«*
+ 1 t-il r^<-i-frHrsieM 1 + + 3 | estrH
1 11+11 T T 1 + 1
m in «n in m m
CMtnrn-^OfH O^^.
«HiHr^«nrop-i esiH
ONr^r^coi^r^ r-co
p*-r>i t*-psi F-^CN fNiin csin r-»tsim r*»r^ r^-cj
CD
•H
c
03 « e
> « C 0 -H
i— ( c to a OQ
(^ « -H S TO C
ID 4J IB (Q M fit O
C C -H J2 6 U
C53tO*M ttfO
o) E o tc oo ^H at
ft- J t- -0 35
*• .D -H CO M
* 4J •> QJ & QJ •
T5 C l-( Z rt C 0)»
>-i -H
-------
summer study analysis. Therefore, these audit data are not included in
this analysis.
The ozone audit data for the RTI-audited network yielded an overall
bias of
A
T •* -1.8 percent,
and a coefficient of variation of
CV = 13.5 percent.
The combined data for the EPA ozone audits result in an overall bias
of
A
T = -4.3 percent,
and a coefficient of variation of
A
CV - 14.9 percent.
From the above estimated population parameters and under the assumption
made previously concerning the accuracy of the reference samples, it can be
stated that, for the summer study, the percent error in the measured value
C of a true 00 value for the RTI-audited network would be in the interval
m 3
A A j, - A /\
T - 1.645 x CV - % Error - T + 1.645 x CV,
or
-24% - % Error - +20%
approximately 90 percent of the time.
Similarly for the network audited by EPA the percent error would be
within the interval
T - 1.67 x CV - % Error < T + 1.67 x CV,
or
-29% - % Error - +21%
88
-------
approximately 90 percent of the time.
An analysis of variance of the network audit data results in the fol-
lowing observations.
(1) The between-audits component of variability is negligibly small
compared to the within- and between-station components of
variability.
(2) Approximately 80 percent of the total variability is represented
by the within-station component of variability.
(3) Approximately 20 percent of the total variability is represented
by the between-station component of variability.
5.4.2 Precision/Accuracy Estimates for Nitric Oxide and Nitrogen
Dioxide Measurements
The results of the NO and N02 audits of the RTI-operated ground sta-
tions are given in table 11. There were nine audit checks performed at
different concentration levels for a total of 15 data points per pollutant
for estimating bias and precision of the measurements reported by the network.
The available audit data for NO show an estimated network bias of
d = -3.8 percent
and a standard deviation of the differences of
s, = 9.6 percent.
The NO audit data yield an estimated bias of
d = -3.0 percent
and a standard deviation of the differences of
s = 7.4 percent.
5.5 Summary of Audit Results
Estimates of the relative bias, coefficient of variation, and 90-percent
confidence interval for the error (deviation of the measured value from the
audit value) in the 0~, NO, and NO measurement data used in the summer study
are as follows:
• Ozone. T = -2.6%, CV = 14%, -26% - % error - +20%.
' Nitric Oxide. T = -3.8%, CV = 9.6%, -22% - % error - +14%.
• Nitrogen Dioxide. T = -3.0%, CV = 7.4%, -17% - % error - +11%.
These bias and precision estimates based on the audit data compare
89
-------
favorably with the results from previous audits of similar monitoring net-
works.— Therefore, it is concluded that the quality of the 0_, NO, and NCL
•J £•
measurements made by the monitoring network and subsequently used in the
summer study analysis is comparable to the quality of similar measurements
made in other well-managed monitoring networks and sufficient to satisfy the
summer study requirements.
90
-------
6.0 SUMMARY OF DATA AND STATISTICS
The results of the field measurement program are presented in this
section and include summary statistics, diurnal pollutant concentration
patterns, and a meteorological summary for July, August, September, and
October 1975. To facilitate discussion and interpretation of these
results, detailed hydrocarbon and halocarbon analyses of grab samples
from Bradford, Creston, Wolf Point, and DeRidder, and pertinent ratios
of compounds in grab samples collected during aircraft flights are
presented in sections 7.0 and 8.0.
6.1 Summary Statistics
Mean hourly ozone concentrations, standard deviations, and case
counts for rural and urban stations, e.g., RTI and State/local agency
stations, are shown in table 12. Valid and/or timely data were received
from only 10 of the original 15 State/local agency stations shown in
figure 1 (section 3.0) during the period June 1 to September 30, 1975.
Consequently, only data for the monitoring stations listed in table 12
were considered in this section and for subsequent analysis and interpreta-
tion. Ozone data above the NAAQS for photochemical oxidants are summarized
in table 13. Statistical data for nitric oxide and nitrogen dioxide,
selected hydrocarbons and halocarbons, and particulates (TSP, NH, , NO ,
SO, ) are presented in tables 14, 15, and 16, respectively. Mean ozone and
oxides of nitrogen data are further summarized for each reporting station
in table 17 and show a breakdown of means on a monthly basis. Table 18
also presents a similar summary for monthly mean hydrocarbon and halocarbon
data.
The mean hourly concentrations of ozone at the rural stations ranged
3 3
from a low of 58 ug/m at Wolf Point to 81 pg/m at Bradford. The range
3
for mean hourly concentrations for urban stations was from 49 to 85 yg/ro .
The standard deviations for all stations, both rural and urban, were
similar in magnitude (see table 12) with notable exceptions—Pittsburgh,
Houston, and Nederland. The overall ozone mean (or ozone burden) for the
rural stations was in most cases higher than for corresponding urban
91
-------
Table 12. Statistical summary of hourly ozone concentration
measurements by station
Station
*
Bradford, Pennsylvania
*
Lewisburg, West Virginia
*
Creston, Iowa
*
Wolf Point, Montana
*
DeRidder, Louisiana
Poynette, Wisconsin*
Port O'Connor, Texas*
4*4?
Pittsburg, Pennsylvania
Columbus, Ohio
Cedar Rapids , Iowa
Des Moines, Iowa
Omaha, Nebraska**
Nederland, Texas**
**
Austin, Texas
**
Houston, Texas
Mean Hourly
Concentration
(yg/m3)
81.4
76.3
70.4
57.6
61.2
75.8
55.3
60.5
44.4
49.9
72.9
71.1
54.6
49.2
51.3
Standard
Deviation
(yg/m3)
31.9
39.2
31.5
25.3
38.0
43.0
43.9
65.4
41.7
34.1
44.1
48.1
52.5
35.7
65.0
Case Count
2332
2386
2117
2160
2994
2428
2912
2841
2885
2781
2528
1787
2714
2504
2104
**
Rural stations [June 27-September 30, 1975, except DeRidder (June 27-
October 31, 1975)].
Urban stations (June 1-September 30, 1975).
stations in the same general area. For example, the mean hourly ozone
3
concentration at Bradford was 81 yg/m , while at Pittsburgh the mean was
3 3
61 ng/m . The lowest mean ozone for the rural stations was 58 pg/m at
Wolf Point. This mean was higher than the mean ozone concentration for
the following urban stations: Columbus, Cedar Rapids, Nederland, Austin,
and Houston. The higher ozone mean at rural stations results from
higher minimum ozone concentrations persisting at night in the rural
environment. Ozone concentrations at night at urban stations decreased
to much lower values than at rural stations and decreased much more
92
-------
Table 13. Summary of ozone data above NAAQS by station
Maximum 99th
Station H°urly ^e™&* Percentile
Concentration , / o,
(^/m3)
Bradford, Pa.*
Lewisburg, W. Va.*
Creston, Iowa*
Wolf Point, Mont.*
DeRidder, La.*
Poynette, Wis.*
Port O'Connor, Tex.*
Pittsburgh, Pa.**
Columbus , Ohio**
Cedar Rapids , Iowa**
Des Moines , Iowa**
Omaha, Nebr.**
Nederland, Tex.**
Austin, Tex.**
Houston, Tex.**
248
225
245
128
256
243
259
490
196
180
196
216
380
206
629
200
180
155
115
170
205
200
300
170
140
185
180
240
158
265
Days Days
Exceeding Exceeding
Standard Standard
(Number) (%)!/
18
11
7
0
10
22
15
41
14
1
22
15
38
10
33
18.5
11.1
7.9
0.0
8.0
21.7
12.4
34.6
11.6
0.9
20. 9
20.1
33.6
9.6
37.6
Hours
Above
Standard
(Number)
100
59
17
0
38
121
99
227
43
6
124
64
138
19
141
Hours
Above
Standard
(%)
4.3
2.5
0.8
0.0
1.3
5.0
3.4
8.0
1.5
0.2
4.9
3.6
5.1
0.8
6.7
**
Rural Stations [June 27-September 30, 1975, except DeRidder (June 27-October
1975)].
i
Urban Stations (June 1-September 30, 1975).
31,
— Based on data available from each station.
93
-------
Table 14. Statistical summary of hourly oxides of nitrogen
concentration measurements - rural stations
(June 27 - September 30, 1975)
Mean Hourly
„ . Concentration
Station , , 3,
(yg/m-5)
NO N02
Bradford, Pennsylvania 2.4 5.1
Lewisburg, West Virginia N.A. N.A.
Creston, Iowa 4.7 4.3
Wolf Point, Montana <1.0 1.5
DeRidder, Louisiana* 1.9 4.9
Standard
Deviation Case Count
(Ug/m3) N0 N0
NO NO 2
3.8 5.8 2265 2259
N.A. N.A. N.A. N.A.
4.8 2.8 2162 2162
1.0 3.4 1318 2136
8.1 9.4 2444 2444
N.A. - (NO not measured).
X
*June 27 - October 31, 1975.
rapidly in the afternoon. Maximum hourly ozone concentrations for the
3 3
rural stations were 248 yg/m at Bradford; 225 yg/m at Lewisburg, 245
yg/m3 at Creston, 256 yg/m3 at DeRidder, 128 yg/m3 at Wolf Point, 259
3 3
yg/m at Port O'Connor, and 243 yg/m at Pittsburgh. Maximum hourly ozone
3 3
concentrations for urban stations were 629 yg/m at Houston, 490 yg/m
3 33
at Pittsburgh, 380 yg/m at Nederland, 216 yg/m at Omaha, 206 yg/m at
3 3
Austin, 196 yg/m at Columbus and Des Moines, and 180 yg/m at Cedar Rapids.
The data summarized in table 13 show that the NAAQS for photochemical
O
oxidants (>_ 160 yg/m hourly average) was exceeded approximately 4, 3, 1,
1, and 0 percent of the hours at the rural stations (i.e., Bradford,
Lewisburg, Creston, DeRidder, and Wolf Point), and from less than 1 to 8
percent of the hours at the urban stations. Ozone measurements at Pitts-
burgh, Houston, Poynette, and Nederland exceeded the NAAQS approximately
twice as frequently as did measurements at the other urban and the rural
stations.
94
-------
Table 15. Statistical summary of selected hydrocarbon
and halocarbon analyses
Compound
Ethane & Ethylene—
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-ll-/
Carbon Tetrachloride
1,1, 1-Tr ichloroethane
Tetrachloroethylene
Wolf Point
_
X CC
25.3 61
3.2 61
1.3 61
2.4 61
2.1 61
0.1 61
0.8 61
1.7 61
0.1 61
2.0 61
8.9 47
7.9 47
278 42
72 61
1.5 61
207 42
Creston
«-
X CC
43.3 53
2.9 53
1.5 53
3.4 53
1.2 53
0.4 53
0.5 53
1.0 53
0.1 53
3.2 53
5.8 45
1.6 45
293 45
47 45
1.0 45
347 45
Bradford
_
X CC
27.8 60
3.9 60
1.2 60
3.4 60
4.5 60
0.2 60
2.3 60
2.6 60
0.2 60
2.7 60
5.3 36
1.9 36
271 50
61 50
1.1 50
426 50
DeRidder*
__
X CC
28.6 103
7.4 103
1.9 103
3.2 103
2.9 103
0.1 103
2.1 103
1.6 103
0.2 103
2.9 103
8.0 83
1.1 83
373 95
77 95
1.4 '95
363 95
July - October, 1975
— Concentration Units are ppbv.
2/
— Concentration units are pptv.
x - mean; cc - case count.
95
-------
Table 16. Summary of mean 24-hour partlculate concentrations:
(TSP, NHj, NC>3, SO^) for rural stations (July to
September 1975)
Station
Bradford
Creston
Wolf Point
DeRidder
TSP
(yg/m3)
34.2
74.5
29.2
41.6
NH*
4
3
0.4
0.1
B.D
B.D
(yg/m3)
0.9
18
0.4
1.3
SQ-
(yg/m3)
10.1
6.2
1.8
5.4
N0~
TST
2.6
2.4
1.4
3.1
so;
TSP
/ o/ \
\/o/
29.5
8.3
6.1
13.0
B.D - Below Detectable
*DeRidder - (July to October 1975)
Based on the percentage of days exceeding the standard and hours
above the standard, a west-to-east gradient is observed in the data.
Ozone measurements at Wolf Point did not approach the NAAQS for photochemical
oxidants; the maximum hourly average was 128 yg/m , with the 99th
3
percentile being 115 yg/m . For comparison, ozone measurements at
Creston exceeded the NAAQS: (1) 17 hours; (2) 1 percent of the hours;
and (3) 8 percent of the days during the period July 1 to September 30,
1975. Ozone measurements at Bradford exceeded the NAAQS: (1) 100 hours;
(2) 4 percent of the hours; and (3) 19 percent of the days during the
same period of time. Ozone measurements at Lewisburg were similar to
those at Bradford and showed a definite increase in the number of hours
exceeding the NAAQS for the eastern stations, as opposed to Creston.
Ozone measurements at DeRidder were similar to those at Creston with
respect to the frequency of levels exceeding the NAAQS, both on daily
and hourly percentage basis. Measurements in DeRidder continued, how-
ever, for an additional 30-day period through October 31, 1975.
The mean hourly concentrations of nitric oxide and nitrogen dioxide
3
measured at the four rural stations ranged from 1 to 10 yg/m , and fell
96
-------
o
e
^-^
^
ID
(3
O
•H
4J
2
C
0)
a
g
U
a
c
>
u
S
R
3
CO
r-x
i~H
0)
rH
•8
H
^^
^
O>
3.
I
J5
I.
4j
C
a>
o
o
u
.c
4J
c
£
c
0)
^
c
c
*r
4-
H
+
V
!_
01
"o
8
s-
01
a.
0)
CO
IA
cn
*
>,
3
at
^
o
4J
c
3
O
CL,
>
j
3
J
•)
o o o °^ ooo o o o ^ oooooooooa
ZZZlOZZZ* Z ZZ r~- ZZZZZZZZZZ
^.incsivor-voro enoro >o o>rs.- envooc\iroo.
p» ro in o 10 oo in o •— «— t\i ooo • r— t\j vo CM in r>-
vo r^in inv vo cviroi— iinpitncomin
mo* CM csioin iooe r-. m f~. vo i- «• o ^- PO r^. en
co f™" ^c n ^o ^^ ^^ i^ '•" 1™* oo f** oo c^ LO ^f ^t f** CNJ f~ ^"*
n in — «• CM i» vo tMo o.
O) CM ^T OO ^^ '^ C\l »•• ^~ CM O ^3 f) CTv ^~ CM n ^f 1O 1O ^f
m in oo r^. in in in CM ^ in
CM CM CM CM
O Z Z O O ZZ OZZ OZZ O O OOO O O O O O
* •»(
10 *
* •>- c4(<0
> T- C C r> f- * ID « X
•— > re >in****fl)
>> *-> T- ift*c+ieoin^-*
u> +J C f C*OlOSi-«lO**
c in * o -r- co u .* o x «* i«
c: cyio "Si :s ai'^mwi— '»att,4/iiD
0> at z o Ck fi- ID in h- oiox
CL. O • _J O3 \- '-O C X CU
• 1—1 +•> - ^i c • t7>«_i»oicu tx-o o (—
I. 3C 15 CU 53*->i-oi:iC-«C
O £O O. -O -O^-U-O^-OCO
IK t/)4-> *o t/)£O)io£&-&.*p-4^
•O i-in It- f 4->3C.CIOO>4->4->V>
* ? O) -- Ct *>r->,gXll.l«3
i. OIL. O 0) «-OOEOlO!O>O2O
m -jo x o a. o a. o o cj z a. < :x
^L
in
r^
en
«
^«
«
t.
01
.a
0
o
J=
cn
|
jC
i—
IO
**
Q
1
cr>
i-
1
1
g
ID
0>
•o
•o
ce
0
CL
CO
U
X
OJ
__
in .-^
r^ in
cn r»
r— en
^- T3
CU
o - -5
n O 3
CO •—
I- U
CU S- C
^ CU O
•M cu E
a. *-> m
ai CL s-
t/> a) cn
I I/) O
r^ i ^>
CM i- 0.
9)
UO IO O
z
r— C
IQ 10 1
i. -O
3 S.
ce => o
* ; s-
97
-------
S
o
•H
4J
cd
01
M
H
4-1
C
>%
J3
CO
4-1
cd
n
4J
o in
g f*
S o>
O rH
a n
o ai
•fi-8
cd u
U 4-1
o p.
r-l .
^3
g
§
4-i
O
w
00
H
*
i
M
§
•>
1
«B
I
*>
&.
«
T
£
^
CD
!
«
!
V)
£
S
i
c
£
i
i
j
L.
^
>.
&
in
*•
!
>»
.?
i
&
v»
*»
i
X
3
I.
H
*»
i
>>
•?
L.
a.
«
*»
<
X
J?
!
|
rv en 00 •- o> t— r» in •- oo •- •»•
<»> 10 »— CM CM o •— *— o CM coo
i/)i*r^r^o »— eft •» CM •" m ©
C* «A f~ CM m O ••- CM G CM GO CM
^" V
f*-f—9iOiir> i— f— r** « •— — f-iceteee CON
F* \O r- «• evi O •— O O •— «•-
eu>m «»wcovo r^r^
CMUif-vio om»now mcvj
r^o-meiO turo^-cxm
•wtnf^diin ocMCMOf— 1
OtfChmcsi •— in CM •— CM r^oo
RCMOI-Of- O O •— O CM COO
V V
»p— itSCMie o«*^ei— vo«t
f— */> O CM O OOOOCM <•*—
GOOOI/*f*t (Mh«.p*CM*JO- »»r^
o» •— m »— *— •— o *~ o m vcu
nwcMtnoi t— « f> f> co CMO
e m •— »• n o^-«no>- m^-
r- v r-
tne>ov »«- r*. CM »— \n «••»•
«•»•-•—•— o o •— e CM inco
•— v
CM-a-vote f— me mve
CM — f- «r o o o o o — icai
in v •—
^
S
i
4-*
uj y
i iis.giis
|£^l| 1 5 | 1 5 |g
5S. o! i" 3 's ^ S. ^ £ S 'x
o o ai co Boeue. ~ £
4Wb£U! IIAIA>|I OI
ujC4.o. tn co
P_
eo r*. p— f*fc
S * §
o
m i-o f— oo
_ in «r
CM m
CM *— f— CO
CO «T CM
CM m
cr>
e $ ° a
csj r\j
o*
0*1 ^ o <*>
i— in i—
n «•
O
CO O» ^- *
r-. «s- m
CM CM
\O
0* CO r- O
h^ ? OO
f\J .—
I * I
1 8 - 1
^y
m Tetrachlorlde
-Trlchloroe thane
chloroethylene
g * .-"S
L. IB » 0;
u. O f- 1—
> i
& Q.
g. &-
i|
s *
*f
C 0
O f
£> +>
1 1
i. I.
i i
Vt (A
t t m
g. g. &
- - fc
£ S 1
" " S
i it
5 i •?
^i*
98
-------
for practical purposes within the noise or detectability level of the
3
measurement method (~ 10 yg/m ). These data, presented in table 14, are
shown to emphasize the extremely low concentrations of measured NO in
.X
the environment at the rural stations. Maximum hourly nitric oxide and
2
nitrogen dioxide concentrations were 34 and 68 yg/m at Bradford, 28 and
33 3
25 yg/m at Creston, 3 and 19 yg/m at Wolf Point, and 17 and 43 yg/m
at DeRidder, respectively.
Grab samples collected at four of the five rural stations were
analyzed for selected hydrocarbons and halocarbons. A summary of these
data is presented in table 15. Each value represents the mean concentra-
tion over the measurement period for that component at the four stations.
Averaged over the measurement period (July through September, except
DeRidder), these data show similar concentrations for selected hydro-
carbons and halocarbons. A definite trend is not apparent in these data.
Mean suspended particulate data given in table 16 show similar
particulate loadings for Bradford and Wolf Point. Suspended particulates
were as high at Bradford as at Wolf Point, with DeRidder falling in
between. Further examination of the data shows a definite trend with
respect to the sulfate and nitrate components of the suspended particulate
matter; the percentage of sulfates and nitrates increases progressively
from western to eastern stations. For example, sulfate represented 30
percent of total suspended particulate (TSP) at Bradford, 8 percent at
Creston, and 6 percent at Wolf Point. Nitrates as a percentage of TSP
showed a similar pattern, although not as dramatically.
Mean monthly ozone concentrations did not change significantly during
the period July through September at Bradford, Creston, and Wolf Point,
or at DeRidder from July through October; however, a gradual decrease
was observed for Bradford, Creston, and Wolf Point and a gradual increase
was observed for DeRidder. Mean monthly ozone concentrations at urban
stations generally decreased during this period of time. Notable exceptions
to this trend were observed at Houston, Austin, and Nederland. Mean monthly
concentrations for nitric oxide and nitrogen dioxide at the four rural
stations showed little variation during this period. Similar results
99
-------
were observed for mean monthly concentrations for selected hydrocarbons
and halocarbons; these revealed no obvious trends.
Cumulative frequency distributions for hourly ozone concentrations
for rural stations are presented for the entire period in table 19 and for
State/local stations in table 20. Cumulative frequency distributions for
hourly nitric oxide and nitrogen dioxide concentrations are given in
table 21.
6.2 Diurnal Patterns
Mean ozone concentrations for each hour of the day for RTI rural sta-
tions are shown in table 22 and for State/local stations in table 23.
Table 24 gives mean nitric oxide and nitrogen dioxide concentrations for
each hour of the day for four of the five rural stations. Mean diurnal
curves for ozone are shown for rural and urban stations in figures 38 to 42.
Figures 38 and 39 present, respectively, the mean diurnal curves for Brad-
ford, Creston, and Wolf Point, and for Bradford, Lewisburg, and DeRidder.
Mean diurnal ozone curves for the stations at Pittsburgh, Columbus, and
Poynette are presented in figure 40; at Des Moines, Cedar Rapids, and Omaha
in figure 41; and at Nederland, Austin, Houston, and DeRidder in figure 42.
The mean diurnal ozone curves for the rural and urban stations show the same
general pattern; however, several observations can be made, as follows:
1. Mean hourly ozone concentrations for rural stations range
from approximately 20-55 yg/m3 in the morning to 80-115 yg/m3
in the afternoon, while for urban stations, the mean ranges
from 10-50 yg/m3 to 70-130 yg/m3.
2. The overall range or spread of mean ozone concentrations
for both rural and urban stations (from low to high) was
approximately the same (i.e., 60-70 yg/m3).
3. Mean hourly ozone concentrations peaked in the afternoon,
at rural stations from 1600-1700 and at urban stations
from 1400-1700.
Figure 43 presents the mean diurnal ozone concentration curves for
Kane, Pennsylvania, in 1973; for DuBois, Pennsylvania, in 1974; and for
Bradford, Pennsylvania, in 1975. These curves represent ozone measurements
during similar time periods at three sites that geographically, clima-
tologically, and meteorologically are similar enough for comparison
100
-------
co
C
o
4-1
5
CO
p^
cfl
M
3
1
CU
e
o
N
O
M-l
O
co
o
4J
Cfl
iJ
a /-*
cu m
o r"»
8 H
*
£*> o
iH fO
3 M
O 01
J"] o
tii S
0 4J
CO CU
O 1
•H r^
4J CN
,0 CU
•H C3
J-J ^
CO "-s
•rl
^
5
a"
cu
4-)
CU
45
•l-J
tB
^H
3
a
3
O
•
~j
CU
H
rrt
H
o
tc
£
g
5
^
«
en
c
^
01
10
£
0
tn
9>
IB
S-
*
t.
^_
O
g
u
I.
c
o
z>
(Of
^
o
0
-1C
c
ID
VI
a
o
_l
,_•
V
•o
C£.
10
c
ID
e
£
c
"5
Q.
<*-
*
1
*"N
|
in
0>
c_5
^
to
c
t.
*
1
„
?
y
J2
1
J
ID
C
to
*>,
(A
C
Ol
D.
f
0
«4_
"O
L.
CO
p"^«
1
3.
HMT
OiOOienr". (TitNirowr— MO000000
»—
oocT>lr,^eo«.^^-cvJ^w«.-oooooooeoooo
Qoeocvin.— o<0noo«->-oooooooooooooo
^7 C? eft 0*1 co r** ^5 *f ^o cvj ^*~
oe09>mineocoi^umootv)rxir>mmo-^ooo ooooo
888S;&!.S^S;5SS2"eo<*cs"~0000000000
»— • f^ r—
oa.in^ocNjcs.^o-r-.wcvj^eoo-iowr-inr-ooooo
§S§:!a!§§2SK?Sf3S20""'0'Cs"~''0000000
^
OOr«»*— ^-f\ic\j^o>tnor*.KDr— f^J^o^•CH^^coror— vcxiooo
oocotft^— ioa*oc»o*cor«wO>^'Or*-u>(v>c\j^-F— p— ooooo
OOOOOOOOOQOOOOOOOOOOOOOOOO
in
Ot
*
f>
1
^
eg
4)
I
101
-------
r-|
to
O
0
,— |
0)
4-1
tfl
4—)
en
I
cu
C
O
N
O
O
a
o
•H
11
cfl
>-i
4_1
ri ji^s.
Q) LO
o r-
C ON
O rH
O
rl
^ Q)
T! "fa
3 CU
O 4J
J3 P.
CU
14-4 C/3
0 1
0)
co C
C 3
0 1^
•rl -~"
4J
3 to
Xi C
•H O
4-1 4-1
CO tfl
•H 4-1
T3 CO
t*<
C
CU
3
o4
0)
>
•rl
tO
, — |
3
O
o
CM
0)
3
H
c
o
1C
L.
C
01
u
c
o
Ol
IO
4->
l/>
C
(C
H-
fc
01
o
0>
Ol
1C
01
^
_>.
11
1
^
o
•^
01
u
1-
o.
c
0
0 t-
3;
c in
•i- (0
*J X
W O)
3 r-
S-
o
c
c
O i/l
X
O 01
t—
i_
o
o-
•1
fD (A
r— ID
i- X
CD Ol
Z
*
t/)
^».
O.
fO
Q£ (O
ID *-i
5
*
IA
0)
•c
O (C
V> I-*
iS
*"«
1 2
i-8
z
. c
+J *^
"S c
>> s
o 42
'** jfc
So
g x:
o
U
«
*1
l>
** i"
^1
c
o
+J *^ ""i
1%
S 3
u^-
I
Ir-i—t—OOOOOOOO
IP-I—OOOOOOOOOOOOO O
ooooooooooooooooooo
t— ooooooooo
§
o>—
oescJo'odo'ooocJcJdcJo'cJodooo
§
r-oO«9-CMr-OOOOOOOOOOOOOOOOO
'— «oc\4r— ooooooooooooo
8
lftfOOfO«fl-
r— OOOOOOOOOOOOOOO
o o d o d d o d d d d <
§
J P— O l/i CM O O
ir-^i-^r-^r-^^i-^dddd
OOOOOOOOOOOOOOOOOO
^cvjrowmiorxcooiO'— tM O O O <
I CM CM O P O O
> ro ^ m *o
102
-------
1
c
0)
60
O
4J
•H
f3
4-<
o
CO
T3
•H
X
O
0 iH
£ u
O M
O 0)
. Cl
iH
•H
4J
ca
3
s
3
U
,
rH
CM
0)
rH
•a
H
|p
re
t.
+^
§
J
^5
+j
(0
4J
00
c
1
01
-i »
re
01
(9
trt
re
?
^
§
3:
n-
o
c
O)
a.
c
c
•V
4
I
<
J
<_
*
4r
li
i/)
3
O
Ol
5Z
0
10
*§*
^1
c
*^>
5
u.
oi
i
I
1
a>
i.
og
10
"c
re cj
V)
01
a.
f
•o
2
CD
)
J^«.
rt
gj
i
3
OO»— OOOrOCMO
§00 CO «»• i— O O O
«»• F—
•"•
OM»-WP-'— oo
oaiaiooooo
'—
O o' o'
o^
'"
oo^-ooooo
OtnvOroOOOO
F"
O O> i— r- U> f— OO
orotntsioooo
*"""
otooomoooo
8Q V O O O O O
ifi
o o o o o o o
•— CM ro » tf> vo
re
c
o>
1-
>
4J
X
e
ja
VI
"» m
* en
~J r—
re »—
•o
ai !.
ID +J
?
O CM
Ol
X C
* «
*
103
-------
o
•rl
4J
nJ
M
H
S
I
a
13
o
M
o
&
CO CT>
(!) H
J-l •>
O O
0) )-(
C QJ
N 'B
O CD
4J
u-i ex
O 01
CO
SrL
O CN
•H
4J 0)
S §
•u •-)
co
o
d
o
u
§
en
C
I
CN
CM
•8
H
en
C
o
c
o
*
ID
C.
ic
•*•
•f*
o
1
OL
1
*J
C
°o
a.
1/1
£
LeMlsburg,
1C
"c
10
in
c
1
»
•o
•o
1C
00
!_
O«TOVOr^l£VOVD(S10OO>
in
o<
§ § § §
§
i 8 § §
*— tsj
CM
Ol
c
s
pu CM rj
104
-------
CO
s
o
£2
ir> &t ID CO O 1C
O"i 00 CO 00 f"* W>
oca
ate
(_> (D
X
O «Jt£>*£>i£>ir>t/>uiu->iri
(4-1
O
1
fl? V)
•6
cfl ,-x
o) m
M CT\
O rH
IW
M
O 0)
g-i
M CU
O 4J
&
M-l 0)
O CO
I
CQ 01
a a
O 3
•o
. 'E.
IP
QJ
C
I|
(A «
£
rn eo «; co C5 r~ oo
<\j .-' oi to co «•' r-'•
concen
I-
S 8 §
3
O
M-l
O
co
cfl
0)
a
CO
CM
Jg
3i
•§
H
I
888§§§88§So888S88oo§8888
fx>COCri
105
-------
CO
(3
O
•rl
Cfl
H
T3
<4H
O
3
O
•s
ffl
CD
c
01
•H O
f* CO
4-1 V-i
O CU
co -I
(
—;
i
a.
c\JpJrorocjp>jcO{M -'.-'.-• — -.
*4->
f—
-------
0. O -
1- C O
O O O.
<4- +J
TT 10 >—
s- i- o
CO U 2
00
CM
O
CM
CM
O
CM
T3
to
->
o -^
N
O 4-1
rH-5
td o
3 M-l
•rl rH
oo
CO
60
•H
o
CNJ
auozo
107
-------
to
C
«J
>
O)
S- «J
-r- C
_
C d) 3
OJ 3 O
Q. _l
-O S- S-
$_ 3 O)
O -Q "O
<4- i/l T3
"S S 5
S- O) 0)
CQ _l Q
• O D
o
«d-
S
HCM
O
. CM
CM
d) •
C ^
3 l-i
I-) 0)
T3
C
cd
00
O)
en
oo.,-
^
(O
Q
«J
10 o
" CO
M -H
01
•H 4J
S ni
"d .-t
^
o •>
pq Q)
*.•§
n) 4-1
O
S?
O r-
•H CM
4-1
td 3
f-
A
, , , J-
i °\ '
O D
J 1 -
o
CM
O
o
o
CO
JlO
o
cu
o >~
0 T3
O !-i
o o
m
Q) T3
c to
O >-i
N cq
O
13
«l
C O
n) co
0)
_
CO -U
P<
a) a)
(-1 t/1
00
•rl
( ui/!5rt)
SUOZQ
108
-------
oo in
UJ O O
O. HH (/)
C5
o:
t = *
E g 5
0. <_> O-
O O •
60
M
a
,0
CO
•u •
4J ^
•H IT)
CO )-l
(3
-------
O IQ
s- -o
-Q -c-
-------
_ _
'7- C T> O
-
T3
4)
< o
o
OJ O
C m
o i-»
•u ON
CO rH
O K
PS CU
,0
" O
C 4J
•H O
•W O
CO I
p >.
a i
to M
rH CU
H 13
CU T3
"O -H
O O
O CO
CU )H
a cu
O ,0
N e
O 0)
4J
i-H CX
cd cu
SC/3
I
3 rH
•rl
T3 CU
0 §
n) f-j
a) ^-
a)
• T3
CM T3
-* -H
&
CU CU
auozo
111
-------
Lf)
^^ r*»
•» 01
^••^ T3
(/) O
•i- <*-
O -O
CD «O
o o
ro
cd co
CO 0
C (1)
o c^
•H
•U •>
CO ^
0)
a
o
u
N H
o m
I
o
ffi
( Ul/6rt)
3UOZQ
112
-------
purposes. Figure 43 shows that the 1973 and 1974 diurnal curves are quite
similar, while the diurnal curve for 1975 is substantially lower. The
mean difference (diurnal variation) from ozone minimum to maximum for
33 3
the 3-year period was 80 Mg/m for 1973, 100 ng/m for 1974, and 60 ug/m
for 1975. Similar observations were observed in diurnal ozone trend at
the urban station in Columbus from 1974 to 1975 (fig. 44).
For each of the four stations where measurements were made, mean
diurnal curves for nitrogen dioxide are presented in figure 45. Mean
3
hourly nitrogen dioxide concentrations ranged from < 1 to 8 yg/m . All
of these measurements were at or below the generally accepted minimum
3
detectable level of 10 yg/m for the measurement method. These data are
shown, as previously stated, to emphasize the extremely low concentrations
of NO measured at these rural sites. Better resolution instrumentation
x
with much lower minimum detectable levels are obviously needed for back-
ground measurements of NO in rural areas.
x
6.3 Summary of Climatic Conditions
In planning this study, RTI estimated that one to two candidate
high-pressure systems could be anticipated for each of the 3 months, July
through September. Six high-pressure systems were investigated. During
the study, high-pressure systems tended to move south, to turn eastward in
the central plains, and to move rapidly toward the east/northeast. It was
difficult to keep up with the system and to perform the sampling at the
desired locations. Of the systems studied using the aircraft, none stag-
nated. Secondary high-pressure systems following those studied did slow
down and remain over the Eastern United States for a longer period of time.
Movements of high-pressure systems for each of the 3 months are shown in
figures 46, 47, and 48.
Due to the high mobility of the systems, annual changes in emission
patterns or quantities are not routinely documented. The great year-to-
year variability of ozone concentrations measured in the Eastern United
States is usually attributed to changes in meteorological conditions. The
correlation of ozone concentrations with climatic indices such as tempera"-
ture, moisture, or wind over a long period of time is very poorly established;
113
-------
IT)
P-
V) in
3 3
-i -i
13 3
"o 'o
CJ O
o
•H
CO
o
o
4J
td
CO
g
•H
4-1
a
CD
O
§
CJ
8
o
N
O
0)
I
•H
13
0)
^
00
( ui/6H)
114
-------
c
> c c
l/> C
C O •!-
C 5> Z 3
Q.
£ t
to o
- D O
H
• D. O
I
^"° 1 \l
1 1 X ?
"O ^ 1
i— oe • O
So X XI
• D Vv
/^/D
/*/!
NM
Xsf^
o •
1 i
{ 1\
I i i
CVJ
O
§
10
ID
Q
a
o
4J
CO
-------
m
0)
4-1
o
-------
g
u
CO
3
CO
co
<0
M
&.
•a
co
•a
n)
0)
M
a
117
-------
0)
P.
-------
as a consequence, it is difficult to assess the influence of climate upon
these conditions. In section 8.0, some of the temporal relationships
of ozone and meteorological conditions are examined.
High ozone concentrations have been associated with the transition
of high-pressure systems in previous studies as well as in this study.—
The upper air flow patterns are examined for the 700-mb circulation dif-
ference for the year 1975 from the 30-year climatic norm.
21/
A. July 1975—7
The average position of maximum winds at 700 mb during July lay just
north of the Canadian border. A ridge line initially over the Midwestern
United States retrograded to the Rocky Mountains by the end of the first
week. The trough on the east coast moved to the central plains for most
of the month. The subtropical ridge began to dominate the circulation
patterns at 700 mb over the Southern United States later in the month.
The ridge over the Rockies, trough over the Central, and ridge over the
Eastern United States, gave greater north-south movement to systems than
is usual. This situation provided the opportunity for more high-pressure
systems to move southeastward out of Canada, into the Central United
States, and northeastward from there. Temperatures were above normal by
as much as 3.3° C in the eastern Dakotas and in southeastern Montana.
Above normal temperatures were also found over most of Kentucky, eastern
Ohio, New York State, and in New England.
221
B. August 1975—
Pressure heights of 700 mb were higher than normal over the eastern
half of the United States, the maximum departure being over the Ohio/
Pennsylvania/West Virginia region. Lower than normal heights prevailed
over the Rockies. The subtropical ridge extended along 32° N, with one
center off the Carolinas coast and another center over Arizona. The
resulting circulation pattern shows strong west-to-east flow from coast
to coast at 42° N. Pressure systems moved quickly from west to east
without north-south movement. None of the fronts penetrated south of
35° N during the month.
Lower ozone concentrations were generally found at urban and nonurban
locations of the northern study. Increased ventilation by more transient
119
-------
systems may have been partly responsible. Temperatures were above normal
east of the Mississippi River in all of Missouri, Kansas, Nebraska, Iowa,
and in eastern South Dakota. Departures of more than 2.2° C occurred in
the Creston area, about +1.0° C at Bradford and Lewisburg, and about
-2.2° C at Wolf Point.
23/
C. September 19 75—
The trough at 700 mb was reestablished over the Central United States
in September. The subtropical pressure ridge appeared slightly weaker
near 30° N, with centers displaced farther eastward over the Atlantic.
Pressure height departures from the mean are negative over the Central
United States while they are most positive over British Columbia and
off the Carolinas coast near 35° N, 70° W. The axis of maximum wind lies
just north of the Canadian border but shows low speeds. The pattern of
departures is more variable, farther west and farther north than in July.
High-pressure systems followed the long wave pattern, entering the United
States near Montana moving southeastward, then accelerating northeastward
off the coast.
In September, there were ::ew events of relatively high ozone concen-
tration at both the urban and the nonurban locations. Most of the average
afternoon concentrations were below mean value of the 3 months. Tempera-
tures were cooler than normal east of the Rocky Mountains except on the
South Atlantic coast. Departures of -2.8° C or more were common along a
240-km-wide swath from southwest Oklahoma to Detroit.
24 /
D. October 1975—
A southwesterly flow at 700 mb over the Gulf Coast area was induced
by a long wave trough at latitudes near 30° N, situated near 115° W.
Heavy precipitation was reported east of the study area. Temperatures
were slightly warmer than normal over most of the area.
Four frontal systems passed the Gulf study area and one tropical
storm moved out of the Yucatan Channel into the Mississippi coast. The
fronts took a day or more to pass the area, since they were generally
weak and did not penetrate much farther southward.
120
-------
7.0 INTERPRETATION OF RESULTS: NORTHERN HIGH PRESSURE OXIDANT STUDY
7.1 Examination of Aircraft Ozone Measurements
7.1.1 Data Analysis Approach
The objective of the northern high pressure oxidant study was to gather
aerometric data from an aircraft and strategically located ground stations
to determine the change in the concentration of ozone in the center of a high
pressure system, as the system moves from a region of low population density
(Canada and northern great plains States), south of the Great Lakes (high
population density), and out into the Atlantic Ocean. Thus, the mission area
was bounded roughly by the Eastern Rockies to the west, the Canadian border
to the north, the Atlantic Ocean to the east, and the 37th parallel to the
south. This area was divided into three sectors, each with relatively uni-
form population density and similar land use patterns:
(1) Western sector: west of a line from Fargo, North Dakota, to Dal-
las , Texas, including the rather sparsely populated Great Plains;
(2) Midwestern sector: between longitudes 96° W and 86° W (between
line from Fargo, North Dakota, to Dallas, Texas, and line from east of Chi-
cago, Illinois, to St. Louis, Missouri) including the more densely populated
midwestern farm belt;
(3) Eastern sector: east of the line from east of Chicago, Illinois,
to St. Louis, Missouri, including the densely populated and highly industri-
alized area from Ohio to the eastern seaboard. All data from a flight were
categorized by the sector containing the center of the high pressure system.
In an attempt to present a roughly homogeneous data set, the aircraft
flights were then screened for altitude consistency, time of day, and dis-
tance to the center of high pressure.
(1) Altitude: due to the areal extent of these missions, terrain
differences, air traffic density, and FAA controls, it was not always pos-
sible for the mission aircraft to maintain a constant altitude throughout
the flight, or day after day. Thus, for analysis, only those flights that
averaged 915 to 1220 m MSL were considered. Also, it was decided that
the flight level (expecting approaches, departures, and vertical profiles)
should not vary more than 305 m during a single mission. Data outside
121
-------
this altitude range were not considered;
(2) Time of day: in order to minimize diurnal effects, only mission
data taken between 1300 and 1900 local daylight time were included;
(3) Distance to high center: to be considered for analysis, the
flight had to be almost wholly within an 800 kilometer radius of the
center of the high pressure system.
7.1.2 Summary of Aircraft Ozone Measurements and Meteorological
Conditions
Based upon this selection process, portions of 14 flights were chosen
for analysis (table 25). Two of the six high pressure systems investigate
produced data in each of the three geographical sectors: July 24-25-26 and
September 5-6-7. Both of these cases showed ozone concentration increasing
from west to east (table 26).
In the July case, the anticyclone moved southeastward out of Alberta
at about 30 knots and was centered in central Nebraska at the time of the
July 24 flight (figure 49). Ozone concentrations in these figures are for
each 10 minutes of the flight which is shown in its entirety. The July 24
flight roughly paralleled the track of the cyclone some hundred kilometers to
the east of the system's center. The July 25 flight also followed the tracks of
the anticyclone, very nearly passing through the center of the system in
south central Iowa (figure 50). On July 26 the aircraft trailed the high
center into Pittsburgh after tailing an excursion to the northeast into
southern Ontario (figure 51). At this point, the high pressure system
slowed drastically and dissipated as Tropical Storm Blanche moved up the
eastern seaboard. In the September case, both anticyclone track and air-
craft flights were geographically similar to the July mission except that
the system was followed into southern Maine.
Ozone concentration was measured continuously during the flight. All
measurements taken at mission altitude were averaged together for each flight
to produce the data in table 27. Spikes associated with urban areas were not
included in these averages. These data demonstrate the general increase in
ozone concentration from the west to east. The average ozone concentration
was computed for each flight. The results are presented in table 27 along
with computed averages by geographical sector. An increasing west-east ozone
122
-------
Table 25. Candidate missions by sector
Sector Date
WEST 09 Jul
10 Jul
11 Jul
24 Jul
13 Aug
05 Sep
MID-WEST 13 Jul
25 Jul
06 Sep
27 Sep
EAST 26 Jul
07 Sep
28 Sep
29 Sep
Local Daylight Time
1300 CDT
1356 MDT
1300 MDT
1300 CDT
1608 MDT
1326 MDT
1306 CDT
1300 CDT
1314 CDT
1300 CDT
1300 EOT
1300 EDT
1336 EDT
1304 EDT
- 1344 MDT
- 1622 MDT
- 1442 MDT
- 1416 CDT
- 1900 MDT
- 1718 CDT
- 1612 CDT
- 1520 CDT
- 1838 CDT
- 1848 EDT
- 1548 EDT
- 1856 EDT
- 1430 EDT
- 1558 EDT
Altitude (s)
(Kft MSL)
4.0
4.0
4.0
4.0
4.0
4.0
4.5 - 3.5
3.5
4.0
3.5-3.0-3.5
3.5
3.5 - 4.0
3.5
3.0-3.5-4.0
Avg Speed of High
Center (Knots)
10
Stationary
10
27
22
Indeterminate
40
30
22
15
13
38
17
33
Table 26. Average ozone concentration at mission altitude
CASE
July 24-25-26
September 5-6-7
WESTERN SECTOR
MID-WESTERN SECTOR
38 yg/m3
65 yg/m3
94 yg/m3
80 yg/m3
EASTERN SECTOR
101 yg/m3
98 yg/m3
123
-------
51
37
37
SOUTH 3
DAKOTA
Figure 49. High pressure system flight on July 24, 1975.
124
-------
L
TEXAS
Figure 50. High pressure system flight on July 25, 1975.
125
-------
WISCONSIN
IOWA
.20
MISSOURI
'MICHIGAN
NEW YORK
(J|37 PENNSYLVANIA
ILLINOIS
I23S
£DTl
N08
OHIO
v
"ft
N.H.
M.D,
INDIANA
KENTUCKY
TENNESSEE
U-»-
O
N.J,
WES]
VIRGINIA
NORTH CAROLINA
Figure 51. High pressure system flight on July 26, 1975.
126
-------
Table 27. Average ozone concentration for each operational sector
NUMBER OF 0.
SECTOR DATE MEASUREMENT!
AVG 0, FOR
FLIGHT STANDARD
i (yS/m ) DEVIATION
MAX 0
(ys/m )
MIN 0
3
(pK/ra )
RANGE
3
WEST 09 Jul
10 Jul
11 Jul
24 Jul
13 Aug
05 Sep
Totals & Avgs
MID-WEST 13 Jul
25 Jul
06 Sep
27 Sep
Totals; & Avgs
EAST 26 Jul
07 Sep
28 Sep
29 Sep
Totals & Avgs
53
70
35
39
23
66
286
78
43
79
72
272
59
111
26
70
266
66
72
60
38
61
65
60
86
94
80
124
96
101
98
131
95
106
4.4
6.0
4.1
2.9
3.4
5.3
4.4
6.5
12.9
6.7
19.6
11.4
18.2
22.4
14-5
7.9
15.8
72
81
72
42
67
78
69
104
123
96
170
123
157
140
146
120
141
42
57
57
32
57
60
51
70
75
66
74
71
55
54
90
78
69
30
24
15
10
10
18
18
34
48
30
96
52
102
86
56
42
72
127
-------
concentration for July and September cases (table 26 can still be seen even
when all data from all flights are combined for averaging (table 27) .
The spread between maximum and minimum ozone concentration values also
increases west to east, which may be the result of a build-up of ozone on
the back side of the high pressure system once it moves east of longitude
96° W. This phenomenon is discussed in Section 7.2.3. Other measures of
ozone concentration that increased west to east were the average maximum
and concentration variability as indicated by the average standard deviation
of the mean values.
7.2 The Relationship Between the High Ozone in the Rural Boundary Layer and
High Pressure Systems
7.2.1 Introduction
3
That high ozone (ozone concentrations in excess of the NAAQS—160 yg/m )
4579 25/
exists in rural boundary layers is a well-established fact.—2—*—!—z— High
ozone has been attributed to the formation of ozone by anthropogenic precur-
sors and correlated with solar radiation and with wind speed; that is, high
ozone concentrations are found with high levels of solar radiation (suggest-
ing synthesis as a source of high ozone) and also with low wind speeds.
A more recent study has suggested that the occurrence of high ozone in
the rural boundary layer is widespread in the eastern portions of the United
States and is correlated with the presence of high pressure systems.— This
study has shown that when a synoptic high pressure system moved into the
eastern portions of the United States, high concentrations of ozone were
reported at a number of rural stations scattered throughout the region.
This condition persisted as long as the environmental conditions (e.g., high
solar radiation, low wind speed) accompanying high pressure systems remained
in the eastern portions of the United States. In this 1974 study, the five
highest ozone concentrations were found at rural stations located near the
central regions of the high pressure system. These regions were character-
ized by weak winds, disorganized flow, and relatively clear skies. The
lowest ozone concentrations were found after frontal passages.
Data collected by the Research Triangle Institute in the summers of
1973, 1974, and 1975 have besen used to investigate the relationship between
high ozone and high pressure systems. In 1973 and 1974, the data were col-
lected in the eastern portions of the United States at various stations
128
-------
located in rural boundary layers. However, in 1975 stations were located
across the United States at Wolf Point, Montana; Creston, Iowa; Lewisburg,
West Virginia; and Bradford, Pennsylvania. The following summarizes the
results of the studies using these data.
7.2.2 Statistics on High Ozone Concentrations Versus High Pressure
Systems
Figure 52 shows the daily area-averaged pressure and the daily maximum
ozone concentration for the summers of 1973 and 1974. The ozone data repre-
sent the average value of the diurnal maximum ozone concentration observed
at each available station. An averaging technique was used in order to re-
move local anomalies which may exist at a given time and station, and to
demonstrate the systematic nature of the increase and decrease in ozone over
a fairly large area. The rural stations used in 1973 were Kane, Pennsylvania;
McHenry, Maryland; Lewisburg, West Virginia; and Coshocton, Ohio. In 1974,
they were McHenry, Maryland; McConnelsville, Ohio; Wooster, Ohio; DuBois,
Pennsylvania; and Wilmington, Ohio. The area-averaged pressure is the
average of the daily average pressure for a synoptic weather station nearest
to the respective rural stations. Figure 53 shows the daily ozone maximum
and the average pressure from the nearest synoptic weather station to three
of the ozone stations for the summer of 1975.
The data in figure 52 and 53 show the relationship between high ozone
and high pressure systems; that is, high concentrations of ozone at the
surface were found near the time a high pressure system was located in the
region of interest. The 1975 data (figure 53) further indicate that high
ozone concentrations were found in high pressure systems most often when
these systems were located in the eastern part of the United States. This
can be seen by noting that many high pressure systems pass through Wolf
Point, Montana, but at no time did the concentration of ozone reach values
greater than the NAAQS. However, relatively speaking, when the high pressure
systems were found over that station, the ozone concentration at times
exceeded the NAAQS at Creston, Iowa, when high pressure systems moved
into the region, but not as often as when similar systems were located near
Bradford, Pennsylvania.
129
-------
oat
ta
§i
is
IA
Z»
Figure 52. Area averaged value of the daily maximum ozone
concentration (solid line) and the area averaged surface
pressure (dashed line) versus day of month for latter
part of the summer of 1973 and 1974.
130
-------
i 985—
I 1 *7S-
2 »
fi I W5-
.». Oi(l —
a*
m
5
950-
940-
930.
920.
Pittsburgh
Oes
-.995
— 985
— 975
das glow
3 12 22 1 11 21 31 10 20 30
JULY AUGUST SEPlBSER
230—.
200 —
-•5 150-
100 —
so —
0 —
ISO —
100.
so —
0-
Wolf Point
12 22
JULY
I
11
-.250
—200
—100
—SO
—0
I I I
1 11 21 31 10 20 30
AUGUST SEPTS0ER
Figure 53. Daily maximum ozone concentration at Wolf Point,
Montana; Creston, Iowa; and Bradford, Pennsylvania
and the daily average pressure obtained from the
nearest synoptic station for the summer of 1975.
131
-------
The data in table 28 support the association between high ozone and
high pressure systems. The table shows that in 1973, between July 4 through
September 3, there were more hours in which a high pressure center or a ridge
was within 720 km (450 mi) of the rural station in Pennsylvania than in 1974
or 1975. There is no significant difference between the total number of
hours of high pressure in 1974 and 1975. However, there is a marked differ-
ence between the total number of hours ozone exceeded the NAAQS over the
3-year period. The greatest number of hours was in 1973, and the least was
in 1975. The data also show that over the 3-year period, on the average,
about 85 percent of the hours in which ozone was greater than the NAAQS
occurred when a high pressure center or a ridge was within 720 km (450 mi) of
the station. On the average, approximately 93 percent of the hours of high
ozone occurred when the station was in the same air mass, regardless of the
distance of the high pressure center from the station.
Table 28. The relationship between the number of hours a high pressure
system is near a station and the number of hours of high ozone
(greater than the NAAQS) observed at that station for the period
4 July to 3 September. In 1975, the station used was Bradford;
in 1974, DuBois; and Ln 1973, Kane
YEAR
1975
1974
1973
A*
773 hrs
781 hrs
1015 hrs
B**
92 hrs
346 hrs
552 hrs
c***
83 hrs
277 hrs
471 hrs
90%
80%
85%
D****
87 hrs
301 hrs
521 hrs
95%
89%
94%
*Column A gives the number of hours a high pressure center or a ridge is
within 450 miles of the station.
**Column B gives the number of hours of high ozone, during the period 4 July-3 Sept.
***Column C gives the number of hours and the percentage of hours in Column B
which coincide with a high pressure center or ridge within 450 miles of
the station.
****Column D gives the number of hours and the percentage of hours in Column B
which coincide the period the station remains in the same air mass regard-
less the distance of the high center from the station.
132
-------
Though these data support the contention that high ozone is found in
the presence of a high pressure system, they also suggest that the converse
is not necessarily true; that is, the presence of a high pressure system does
not necessarily imply the existence of high ozone. It is noted that in 1973
and 1974, high ozone occurred for about 50 percent of the total number of
hours of high pressure. In 1975, it only occurred about 12 percent of those
hours. If the ratio of hours of high pressure to that of high ozone is com-
puted for daytime periods only, they are 65 percent for 1973, 50 percent for
1974, and 18 percent for 1975.
Figure 54 shows the results of computing a nine-point running mean
using the data in figures 52 and 53. For 1975, the Bradford, Pennsylvania,
data were used so that the figure shows variations in the eastern portion
of the United States only. The nine-point running mean filters synoptic
scale and mesoscale fluctuations, leaving only long-term trends (or macro-
scale fluctuations with periods greater than 9 days). In 1973 and 1974,
the summer months were characterized by macroscale high pressure systems,
which last for approximately 30 days or more. However, in 1975, they were
characterized by a series of macroscale pressure systems, the longest of
which was approximately 23 days and was centered around the 30th of July 1975,
Coinciding with the pressure systems in 1973 and 1974 were macroscale ozone
concentrations in excess of the NAAQS, which also lasted for approximately
30 days. However, in 1975, there was only one period, lasting approximately
4 days, in which the macroscale ozone exceeded the NAAQS centered at the
30th of July. These data suggest that ozone concentrations in excess of
the NAAQS are associated with macroscale high pressure systems lasting longer
than 20 days. If such systems do not develop, the occurrence of ozone con-
centrations greater than the NAAQS in rural regions appears to be minimized.
This is supported by the facts that, though the number of hours a high pres-
sure system or a ridge was near Bradford and Dubois in 1975 and 1974 were
about equal, Bradford (1975) had considerably fewer number of hours of high
ozone than DuBois (1974) (table 28).
7.2.3 Distribution of Ozone Relative to a Moving High Pressure System
Figure 55 shows the average daily variation of the diurnal ozone maxi-
mum relative to a moving high pressure system using data from Kane, Pennsyl-
vania, in 1973, from DuBois, Pennsylvania, in 1974 and from Bradford,
133
-------
1974
A
Days
Figure 54. Nine-point running average of the data presented in
figure 53 and of the Bradford ozone data and the
Pittsburgh pressure data presented in figure 53.
134
-------
4320 3600
250.
2380 2160 1440
Distance (Ton)
Days
4 3 2
130.
30-
COLD
FRONT
HIGH PRESSURE
SYSTEM CENTER
COLD
FRONT
488040003200240016003000
Distance (km)
Days
250
200.
ISO.
100.
SO.
COLD
FRONT
1975-BRADFORD
Hlffl PRESSURE
SYSTEM CfcNIkK
3750 3125 2500 1875 1250 625 0
Distance (Ion)
Figure 55. The temporal and spatial variation of the diurnal
maximum ozone concentration through a high pressure
system located in the east in the summers of 1973,
1974, and 1975.
135
-------
Pennsylvania, in 1975. The period represented in the data coincides with
the period from an initial frontal passage through a high pressure system
to the next frontal passage. Only the data around 1700 EDT for each day
represent an actual data point. A smooth curve was passed through these
points in order to represent the average variation of the diurnal ozone
maximum with distances from the initial frontal passage. Distances were
computed based on the mean speed of the high pressure system. It is noted
that the distances span the entire continent of the United States. These
distances are representative of synoptic wave numbers three and four, which
characterize the particular summer situations in each case. The mean period
from frontal passage to frontal passage through the high pressure system was
exactly 6 days in each of the 3 years.
The data indicate that a relative minimum in the diurnal ozone maximum
occurs somewhere in the region between the initial frontal passage and the
high pressure center. For a slow-moving high pressure system (the Bradford
data), the relative minimum occurs soon after frontal passage (less than a
day) or approximately 400 km upstream of the front. However, for fast-
moving high pressure systems such as that found in 1974 (the DuBois data),
the relative minimum occurs about 2 days after frontal passage or approxi-
mately 1,400 km upstream of the front. The 1973 data indicate that the
position of the relative minimum (approximately 1.5 days after frontal
passage or 1,000 km upstream) falls between that found in 1975 and that
in 1974. The mean speed of the high pressure system in 1973 was greater
than that found in 1975, but: less than that found in 1974.
The largest value for the diurnal ozone maximum occurs after the high
pressure center passes the station or on the back side of the high pressure
system. In each of the 3 years, the concentration of ozone on the back
side of the high pressure system exceeded the NAAQS.
These data support some of the conclusions resulting from the 1974 RTI
summer field program,— but are more significant in the statistical sense.
However, one result of this study differs from that found in the 1974 summer
study program. Though the 1974 data indicate that high ozone is found near
the center of the high pressure system, the highest ozone is found on the
back side of the high (or at least 2 days after the high pressure system
passed the station). Distances given in the figures cannot be considered
136
-------
realistic because variations in the speed of the high pressure system could
produce both larger and smaller distances. Differences between the results
of this study and the 1974 study can be explained from the facts that the
1975 study dealt with time-dependent variations in ozone whereas the 1974
study dealt with space dependencies with no regard to time at all.
Figure 56 is the average variation of the diurnal ozone maximum rela-
tive to a moving pressure system determined using the 3 years of data in
figure 55. These overall data summarize that found in figure 55.
Figure 57 is a similar representation to that of figure 56, except
that the 1975 data for Bradford, Pennsylvania; Creston, Iowa; and Wolf
250-
200-
>r 15°-
s
=0
I 100 -j
50-
COLD
FROOT
EAST -
o=7.5 ms"1
HIGH PRESSURE
SYSTEM CENTER
COLD
FRONT
1
6
4200
i
5'
3500
i i
4 3
Days
2800 2100
Distance (Ton)
i
2
1400
i
1
700
i
0
1
0
Figure 56. The average temporal and spatial variation of the
diurnal maximum ozone concentration through a high
pressure system located in the east based on the
1973, 1974,and 1975 data.
137
-------
4800
200 —
,= ISO-I
100 —
50'
4000
I
Distance (Ion)
3200 2400 1600
800
BRADFORD
CRESTCN
WOU: POINT
Pressure
Center
6 S 4
Bradford § Wolf Point
1 1
4 3
Creston
1
2
1
1
Days
1
0
Figure 57. The temporal and spatial variations of the diurnal maximum ozone
concentration through a high pressure system based on the 1975
data at Wolf Point, Montana; Creston, Iowa; and .Bradford,
Pennsylvania.
138
-------
Point, Montana, are given. The period from frontal passage through the
high pressure system to the next frontal passage was 6 days at Bradford and
Wolf Point, but was approximately A days at Creston. At Creston it took,
on the average, about 1 day for the high pressure center to reach the sta-
tion after frontal passage, but the next front did not arrive until 3 days
after the high pressure center arrived. Apparently, there was an initial
acceleration of the high pressure system in the Great Plains regions of the
United States. Distances were computed based on the mean speed of the high
pressure system at Bradford.
The data indicate that there is no marked variation in the ozone con-
centration at Wolf Point as the high pressure system passes through that
station. The ozone concentration was generally higher on the back side of
the high, but only by approximately 10 percent of that in the front side.
The variation of ozone concentration at Creston is similar to that shown in
figures 55 and 56, except in this case the relative minimum was found when
the high pressure center had passed the station and the ozone concentration
on the back side of the high did not exceed the NAAQS. It was not until
the high pressure system moved into the eastern portions of the United States
that the ozone concentration on the back side of the high pressure system
exceeded the NAAQS. This suggests an enhancement of a mechanism to increase
the ozone concentration in local air parcels as the high pressure system
moves eastward from Montana to Pennsylvania.
7.2.4 Source Regions and Residence Times of Air Relative to a Moving
High Pressure System
A potential source for ozone is the upper atmosphere. However, analy-
sis of available meteorological data did not show any evidence for an en-
hancement of downward transport of ozone, on the average, as a high pressure
system moved from the west to the east by either subsidence or by vertical
mixing. Low-level divergence and downward vertical motions are usually
greatest near the center or in the forward portions of an eastward moving
high pressure system, not on the back side. It could be hypothesized that
the vertical gradient of ozone is greater through the troposphere in the
East than in the West, which could be responsible for enhanced downward
transport of ozone, but available vertical profile data from aircraft
measurements did not support this.
139
-------
Since vertical transport of ozone does not appear to be the mechanism
to produce high ozone and since the number of industrial complexes and popu-
lation centers (sources for ozone precursors) increases markedly from the
west to the east, enhancement of tropospheric synthesis is suggested as the
most probable mechanism. Since the industrial complexes and population
centers are large in number aad widespread in the East, injection of ozone
precursors into the front side of a high pressure system should be identi-
cal to that into the back side. The fact that the ozone concentrations are
larger in the back side of a moving high pressure system suggests a build-up
of ozone precursors or the establishment of a critical distribution of ozone
precursors takes place as parcels of air circulate in the high pressure
system. It is important for air parcels to remain within the high pressure
system since the system offers the critical environmental factor necessary
for synthesis; that is, relatively cloudless skies which allow unimpeded
exposure to solar radiation.— The longer parcels of air remain in the
high pressure system and travel through the industrialized and highly popu-
lated eastern portions of the United States, the greater the potential is
to increase significantly the concentration of ozone precursors and to
produce high ozone. The following calculations offer some insight into the
residence time of air parcels, in high pressure systems.
It was assumed in the initial calculations that the pressure distribu-
tion, p, in the high pressure system may be expressed by the following
equation:
p = p + p sin k(x - c t) sin A.y, (1)
O _L X
where
k is the wave number in the x direction (west-east direction) ,
k = 2 /L and L is the wavelength along x (see figure 58)
X X
c is the wave speed assumed to have an x component only,
X
is the wave number in the y direction (south-north direction) ,
= 2 /L and L is the wavelength along y (see figure 58)
p is the mean of pressure,
p is the amplitude of pressure for the harmonic, and p is pressure.
140
-------
=4R
Figure 58. Hypothetical high pressure system for
which residence times were calculated.
141
-------
Accelerations were assumed to be zero, which yield the following equations
of motion in the high pressure system
fv - fv + Ku = 0
S
fu - fu - KV = 0 ; (2)
&
where
f is the Coriolis parameter and is allowed to vary in the y direction
v is the south-north component of the wind (y-component),
u is the west-east component of the wind (x-component),
K is the friction coefficient,
u is the west-east component of the geostrophic wind, and
&
a 3p /0,
ug = - ? at • (3)
v is the south-north component of the geostrophic wind, and
o
a 8p , /7s
vg = I 3y ' and (4)
a is the specific volume.
This system of equations was used to compute the boundary layer trajec-
tory of air parcels in a circular symmetric high pressure system which was
used to determine the residence time for air parcels. All motion was con-
strained to be horizontal. The calculations were subject to the following
conditions. The high pressure: system had a radius of approximately 1,000 km
and at that outer boundary, the geostrophic wind speed was set equal to 10 m/s,
-4 -1
The value of the friction coefficient was 10 s . These are typical param-
eters associated with high pressure systems.
Figure 59 shows the number of days air parcels that start at various
locations in a high pressure system will spend within that system. Varia-
tions in the three high pressure systems are a result of differences in the
speed of the systems. The system speeds were 5 m/s, 7.5 m/s and 10 m/s, and
the direction of motion was to the east. Calculations indicate that, regard-
less of system speed, the source region for air parcels which will spend more
142
-------
-1
c « 10 ms
-------
than 1 day in the high pressure system is the north-eastern quadrant of the
system. The number of days an air parcel in that quadrant will spend in the
high pressure system depends on the system speed. The slower the speed, the
greater the number of days. Air in the back side of the high pressure sys-
tem, regardless of system speed, will spend less than 1 day in the system.
Figure 60 shows the number of days air parcels in the high pressure
system already have spent within the system. The system speeds are the
same as before. On the whole, the air in the back side of the high pressure
system has spent more days in the high pressure system than the air in the
front side. The region of maximum residence time shifts from the south-
western sector for small speeds to the northwestern sector for large system
speeds. As the system speeds increase, the contour depicting one-day resi-
dence time shifts westward. This suggests the hypothesis that residence
times greater than one day are required for air parcels in a high pressure
system to build up critical concentrations of ozone precursors. Subsequent-
ly, the rise in ozone to the concentration experienced on the back side of
the high pressure system should commence nearer to the leading edge of a
high pressure system (soon after frontal passage) for a slow-moving system,
and near or at the center of the high pressure system for faster-moving
systems. This agrees well with the data in figure 55 which shows that the
minimum concentration (afterwards there is a rise in the ozone concentra-
tion) is found soon after the front passed the station for the slow-moving
system and nearer to the center of the high pressure system for the faster-
moving system. The data in figure 57 shows that at Creston the ozone con-
centrations did not begin to rise until after the high pressure center
passed the station. Whereas, at Bradford, the rise in concentration occurred
soon after the initial frontal system passed the station. The difference is
probably a result of the pressure system moving through Creston at twice the
speed it moved through Bradford.
The distribution of residence times also is a function of the shape of
the high pressure system. Figures 61 and 62 show similar distributions as
those shown in figures 59 and 60 but for an elliptically shaped high pres-
sure system. In the first case, the major axis (2,600 km) was oriented
north-south, and the system was moving from west to east at a speed of
7.5 m/s. In the second case, the major axis was oriented west to east, and
144
-------
= 5ms
-1
c = 10 ms
.X
-1
Figure 60. The number of days air parcels in high pressure systems
have spent (residence time) within the system versus
the speed of the system for a circular symmetric high
pressure system.
145
-------
Figure 61. The number of days air parcels in various locations in
a high pressure system will spend within that system
versus differently shaped elliptical high pressure systems.
The system speed is 7.5 ms~^-.
146
-------
Figure 62. The number of days air parcels have spent (residence
time) within a high pressure system versus differently
shaped elliptical systems. The system speed is
7.5 ms"1.
147
-------
the system also was moving eastward at a speed of 7.5 m/s. The minor axis
was 1,600 km. The source regions for air that will have large residence
times remain in the northeastern quadrant, regardless of the orientation of
the major axis. The system with the major axis oriented west to east has
the air in the northwestern quadrant having the largest residence time
(figure 62) identical to that found using the circular symmetric high pressure
system; but systems with the major axis oriented south to north have the largest
residence time in the northern sector with a secondary maximum in the south-
western quadrant. This portrays differences which can be encountered in the
distribution of residence times within the high pressure system as the shape
of the system changes and suggests that if residence times of air parcels
are important in the production of high ozone as previously implied, there
may be marked differences in the distribution of high ozone within the high
pressure systems as the shape of the system changes.
7.2.5 Some Aspects of the Mechanisms Governing the Chemistry of Ozone
in High Pressure Systems
In order to examine the cheirical behavior of ozone in the high pressure
system, it was necessary to examine theoretically the governing equation for
the mean diurnal behavior of ozone in the lower portion of the boundary
layer (e.g., first 100 meters). The theoretical effort is described below.
7.2.5.1 Theory
A chemical and environmental system is considered in which synthesis,
transport, and destruction are the principal mechanisms governing the be-
havior of ozone. Mathematically, this can be described by the following
equation:
8°
G + M - D , (5)
o t
where
30»/3t is the local rate of change of ozone,
G is a contribution to the local rate of change due to synthesis
of ozone,
M is the contribution 1:o the local rate of change due to transport
of ozone , and
D is the contribution to the local rate of change due to destruc-
tion of ozone either in the gas phase or at the surface.
148
-------
The mean diurnal variation of ozone is usually determined by averaging
of the hourly value over a very large number of days. In the summation of
daily data, most synoptic weather regimes should be representative. In the
average, therefore, the contribution due to horizontal transport is minimized
since it is doubtful that horizontal gradients of ozone would exist over
continental regions such that mean positive or negative contributions would
o/• /
exist. Furthermore, Bach— has shown that there is no unique correlation
between air trajectories and high or low ozone over the eastern half of the
United States. It should be pointed out that earlier in this report, it was
shown that ozone gradients may exist between oceanic regions and continental
regions, due apparently to a lack of ozone precursors in the oceanic regions.
However, in this case we are dealing only with continental regions sufficiently
inland so the influence of the oceanic regions would not affect the distribu-
tion of ozone. Since on the average, ozone does increase upward through the
27 /
boundary into the mid and upper troposphere,— positive contributions
through vertical transport could exist.
If a point in space is considered not at the surface, then gas phase
destruction is a process which dominates the destruction term. The predomi-
nant reactions are second-order; that is, the reaction of ozone with one
other gas. This can be expressed mathematically by the following expression:
D = E k ¥ 00 (6)
n n j
n
where
k is the rate constant,
n
¥ is the concentration of destructive agents, and
n
n is the index of summations since there is more than one destructive
agent for ozone.
It is assumed that the concentration of destructive agents is independent
of time. This assumption severely limits the quantitative results, but it
should allow reasonable qualitative results and also greatly simplifies the
mathematics. Based on this assumption, a pseudo-first-order rate constant
can be defined such that
149
-------
K = I k Y , (7)
o n n
n
where
K is the first-order rate constant.
o
Therefore, the resultant expression for the destruction term is
D = KQ 03 . (8)
Figure 63 shows the average daytime variability of the magnitude of
the synthesis and of the (vertical) transport term in equation of continuity
28/
for ozone determined from data collected by Jeffries— in a rural North
Carolina boundary layer near the surface. Jeffries data have been smoothed
and normalized to produce the curves in figure 63. At night (2100 to 0800
EST), the synthesis term is 0 and the magnitude of the transport term was
relatively small. The nighttime value of the transport term was subtracted
from the data to produce a system governed by destruction at night in order
to simplify the mathematics. Generally, the nighttime behavior of ozone is
29/
governed by destruction, although some vertical transport exists.—
Both the transport and synthesis terms begin to increase after 0800 EST,
reach a maximum at 1500 EST, and return again to zero at 2100 EST. Figure
63 also contains a truncated Fourier series made up of the mean value and
one harmonic. The Fourier series preserves the initial time and the time
of maximum for both the transport and synthesis terms, but the evening mini-
mum occurs 1 hour later (2200 EST). The truncated Fourier series represents
the variations observed by Jeffries reasonably well and will be used as a
mathematical representation of the daytime contribution for both the trans-
port and the synthesis term to the diurnal variation of ozone.
Therefore, the continuity equation, which governs the daytime behavior
of ozone and which is valid from t=t tot = t +T, is
o oo
80
3-^ + K 0, = A {1 - cos [co (t - t )]} . (9)
9t o 3 o
The equation governing nighttime behavior of ozone, which is valid from
150
-------
t = t + T to t , is
o o
90
9F + Ko°3 = ° ;
where A = a + b, a is the amplitude of the transport term, b is the ampli-
tude of the synthesis term, to = 2TT/T, and is the frequency for synthesis
and/or transport, T is the period for both the transport and the synthesis
term, and t is the initial time for both the synthesis and transport term
and in the applied sense, can be considered the time of minimum ozone. The
righthand side of equation 10 is zero because at night the synthesis term
is zero and the vertical transport term is negligible. Under conditions
a>b,b>a, ora=b, the qualitative behavior of equation 9 will be
unchanged, provided the period is the same for both the synthesis and the
transport terms.
The solutions of the equations 9 and 10 were found subject to the fol-
lowing boundary conditions:
0 = 0 at t = t for equation 9,
00 - 0OT, at t = t + t for equation 10, (11)
J J.h O
where 0_ is the value of 0_ at t + T and is obtained from the solution
of equation 9. These boundary conditions deal with variations observed in
the lower boundary layer only.
The solution to equation 9 under the boundary condition is:
03 = ^ {1 - AQ cos [u)(t - to) - 6]} . (12)
? -1 /?
X = u/K , X = (X + 1) ' , and 0 = arctan (X).
o o
The above solution is valid under the condition that t > t (for example,
t = time of maximum synthesis and/or transport). According to equation 12,
ozone is a maximum when the argument of the cosine is pi (TT) . The time of
maximum ozone (t ) is defined by the following formula:
IQ3.X
151
-------
(,j(t - t ) - 6 = IT
03 max o
or (13)
t = t' +6
max max
where
t' - t + 0.5 T which is the time of maximum transport and/or
max o
synthesis,
6 = 6/0) which is the difference between the time of maximum ozone
and the time of maximum transport and/or synthesis.
Figure 64 shows the variations of 6 with X for periods T = 10 hours,
T = 12 hours, and T - 14 hours. Variations in the period are due primarily
to durations of daylight and represent seasonal influences. The figure
shows that when X = 0, 6=0. Since T is finite and greater than zero, u) is
finite and greater than zero. Therefore, if X = 0 (X = 0)/K ), K must be
infinite which requires an infinite concentration of destructive agents
since the rate constants are finite. As X approaches infinity, 5 approaches
a maximum value which depends ort the period. K must approach zero or the
concentration of destructive agents approaches zero in order for X to
approach infinity. These results indicate that if large amounts of ozone-
destructive agents are present, the time of maximum ozone approaches the
time of maximum synthesis and/or transport. However, if the concentration
of ozone destructive agents is small, there will be a lag between the time
of maximum ozone and the time of maximum synthesis and/or transport with the
time of maximum ozone occurring later in the day.
Figure 65 gives theoretically derived concentrations of NO, N0_, and
a-pinene as a function of 6 for various values of T, using data in figure 64.
These concentrations were computed under the assumption that the particular
gas was the only available ozone-destructive agent. Though the specific
values of the concentrations computed are unreliable due to the simplifica-
tions made, it is believed that the order of magnitude may be reasonable.
This point will be clarified later. The figures show that NO must be of
152
-------
2.0 -
c
o
•H
ec
c
9.
— Transport [Jeffries, 1971]
Sy.ithtciis [Jsffrici, 1971]
l-ccs[(u(c-t0)]
-]'ii•ii•ii•i
08 10 12 14 16 18 20 22
Time (EST)
Figure 63. The diurnal variation of the transport and synthesis
term in a rural boundary layer in North Carolina.
4.o_
3.0-
2.0-
i.o-
T-I
0 1
I I I 11 I I I I I I I I
5 10 50
100
> (non-dlnMnaional)
Figure 64. The variation of 6 versus X.
153
-------
l.O—i
I
sr
i o.oi-
0.001.
60—,
I I
Figure 65 . The variation of the concentrations of NO, NO-, and
ct-pinene versus 6 for various values of T.
154
-------
the order of 0.07 to 0.006 pphm to produce 6's of the order of 1 hour to 3
hours, respectively; NO must be of the order of 17 pphm to 2 pphm; and
a-pinene must be of the order of 14 pphm to 1 pphm. The data indicate that
extremely small and normally undetectable changes in the NO concentration
can produce large differences in 6. However, large and detectable changes
in the concentration of both a-pinene and N0« must occur in order to pro-
duce similar changes in 6.
According to equation 12, the maximum concentration of ozone at t
max
(that is, when the argument of the cosine is equal to TT) is given by the
following formula
0 - — [1 + X ]
3max to o
(14)
Consider two regimes, which are defined by the following set of parameters.
REGIME #1
0 * - 0. in #1
3 3max
t* = t in #1
max
A* = amplitude of synthesis
and/or transport in #1
6* = 6 in #1.
REGIME #2
0 ** = 0. in #2
3 3max
t** = t in #2
max
A** = amplitude of synthesis
and/or transport in #2
6** = 6 in #2.
(15)
If 0) was identical in each regime, then the ratio of the maximum concentra-
tions is given by
(16)
where
a
A*
A** '
(17)
and
3 =
(i + X *)
o
(18)
155
-------
Che parameter, a, treats the differences in synthesis and/or transport in
the two regimes, and B, the differences in the destruction in the two
regimes.
Figure 66 yields the variation of 3 as a function of the differences
in & in the two regimes. The data generally show that for large differences
in S, with the & in regime #1 greater than the 8 in regime #2, 3 is much
greater than 1. For example, Lf 6* = 3 hours and 6** = 1 hour, g = 6.7.
This suggests that in the comparison of two regimes, one having smaller
concentrations of ozone destructive agents than the other, the regime having
the small concentration of ozone destructive agents has the potential to
allow the achievement of a larger maximum ozone concentration. Whether or
aot the regime with the small concentration of ozone-destructive agents
actually produces a larger maximum ozone concentration will depend on the
parameter, a.
The solution to equation 10, the governing equation for the nighttime
behavior of ozone, is
[-Ko(t- to>] . (19)
The expression for Oor, is determined from equation 12 and is the value of
.3E
equation 12 at t = t + T; that is,
0 = — - (20)
3E 0) X2 + i
The value of the ozone concentration at the end of the diurnal cycle (0,, )
was calculated using equation 19. The results are expressed in terms of
the ratio 0_ /00 ; that is,
3o 3max
3max
- V"
where At is the time interval over which equation 10 is valid.
Figure 67 yields the variations of the ratio 0» /00 as a function
° J 3o 3max
of 6 and T. The data indicate that for small values of 6, or large concen-
trations of destructive agents, the concentration of ozone at the end of
156
-------
30—i
60 —
3 40 —
20—'
5 « 3 hrs
4* « 2 hrs
* » 1 hr
1 2 3
4**(hrs)
* **
Figure 66. The variation of 6 versus & and &
l.O-i
.8-
x .6-
I
.4-
.2-
1.0 2.0 3.0 4.0
Figure 67. The variation of the ratio, 0
-
versus 6,
157
-------
the diurnal cycle is zero. However, for large values 6, or small concen-
trations of destructive agents, a residual ozone concentration will remain
at the end of the diurnal cycle. It appears that 6=2 hours is the crit-
ical value. It should be noted that the actual amount of the residual will
depend on the value of ()„
3max
Due to the assumptions required to linearize governing equations for
the diurnal behavior of ozone, the resulting model is an oversimplification
of ozone behavior. However, though the quantitative results stemming from
this modeling effort are unreliable, it is believed that the qualitative
results are reasonable. The results of the model effort will now be ap-
plied to examine the chemistry of ozone in high pressure systems.
7.2.5.2 Variation of the Diurnal Cycle of Ozone in a Moving High
Pressure System
The synoptic surface weather data for the period 4 July through 30
September 1975 were examined to determine when a high pressure system moved
out of Canada and passed consecutively over Wolf Point, Montana; Creston,
Iowa; and Bradford, Pennsylvania. During the period, eight such systems
were found. At each of these locations, the diurnal ozone distributions
were examined during the period the high pressure system was in the vicinity
of the station, and the day having the largest diurnal maximum ozone concen-
tration for each system was used to compute the average diurnal variation
over all eight systems. Figure 68 shows the results of the computations.
As the high pressure system moved through Wolf Point and Creston to Brad-
ford, the ozone concentration increased. The percent of increase was
greater between Creston and Bradford than between Wolf Point and Creston.
It is noted again that only at Bradford was the ozone concentration about
equal to the NAAQS.
Table 29 summarizes some of the significant data in figure 68. The
t at Wolf Point and Creston are identical, but the t at Bradford is
max max
3 hours earlier in time. Assuming that the time of maximum transport and/or
synthesis are identical and that this time is near the time of maximum solar
radiation (t? = 1400 LOT), the data indicate that there were larger con-
max
centrations of ozone destructive agents at Bradford (6 = 1/2 hr) than at
either Creston or Wolf Point (6 = 3 1/2 hrs.).
158
-------
ISO —
125
100 —
BRADFORD
— - CRESTCN
— WOLF POINT
7S —
30 —
2S_
TIT n -I i M i i M ii M M
24 6 8 10 12 14 16 18 20 22 24
Chrs)
Figure 68 .
The average diurnal variation of ozone concentration
at Wolf Point, Creston,and Bradford based on eight
high pressure systems which consecutively passed
through these stations. The day with the largest
diurnal maximum ozone concentration, when the high
pressure system was in the vicinity of the station,
was used to compute the average for all eight systems.
159
-------
Table 29. The values of tmax, 03*/03** (where 03* is Wolf Point ozone
amplitude in all cases), 3 (estimated from tmax and letting
1400 LDT), a, diurnal ozone amplitude (diurnal maximum
Lmax
ozone minus diurnal minimum ozone), and the diurnal minimum
ozone concentration for Wolf Point, Montana, Creston, Iowa, and
Bradford, Pennsylvania obtained from the diurnal curves given
in figure 68
STATIONS
Wolf Point
Creston
Bradford
t
max
1730 MDT
1730 CDT
1430 EDT
* A*
°3 /03
1.0
1.1
0.85
3
1.0
1.0
20.0
a
1.0
1.1
0.04
Ozone
Amplitude
(jjg/m3)
71
65
83
Diurnal
Minimum
(yg/m3)
24
45
7?
Table 29 also gives the values of the ratio 0 */0 ** where Wolf Point
was considered to be Regime #1 in each case, estimates of 3 based on values
of 6 (computed using the values of t and assuming a value of 1400 LDT
in 3.x
for t1 ) and on data in figure 66, and the computed values of a. The
ratio 0 */0 ** was computed using the ozone amplitude (maximum minus mini-
mum) in this case and in all following cases because the theory requires
that the initial value of ozone is zero. This requirement can be satisfied
by subtraction of the minimum ozone concentration. The calculations indi-
cate that the amplitude of synthesis and/or transport at Creston and Wolf
Point are identical; however, for Bradford the amplitude is considerably
larger. Since available meteorological evidence does not support a larger
vertical transport when the high pressure system was in the eastern portion
of the United States than in the western portion, the ratio a must treat
only amplitudes of tropospheric synthesis.
These data indicate that, as the high pressure system moves from the
western portions of the United States to the eastern portions, there is a
marked change in the local O2;one chemistry. Both destruction and synthesis
of ozone are larger when the high pressure system is in the eastern portions
of the United States. This supports the notion that there is widespread
160
-------
injection of ozone precursors, and apparently ozone-destructive agents also,
in the east.
Table 29 also presents the amplitude of ozone (maximum ozone concentra-
tion minus minimum ozone concentration) and the minimum ozone concentration
for each of the three stations. These data indicate that though there is a
larger synthesis of ozone at Bradford, the amplitude of ozone is not much
different than at Creston and Wolf Point. This is due to the increased
(decreased) ozone destruction at Bradford (Wolf Point and Creston). The
overall increase in ozone appears to be due to a buildup of the minimum
ozone concentration. Since theory indicates that a residual ozone concen-
tration will be found at the end of a diurnal cycle when 6 is of the order
of two hours or greater, it is suggested that the small concentrations of
ozone-destructive agents in air parcels in the west allowed a buildup of
the minimum ozone concentrations as they drifted westward. (Nota Bene:
an air parcel does not travel with a high pressure system across the United
States. It would take the circulation associated with six simultaneous high
pressure systems moving from Wolf Point through Creston and Bradford at a
_i
speed of 10 ms to transport a parcel from Wolf Point to Bradford.) In
the east, injection of ozone destructive agents reduced 6 to less than 2
hours, and no further build-up of the minimum concentration was allowed.
7.2.5.3 Variation of Ozone Chemistry in a High Pressure System in the
Eastern Portion of the United States
Figure 69A gives the average diurnal variation of ozone for days when
the maximum ozone concentration exceeded the NAAQS and for the remaining
days when it was less than the NAAQS at Kane, Pennsylvania, during August
1973. Table 30 summarizes pertinent data from figure 69A. The t for
fe max
Table 30. The values of tmax, 03*/03** (03* is the amplitude of ozone
for the high ozone case), 3 (using tmax and letting t^x =
1400 LOT), and a for Kane, Pennsylvania obtained from the
1973 diurnal curves given in figure 69
CASE
High Ozone
Low Ozone
max
1630 EDT
1430 EDT
* , **
o3/o3
1.0
1.S9
s
1,0
9.6
a
1.0
0.19
161
-------
II I I I I I I I I I I I I I I I I 1.1 I I i 1
00 02 04 06 OS 10 12 14 16 IS 20 22
Tin (hn)
200 —
ISO—
160 —
140 —
120 —
100 —
30 —
60—
40—
1)1*015--1974
II I I II I I I I I I I I I I I I I I I I I
00 02 04 04 0< 10 12 14 16 18 20 22
Tlat (hn)
130
160 —
140-
„- IZ°-
J 100—
f $0-
60 —
10—
,•
/ ^-^\
Sft««O"l9TS / >t
^^r^
^•^/
CC)
II i m iriTTirriTTiirr
00 02 04 36 OS 10 12 14 16 13 :0 U
Tin* Oiri)
Figure 69. The average diurnal variation of ozone for those days
when the diurnal maximum ozone concentration exceeded
the NAAQS (solid line) and when the diurnal maximum was
less than the NAAQS (dash-dot line) based on the data
for August 1973(A), 1974(B), and 1975(C) and at Kane,
DuBois, and Bradford, Pennsylvania, respectively
162
-------
the high ozone curve (greater than the NAAQS) occurs later In the day than
that for the low ozone curve. Assuming that t' = 1400 LDT, the data indi-
max
cate that there were more ozone destructive agents on the days of low ozone
(6 = 1/2 hr) than the days of high ozone (6 «• 2 1/2 hrs). Also shown in the
table is the ratio 0 */0 ** where the high ozone curve was Regime #1, esti-
mates of 3, and the computed value of a. These data indicate that there was
considerably more synthesis of ozone (vertical transport being ruled out) on
days of low ozone than days of high ozone. The small concentrations of
ozone destructive agents on days of high ozone versus days of low ozone is
further supported by the fact that the minimum value on high ozone days is
considerably greater than the minimum value on low ozone days (see figure
69A).
Figure 69B yields data similar to figure 69A except these data were
obtained at DuBois, Pennsylvania in August 1974. Table 31 summarizes the
pertinent data in the figure. The t for the low ozone curve occurred
v 6 max
later in the day than for the high ozone curve which indicates more ozone
destructive agents on days of high ozone than on days of low ozone. This
is further supported by the fact that the minimum concentration for the
high ozone curve is less than the minimum concentration for the low ozone
curve (see figure 69B). Table 31 also gives the values of the ratio 0,*/0 **,
estimates of 6, and the computed value of a. Results indicate that the
synthesis of ozone was greater on days in which the ozone was high than
days when the ozone was low. This is a complete reversal of the role of
destruction and synthesis found in 1973.
Table 31. The values of t^^, 03*/03** (03* is the amplitude of ozone
for the high ozone case), 3 (using t^x and letting t^^ =
1400 LDT), and a for DuBois, Pennsylvania obtained from the
1974 diurnal curves in figure 69
CASE
High Ozone
Low Ozone
max
1430 EDT
1530 EDT
* **
o3/o3
1.0
3.36
3
1.0
0.33
a
1.0
10.3
163
-------
Figure 69C yields data identical to figure 69A, except in this case
the data were collected at Bradford, Pennsylvania, in August 1975. Table
32 summarizes the pertinent data in the figure. As in the case of the
1974 data, the low ozone curve has a t that occurs later in the day
max J
than the high ozone curve, which indicates that the concentration of ozone
destructive agents was greater on days of high ozone than on days of low
ozone. That there were larg&r concentrations of ozone destructive agents
on days having high ozone is supported by the fact that the minimum value
on days of high ozone is smaller (see figure 69C). Also shown in table 32
is the value of the ratio 0 */0-**, the estimated value of $, and the com-
puted values of a. In this case as in the case of 1974, the synthesis of
ozone was greater on days having high ozone concentration than on days
having low ozone concentration. These data, too, show a complete reversal
of the role of destruction and synthesis found in 1973.
Table 33 summarizes pertinent data from figures 69A, B, and C using
the high ozone curve only. The t for 1973 occurs later in the day than
6 J max J
that for 1974 and 1975, indicating larger concentration of ozone destructive
agents in 1974 and 1975. The calculated values of a indicate that synthesis
of ozone was greater in 1974 and 1975 than in 1973. The amplitude of ozone
supports the contention that there was greater synthesis of ozone in 1974
and 1975 than in 1973. The data also show that the high ozone in 1973 is
basically a result of having a large minimum concentration of ozone. The
large minimum concentration of ozone is due to the lack of a large concen-
tration of ozone destructive agents, and suggests that in 1973 minimum value
Table 32. The values of tEiax» 03*703** (03* is the amplitude of ozone
for the high ozone case) , g (using tmax and letting t^x =
1400 LDT), and a for Bradford, Pennsylvania obtained from the
1975 diurnal curves in figure 69
CASE
High Ozone
Low Ozone
t
max
1430 EDT
1630 EDT
* **
0_ /O,
3 3
1.0
2.64
B
1.0
0.1
a
1.0
26.4
164
-------
Table 33. The values of tmax, 03*/03** (when 03* was the amplitude
of ozone in 1973 in all cases), 0 (using tmax and letting
'max = 1400 LOT), a, the diurnal amplitude of ozone (ozone
maximum minus the ozone minimum), and the diurnal minimum
ozone concentration for the high ozone cases in 1973, 1974,
and 1975 from figures 69A, B, and C
YEAR
1973
1974
1975
max
1630 EDT
1430 EDT
1430 EDT
„ */„ **
°3 /03
1.0
0.55
0.77
6
1.0
9.6
9.6
a
1.0
0.04
0.08
Ozone
Amplitude
(yg/m3)
92
168
119
Minimum
Ozone
(yg/m3)
123
28
46
continued to increase in the eastern portions of the United States. Though
the 1975 data (figure 69) indicated that there is a buildup of the diurnal
minimum ozone concentration in the West, the buildup ceased in the East where
air parcels in the system had acquired larger concentrations of ozone destruc-
tive agents. The smaller concentrations of ozone destructive agents, the
smaller value of the synthesis term for ozone, and the larger number of
hours of high pressure in the eastern portions of the United States suggests
that the ozone system in 1973 may have been a spent system; that is, the
high pressure systems in 1973 were relatively stagnant compared to those in
1974 and 1975, and the ozone precursors and ozone destructive agents were
depleted as a result of the large number of diurnal processes experienced.
Figures 70A and 70B show the average diurnal variation of ozone on days
of high ozone and on days of low ozone at Lewisburg, West Virginia, in
August 1973 and 1975, respectively (1974 data are not available for this
station). Tables 34, 35, and 36 summarize the pertinent data from these
figures. Though specific factors on the year-to-year variation of high and
low ozone do not coincide exactly with those found at the Pennsylvania sta-
tions, these data do show that there were greater synthesis and larger con-
centrations of ozone destructive agents on high ozone days in 1975 than in
1973, and that the minimum ozone concentration in 1973 was almost three
165
-------
220—i
200.
130—
160—
~* 140—
***
"^ 120—
o^ 100—
30—
60—
40—
LEWISBURG--1973
(A)
\
\
\
tin rm nTin rn i r i i i i t
00 02 04 06 03 10 12 14 16 18 20 22
Time (hrs)
(B)
i i i iTTTTTTM T in i NTT IIT
00 02 04 06 08 10 12 14 16 13 20 22
Time (hrs)
Figure 70. The average diurnal variation of ozone at Lewisburg,
West Virginia for thpse days when the diurnal maximum
ozone concentration exceeded the NAAQS (solid line)
and when the diurnal maximum was less than the NAAQS
(dash-dot line) based on data for August 1973(A) and
1975(B).
166
-------
Table 34. The values of tmax, 03*703** (where 03* is the amplitude of
ozone in the high ozone case), 3 (using tmax and letting
tmax = !400 LDT), and a for Lewisburg, West Virginia
obtained from the 1973 diurnal curves given in figure 70
CASE
High Ozone
Low Ozone
t
max
1530 EDT
1530 EDT
* **
o3/o3
1.0
1.8
3
1.0
1.0
a
1.0
1.8
Table 35. The values of t^x, 03*703** (where 03* is the amplitude
of ozone in the high ozone case), 3 (using tmax and
assuming tmax = 1400 LDT) , and a for Lewisburg, West
Virginia obtained from the 1975 diurnal cirves given in
figure 70
CASE
High Ozone
Low Ozone
max
1430 EDT
1630 EDT
* **
o3/o3
1.0
1.5
a
1.0
0.1
a
1.0
15.0
Table 36. The values of tmax, 03*/03** (where 03* was the amplitude
of ozone in 1973 in all cases), 3 (using tmax and letting
Lmax
= 1400 LDT), a, the diurnal amplitude of ozone
(maximum ozone minus minimum ozone) , and the diurnal minimum
ozone concentration for the high ozone cases in 1973 and
1975 from figures 70A and 70B
YEAR
1973
1975
t
max
1530 EDT
1430 EDT
* **
o3/o3
1.0
1.02
s
1.0
3.0
a
1.0
0.34
Ozone
Amplitude
(yg/m3)
148
144
Minimum
Ozone
(yg/m3)
64
24
167
-------
times larger than in 1975. This suggests that the change in mechanisms be-
tween 1973 and 1974 or 1975 was not a localized phenomena in Pennsylvania
but was widespread.
Table 37 contains the average maximum concentration of NO found in
the period between the time of maximum ozone and midnight of the same day,
using data for August 1974 at DuBois and 1975 at Bradford (N09 data were
not available at Kane in 1973). The average NO data represent only days
when the maximum diurnal ozone concentration exceeded the NAAQS. Comparison
of these concentrations with the data in figure 65 suggests that NO alone
could not be responsible for the shift of t from late afternoon to mid-
r max
afternoon between 1973 and 1974 or 1975.
The NO concentrations given in the table were computed assuming a three-
gas system containing 0», NO and N02 in which N0? is produced through the
reaction of NO with 0_. The concentrations of NO are based on the values
of NO given in the table which are the maximum concentrations [that is,
the local rate of change of N02 is zero (8NO /3t - 0) when the concentra-
tion of N0? is a maximum]. The order of magnitude of the computed concen-
tration of NO compares well with that required to produce a 6 of the order
of 2 hours or more, according to figure 65.
For NO to account for most, if not all, of the destruction of ozone
between the time of maximum ozone and the time of minimum ozone (approxi-
mately 80 percent of the maximum concentration of ozone was destroyed in
Table 37. The average maximum concentration of N02 found between
the time of maximum ozone and midnight using only the
data on high ozone days for August 1974 at DuBois,
Pennsylvania and in 1975 at Bradford, Pennsylvania; and
the computed NO concentration at the time of maximum N02
assuming a three-gas system (NO, N02, and N03)
YEAR
1974
1975
NO 2
(pphm)
1.0
0.5
NO
(pphm)
.004
.002
168
-------
that period according to the 1974 and 1975 data, and on the average the maxi-
mum concentration of ozone was approximately 10 pphm) would require a continu-
ous injection of NO to replenish that destroyed if NO were at the small con-
centrations indicated in table 37. The reaction of NO with ozone would
produce NO^ one-for-one for each reaction. The 1974 and 1975 data suggest
that approximately 8 pphm of N0« should be found some time in the period
during which the ozone is being destroyed. Since the observed concentra-
tions of NO were never much more than 1 pphm, it must be concluded that
either reaction with NO is not the mechanism by which ozone is being de-
stroyed or that NO is being removed almost as fast as it is being produced
Data do not exist from the field programs accomplished over the 3 years to
discriminate which condition is real.
There is, of course, the question of whether the requirement that NO
is replenished by continual injection, presumably from the surface, could
be satisfied. If it were not, this would lend support to the notion that
the ozone is destroyed through reactions with gases other than NO. Again,
this is a point which could not be resolved with available data. However,
since the required concentrations of NO appear to be very small (the order
of 1/100 ppb), it is certainly possible that continual injection of NO
could replenish such small concentrations.
7.2.6 Summary of Northern High Pressure Analysis
Many of the results obtained in the preceding analysis were based on a
comparison between the results of a linearized mathematical model which
demonstrated the diurnal behavior of ozone and observed data in various
rural regions of the United States. The model mainly treated the mechanisms
synthesis and destruction of ozone, with no regard for the vast number of
chemical processes required to produce, accumulate, and destroy ozone. The
amplitude of synthesis treated the production of ozone, and the parameter
X treated the existing concentration of ozone destructive agents. Because
of the assumption made in the model, specific numerical values are unreli-
able. However, application of the model results was in the domain of com-
parison between two ozone regimes using observed data. In the relative
sense (e.g., stating one regime has more ozone destructive agents than
another regime), these results are believed to be reliable. Furthermore,
some of the explicit results determined by applying model results with ob-
169
-------
served data could have just as easily been determined by deductive reason-
ing. The use of the model simplified this effort.
One important assumption made in applying modeling results with ob-
served data was that the time of maximum synthesis (t* ) was near the time
J max
of maximum solar radiation. The absolute rate of synthesis is closely re-
lated to the interactions between solar radiation and NO . Since observed
4/ *•
diurnal variations of the NO in rural regions— show that NO varies by
about 10 to 20 percent of its mean value, and the mean value is approximately
1 pphm or less, this suggests strongly that the time of maximum synthesis is
related to the time of maximum solar radiation.
The data presented support the premise that high ozone (ozone concen-
trations in excess of the NAA.QS) found in the summer months is associated
with high pressure systems. The data further indicate that sustained peri-
ods of high ozone are associated with a macroscale high pressure system
having periods greater than 20 days. The largest concentrations of ozone
were found on the back side of a moving high pressure system, and a relative
minimum was found on the front side or near the center when the system was
in the Midwest and East. The location of the maximum and minimum concentra-
tion of ozone in a moving high pressure system correlated with the location
of air having maximum and minimum residence time in that system; that is,
largest concentrations of ozone occurred on the back side of a high pressure
system where air parcels had the largest residence time, and the smallest
concentration of ozone was found in the front side of the system where the
air parcel had the smallest residence time. For an eastward moving high
pressure system whose speed is of the order of 5 m/s or more, the air ini-
tially in the northeastern quadrant of that system will have large residence
times.
The data from the stations in the western portions of the United
States did not display a preference for high ozone in the back side of the
high pressure system. This, suggested that a mechanism which increases the
ozone concentrations in air parcels was enhanced as they moved from the
west to the east. Meteorological analysis did not reveal any reason why,
on the average, downward transport in high pressure systems, either by the
general subsidence or by enhanced vertical mixing, should increase the ozone
concentration in the back side of the system. Low-level divergence and down-
170
-------
ward motions are usually greatest near the center or in the front side of an
eastward moving high pressure system. Furthermore, available aircraft data
showed no evidence that, on the average, the vertical gradient of ozone
through the troposphere is greater in the eastern portions of the United
States than in the western portions so that this factor could not be respon-
sible for enhanced transport. It can only be concluded that transport of
ozone was not the mechanism creating the larger concentrations on the back
side of the high pressure system, leaving enhancement of the synthesis of
ozone as the potential mechanism.
Examination of the change in the diurnal variation of ozone using the
linearized model and the 1975 data indicated that both the concentration of
destruction agents and the synthesis of ozone were larger when the high
pressure system was in the East than in the West. This suggested that the
concentration of ozone precursors and ozone destructive agents (often ozone
precursors and ozone destructive agents are identical) were larger when the
system was in the east than in the west. The number of industrial complexes
and population centers is considerably larger in the East, suggesting that
the source of the ozone precursors and destructive agents is anthropogenic.
Ozone concentrations were highest on the back side of a high pressure system,
where air parcels had the largest residence time. This suggests that synthe-
sis of high ozone requires sufficient time for injection of sufficient amounts
of ozone precursors in the system where proper environmental conditions exist.
Even though there was a significant increase in synthesis when the high
pressure system was in the East, the amplitude of the diurnal curve for ozone
in the East was not significantly greater than that in the West. This is
thought to be a result of the increased destruction of ozone. The larger
maximum ozone concentration found when the high pressure system was in the
east was a result of a diurnal minimum ozone concentration. The data com-
bined with the modeling results indicated that the diurnal minimum could be
increased due to the small concentrations of ozone destructive agents in
local air parcels in the western portions of the United States. The negli-
gible ozone destruction will allow ozone residuals to exist at the end of
a diurnal cycle which can "pump up" the level of minimum ozone concentra-
__ «*
tions as these air parcels drift eastward. (Nota Bene: air parcels do not
move with high pressure systems since their residence time is much less
171
-------
than the period of the system. See section 7.2.4.) The data further indi-
cate that the apparent injection of ozone destructive agents into the sys-
tem in the East should prevent further build-up of minimum concentrations.
The building up of the level of the diurnal minimum ozone concentration
and the diurnal synthesis of ozone were sufficient to produce ozone concen-
trations in excess of the NAAQS in high pressure systems located in the
East.
The evidence further indicates that there has been a reversal of the
role of synthesis and destruction of ozone in high pressure system located
in the East from 1973 to 1974 and 1975. Smaller concentrations of ozone
destructive agents in 1973 allowed a larger build-up of the level of the
diurnal minimum ozone concentration and were influential in producing a
reasonable diurnal amplitude for ozone (the maximum ozone concentration
minus the minimum ozone concentration). These two factors combined to
produce diurnal maximum ozone concentrations in the boundary layer in ex-
cess of the NAAQS. In 1974 arid 1975, the concentration of ozone destruc-
tive agents was considerably larger such that the level of the minimum was
somewhere between one-half and one-fourth that found in 1973. However,
there was a marked increase in the synthesis of ozone in 1974 and 1975 com-
pared to 1973, which produced ozone in excess of the NAAQS. The maximum
value of ozone and the number of hours of high ozone in 1973 were greater
than that found in 1974 and 1975, suggesting that the mechanism of producing
a very high level for the diurnal minimum ozone concentration was more im-
portant in terms of producing a greater number of hours [this was not only
true for the Pennsylvania stations, but Lewisburg, West Virginia, also had
a greater number of hours of high ozone in 1973 (156 hrs) than in 1975 (26
hrs)] and larger concentrations of high ozone.
The summer of 1973 was characterized by having the largest total number
of hours of high pressure and the largest total number of hours of high
ozone in the east, compared with that of 1974 and 1975. Some of these
high pressure systems became relatively stationary. In particular, one
high pressure system remained stationary in the East for approximately 14
days in August. Long periods of relatively stagnant conditions would allow
air parcels within the high pressure system to have large residence times
and to experience many diurnal ozone cycles increasing the level of the
172
-------
diurnal minimum and depleting the concentration of ozone precursors and
ozone destructive agents. The results would be to produce a mean diurnal
cycle for ozone similar to that obtained for high ozone days in 1973
(fig. 69A). However, this requires that the injection of ozone precursors
and ozone destructive agents was not large enough to compensate for the
depletion of these constituents. It may also be possible that the high
level of ozone, both day and night, removed both NO and NO- (where NO is
an important destructive agent, and NO is not only a destructive agent,
but also an important ozone precursor) after it was injected into the sys-
tem without markedly affecting the ozone concentration.
An alternative hypothesis to that dealing with differences in high
pressure systems, which might explain the differences in 1973 compared to
1974 and 1975, is that there was an increase in the injection of ozone
destructive agents from anthropogenic sources in 1974, and a further in-
crease in 1975. This hypothesis would explain the systematic decrease in
the value of the diurnal maximum ozone concentration and the decrease in
the total number of hours of high ozone. Data were not available to ex-
amine this hypothesis completely.
The results of the analysis are highly indicative of the nature of
ozone behavior in a rural boundary layer and in the presence of a high
pressure system. However, since many of the results of the analysis are
based on a simplified and unvalidated mathematical model, they are not con-
clusive. It is suggested that in the future, a more sophisticated mathe-
matical model, which describes the ozone chemistry in considerably more
detail, be used to test some of the hypothesis concerning the synthesis
and destruction of ozone made in this analyses. It is further recommended
that the results of the more sophisticated mathematical model be used as
a guideline for a limited number of chamber experiments to validate the
mathematical model. It is only in this manner that the chemistry of ozone
in the rural boundary layer and in the presence of a high pressure system
can be completely understood.
7.3 Chemistry of Ozone Generation
In the 1950's, many writers spoke of the "smog barrier." This was the
point at which emission of pollutants to the air or concentration of pol-
lutants in the atmosphere was presumed to reach such a level as to produce
173
-------
the symptoms of photochemical smog.
Today, 20-odd years later, a satisfactory three-dimensional character-
ization has not been made of the conditions of precursor identity, concen-
trations, ratios, time, and physical environment needed to produce certain
objectionable levels of photochemical smog. The inflection point which
would locate the "smog barrier" has turned out to be more nebulous than
earlier optimism would have imagined. With the 1975 summer data, however,
a fairly clear verbal model s.nd some quantitative limitation for the gener-
ation of high ozone concentrations in rural areas in the Midwest and Mid-
Atlantic areas of the United States can be constructed.
As a result of data collected in the northern part of this study, an
explanation of observed ambieint ozone levels in high pressure systems was
given in section 7.2. Within a high pressure system the ozone concentra-
tions begin to drop after the passage of the leading front. After the
frontal passage the ozone concentration drops to a minimum for that high
pressure system, then begins to rise before the passage of the center of
the high. The ozone concentration continues to rise to a maximum for the
system in the backside of the high (i.e., western side of an eastward-moving
high pressure system).
On the front side of the high, the air has been in the system usually
less than a day. Theoretical treatment indicates that the air in the trail-
ing side of the high has remained in the system 2 or more days and has
moved from 0 to 160 km per day, depending on the speed of the high pressure
system and the location of the air parcel at the beginning of the period of
calculation.
Additionally, it was pointed out that in 1975, while approximately 95
O
percent of the hours of high ozone concentrations (- 160 yg/m ) occurred in
a high pressure system regardless of the distance of the high pressure center
from the station, only 12 percent of the hours of high pressure were associ-
ated with high ozone. This can be related to the fact that air on the back-
side of the high has had more time to accumulate ozone precursors and more
reaction time for ozone to be generated and accumulated.
The probable chemistry of ozone generation as it relates to the move-
ment of a high pressure system from the northern Great Plains, south of the
Great Lakes, and on to the eastern seaboard can be explained in general
terms.
174
-------
In the western Plains area away from large cities like Denver there are
few people and few precursor emissions. As a high pressure system moves
through this area, the precursor concentrations tend to increase and with
the extended period of limited ventilation (and therefore greater times of
reaction) ozone concentrations will tend to increase. Seldom if ever, how-
ever, will the ozone concentrations reach the NAAQS. The key here is that
the emission rates of precursors are not sufficient enough to allow the
o
generation of high concentrations (i.e., ^ 160 yg/m ) of ozone.
Somewhere along a line drawn from Fargo, North Dakota, to Dallas,
Texas, the population density (going from west to east) increases by nearly
16-fold (~13 to -213 people per square mile) (tables 38, 39, and 40). Be-
cause of the increase in population density, there is a substantial increase
in anthropogenic pollution emissions so that, under a high pressure system,
the pollution emissions are sufficient enough and have time, occasionally,
to concentrate and react to form net ozone concentrations equal to or sur-
passing the NAAQS.
At this point, the role of ozone in "spent" photochemical systems and
of ozone left over from the previous day's generation should be described.
Table 38. Population density for States west of
Fargo, N.D. •> Dallas, Texas line*
States
Montana
Wyoming
Colorado
North Dakota
South Dakota
Nebraska
Kansas
Population
694,409
332,416
2,207,259
617,761
666,257
1,483,791
2,249,071
Density of Population
(By Square Mile, 1970)
4.8
3.4
21.3
8.9
8.8
19.4
27.5
Area (Square Miles)
145,587
97,203
103,766
69.273
75,955
76,483
81.787
Total 8,250,964
Average Density 12.7
*
U.S. Bureau of the Census, Census of Population and Housing: 1970.
175
-------
Table 39. Population density for states between Fargo, N.D. -»• Dallas,
Texas line and east of Chicago, Illinois •> St. Louis, Missouri line*
States
Iowa
Missouri
Illinois
Population
1970 Census
2,825,041
4,677,399
11,113,976
Density of Population
(By Square Mile, 1970)
50.5
67.8
199.4
Area (Sq. Miles)
Land Area Only
55,941
68,995
55,748
Total 18,616,416
^Average Density 103.0
U.S. Bureau of the Census, Census of Population and Housing: 1970.
Table 40. Population density for states east of
Chicago, Illinois -> St. Louis, Missouri line*
States
Maine
New Hampshire
Vermont
Massachusetts
Rhode Island
Connecticut
New York
New Jersey
Pennsylvania
Ohio
Indiana
Michigan
Delaware
Maryland
Virginia
West Virginia
Kentucky
Total
Population
1970 Census
993,663
737,681
444,732
5,689,170
949,723
3,032,217
18,241,266
7,168,164
11,793,909
10,652,017
5,193,669
8,875,083
548,104
3,922,399
4,648,494
1,744,237
3,219,311
87,853,839
Density of Population
(By Square Mile, 1970)
32.1
81.7
47.9
727.0
905.0
623.7
381.3
953.1
252.3
260.0
143.9
156.2
276.5
396.6
116.9
72.5
81.2
Area (Sq. Miles)
Land Area Only
30,920
9,027
9,267
7,826
1,049
4,862
47,831
7,521
44,966
40.975
36,097
56,817
1,982
9,891
39,780
24,070
39,650
412,531
Average Density 213.0
*
U.S. Bureau of the Census, Census of Population and Housing:
1970.
176
-------
In an urban air pollution system there is continued input of ozone precur-
sors, many of which are also ozone destructive agents (e.g., NO, alkenes)
throughout the 24-hour day. In an urban area these ozone destructive agents
frequently are produced in quantities sufficient enough to drive the ground-
level ozone concentration to a measured zero at night. As a photochemical
system moves out of the urban area, the rate of injection of fresh reactants
decreases, and, when the reactant injection rate is relatively low enough,
some ozone concentration is maintained overnight. Aloft, between the top of
the previous day's mixing layer and the top of the ground-based radiation in-
version, the ozone concentration is often higher than it is at the ground,
since it is temporarily out of contact with sources of pollution at the
ground. After the sun rises, the incident solar irradiation begins both
to erode the radiation inversion, bringing higher ozone concentrations to
the ground, and to initiate photochemical activity, which leads to ozone
generation. Often, but riot necessarily always, exceeding the NAAQS is a
matter of de novo synthesis of only a relatively small amount of ozone,
which, added to the ozone left in the air overnight, allows the atmospheric
ozone concentration to reach or exceed the NAAQS.
7.3.1 Relationship of Ozone and Population Density
Consideration of the following data was instrumental in deriving the
foregoing picture of the ozone behavior with the progression of a high pres-
sure system from the northern Plains to the Atlantic Ocean.
The major objective around which the northern high pressure study was
designed was that, under conditions of high pressure, ozone concentration
is correlated with population density. Based on visual observation of a
number of population maps, the study area was divided into three parts.
This was done by drawing two lines, one from Fargo, North Dakota, to Dallas,
Texas, and one from Chicago to St. Louis (fig. 71). The western area,
-2
containing Wolf Point, Montana, was the least populous (~13 persons mi ,
table 38); the middle portion, containing Creston, Iowa, was intermediate
_2
in population density ("103 persons mi , table 39); and the eastern por-
—2
tion containing Bradford, Pennsylvania, (~213 persons mi , table 40) had
the highest population density. Only the States over which the plane
actually flew were considered in this analysis.
177
-------
The ozone concentrations associated with the various stations and areas
are presented in tables 41 and 42. Table 43 summarizes the ozone half-life
information. Half-life was calculated from the 0200 to 0500 data, assuming
a first-order decay rate. (A number of cases were not counted, those in
which ozone concentration increased and several in which the t ,„ was ~1700
hours.)
It is unlikely that the ozone destruction is entirely gas phase, and
it is probable that fresh material injected into the air near the ground in
addition to residual pollution left from the previous day has reacted with
the ozone at night. Undoubtedly the air above the radiation inversion was
less affected by ozone destruction. The numbers presented in table 43, how-
ever, do represent an upper ILmit for the destruction of ozone by residual
250 or Over
50-249.9
10-49.9
Under 10
Figure 71. Population density by counties: 1970.
178
-------
Table 41. Summary of ozone data for rural stations
Stations
Wolf Point, Montana
Creston, Iowa
Bradford, Pennsylvania
. _ No. of Hours
Average 03 3
uem J
June-September
57.6 0
70.4 17
81.4 100
Number of Hourly
Samples
2160
2116
2332
Table 42. Summary of aircraft ozone data for
population density areas
Sample Area
Average 0,
Np_. of Samples
~160
Number of
Individual Samples
Least Populous Area
Intermediate
Most Populous Area
68.0
88.0
105.8
3*
0
95
1429
578
1163
These samples were taken between Sioux Falls and Sioux City almost on the
Fargo-Dallas line.
Table A3. Dark phase ozone half life
Station
Bradford,
Pennsylvania
Creston, Iowa
Wolf Point
, Montana
Half-Life
Mean (hrs)
13.
20.
15.
1
4
4
Stcl.
15.
28.
15.
Dev.
2
5
9
2.
3.
3.
Range
8-75.9
4-180.9
6-70.7
Case Count
65
60
61
179
-------
gases from the day before. The actual half-lives calculated should at times
allow enough ozone to remain overnight to furnish a significant base on
which to build a concentration over the NAAQS the next day.
The concentration of 0 occasionally increased from 0200 to 0500 LT.
Since there was no sunlight to initiate photochemical reactions, this is an
indication of the advection or movement of air containing high concentra-
tions of ozone into the vicinicy of the sampling site.
With minor variations the population density and pollution emission
rates increase from the Great Plains to the Atlantic seaboard. In general,
the farther east the more quickly the pollution concentration conditions
3
which will generate concentrations of 160 yg/m of ozone are achieved, so
that both the average concentration of ozone and the frequency of exceeding
the standard should increase. The data substantiate this reasoning.
The nature of the high ozone system in the midwest is that of an air
mass characteristic rather than an urban plume or "ozone-shadow." Although
25/
the Bell Telephone interpretation— shows a large urban effect of the New
York Megalopolis, cities in other areas usually show only as perturbations
in areawide systems, which extend upwind, downwind, and crosswind of the
city. No individual area appears to be the sole source of the plume that
trails across half the continent, but the air on the backside of the high
pressure could have come, from several areas from a few kilometers to a few
hundred kilometers away.
The amount of ozone generated, as well as the concentrations of ozone
generated, may depend more on the mass of injection of pollutant precursors
(particularly NO ) than on the concentrations of these materials at any given
X
time. Thus, one might find that ozone concentration correlates better with
population density than NO concentration correlates with either ozone or
X
population density. Computer simulation shows that under some circumstances
the same mass of precursors can.generate higher concentrations of ozone
with lower ambient concentrations of NO . Computer simulations were run
with the following parameters: hydrocarbon at 0.25 ppmC; .071 ppm batch
—1
NO and .005 ppm initial NO (NO + NO ) with an injection of .001 ppm m
X X ^-
for 660 min. (total 0.071 ppm NO calculated). The results are presented
X
in table 44.
180
-------
Table 44. Results of computer simulation runs
HC ppmC
0.25
0.25
0.25
0.25
Initial NOX
(NO + N02)
(ppm)
0.005
0.005
0.005
0.071
Rate
Injection (ppm m )
[NO] [N02] N
____ _. n
— lxlO~4
lxlO~4 —
— —
Total
Ox (Calc.)
X(ppm)
0.005
0.071
0.071
0.071
Avg.
NO (Cone.)
^ppm)
0.003
0.023
0.023
0.045
Max. 03
ppm
0.07
0.17
0.13
0.11
There are considerable difficulties in relating aerometric data to
ozone concentration, especially on a one-to-one basis. First, it is well
known that, even in a batch ozone-generating system, the ozone increases
with time up to a point, while hydrocarbon and NO concentrations decrease.
Nitrogen dioxide, if not present initially, will increase, then decrease.
The ratio of alkenes to alkanes will decrease as will the ratio of NO to
NO . The NMHC/NO ratio will increase. Depending on the nature of the
<£> X
hydrocarbons emitted to the atmosphere and the length of time reactions
have been occurring, the maximum ozone generable by a system will change.
Figure 72 is a comparison of the 0.08 ppm isopleths for ozone concentration
obtained with a computer model utilizing propylene only and using isopentane
only as the hydrocarbon. It can easily be seen that the ozone-hydrocarbon
relationship is different for the two hydrocarbons. The relationship,
0.,-NMHC, will shift with time from somewhere near the alkene (as one ex-
treme) to near the alkane (as the other extreme). The downtown city system
should be closer to the propylene-NO system, and the rural high ozone sys-
X
tern closer to the isopentane-NO system. Tables 45 through 50 show the
variation in relationship between hydrocarbon and ozone at different times
at the sampling station and in the aircraft samples. A relationship with
ozone generation is not readily apparent, but the major purpose of the
hydrocarbon sampling was not elucidation of the chemical mechanisms of ozone
generation, but to demonstrate the presence or absence of an anthropogenic
input.
181
-------
o
25
a.
o.
o
X
1.00
.90
.80
.70
.60
.50
.40
.30
.20
.10
SLOPE ~25
SLOPE ~4
J_
JL
J_
JL
4
0 10 20 30 4O 50 60 7O SO 90 100
N02> ppb
Figure 72. General equal response curves (0.08 ppm ozone) for
an alkane-N02 system (solid line) and for an olefin-N02
system (broken Line). The two diagonals with their
indicated slopes define the HC/N02 ratios which
correspond to maximum ozone production for each type
of hydrocarbon.
182
-------
Table 45. Mean hydrocarbon and halocarbon concentrations
for selected ozone concentration ranges at Wolf
Point, Montana (July-September 1975)
Compound
Ethylene /Ethane (ppbV)
Propane
Propulene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1, 1-Trichloroethane
Tetrachloroethylene
0 Maximum
0-53
(pg/m3)
6.4
0.3
0.8
1.8
0.6
Not
Detected
0.2
0.4
Not
Detected
2.1
35.6
2.1
219.0
53.0
1.0
568.0
Hourly
54-106
(Mg/m3)
28.2
3.3
1.3
2.5
2.3
0.1
0.8
2.0
0.2
1.9
3.8
3.5
279.0
58.0
1.7
153.0
Average Concentration Range
107-159
(yg/m3)
10.0
5.2
1.4
1.5
5.6
Not
Detected
1.8
2.1
0.2
2.1
29.9
1.5
69.7
78.2
1.7
402.0
>_ 160
(yg/m3)
No
Samples
In This
Range
183
-------
Table 46. Mean hydrocarbon and halocarbon concentrations
for selected ozone concentration ranges at
Creston, Iowa (July-September 1975)
Compound
\
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1, 1-Tr ichloroethane
Tetrachloroethylene
03 Maximum Hourly Average Concentration Range
0-53
(yg/m3)
13.2
1.6
1.2
1.2
0.8
Not
Detected
0.4
0.3
Not
Detected
2.5
5.7
2.2
103.0
28.0
0.6
161.0
54-106
(yg/m3)
36.8
2.3
0.8
4.0
0.9
<0.1
0.4
1.1
0.1
1.8
8.4
0.8
300.0
43.0
1.0
327.0
107-159
(yg/m3)
51.1
3.0
2.8
2.3
0.9
1.1
0.7
1.0
0.2
5.6
2.3
1.1
178.0
51.0
0.9
286.0
>. 160
(yg/m3)
24.6
2.0
0.7
2.6
0.8
Not
Detected
0.4
0.4
Not
Detected
4.1
0.5
Not
Detected
423.0
46.0
1.5
425.0
184
-------
Table 47. Mean hydrocarbon and halocarbon concentrations
for selected ozone concentration ranges at
Bradford, Pennsylvania (July-September 1975)
Compound
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1,1-Trichloroethane
Tetrachloroethylene
0« Maximum Hourly Average Concentration Range
0-53
(yg/m3)
51.7
2.7
0.9
4.6
2.0
0.3
0.9
3.5
Not
Detected
9.3
6.6
3.6
213.0
216.0
2.3
791.0
54-106
(yg/m3)
26.6
3.8
1.0
3.0
2.0
0.1
0.8
1.1
Not
Detected
2.4
7.4
2.3
196.0
59.0
0.9
385.0
107-159
(yg/m3)
30.0
4.4
1.3
2.6
4.6
0.1
2.6
2.1
Not
Detected
1.8
4.0
1.4
409.0
53.0
1.3
400.0
>. 160
(yg/m3)
15.3
3.1
2.1
6.5
8.8
0.2
4.1
7.0
1.3
3.2
2.3
2.7
1022.0
69.0
0.8
2322.0
185
-------
Table 48. Mean hydrocarbon and halocarbon concentrations for
selected ozone; concentration ranges for aircraft
samples - Region 1 - west of Fargo, N.D. -> Dallas, Texas
()„ Maximum Hourly Average Concentration Range
Compound
0-53
(yg/m3)
54-106
(yg/m3)
107-159
(yg/m3)
>_ 160
(yg/m3)
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
N-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
N-Pentane
Toluene
0-Xylene
Freon 11 (pptV)
Carbon Tetrachloride
1,1,1-Trichloroethane
Tetrachloroethylene
11.6
0.4
0.7
0.4
0.4
Not
Detected
0.1
0.6
Not
Detected
0.7
3.4
Not
Detected
156.0
50.0
7.3
976.0
177.4
0.4
0.7
1.0
0.7
0.1
0.3
1.4
Not
Detected
0.8
95.3
67.8
92.0
38.0
1.1
152.0
No
Samples
In This
Range
No
Samples
In This
Range
186
-------
Table 49. Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges for aircraft samples -
Region 2 - east of Fargo, N.D. and Dallas, Texas line and
west of Chicago, Illinois -> St. Louis, Missouri line
0- Maximum Hourly Average Concentration Range
Compound
0-53
(yg/m3)
54-106
(yg/m3)
107-159
(yg/m3)
>. 160
(yg/m3)
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
N-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
N-Pentane
Toluene
0-Xylene
Freon 11 (pptV)
Carbon Tetrachloride
1 , 1, 1-Trichloroethane
Tetrachloroethylene
11.9
1.5
1.6
1.0
6.3
0.7
0.2
4.5
0.1
1.7
5.7
Not
Detected
88.0
21.0
1.0
218.0
8.4
0.4
1.1
0.4
1.3
0.2
0.1
1.3
Not
Detected
0.7
1.5
0.2
103.0
56.0
1.0
224.0
9.4
0.1
0.2
0.7
0.0
Not
Detected
0.0
0.2
0.0
0.2
253.0
50.0
Not
Detected
133.0
No
Samples
In This
Range
187
-------
Table 50. Mean hydrocarbon and halocarbon concentrations for
selected ozone concentration ranges for aircraft samples -
Region 3 - east of Chicago, Illinois ->- St. Louis, Mo. line
0- Maximum Hourly Average Concentration Range
v_.ompouna
0-53
(yg/m3)
E thy lene/ Ethane (ppbV)
Propane
Propylene
Acetylene
S-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
N-Pentane
Toluene
0-Xylene
Freon 11 (pptV)
Carbon Tetrachloride
1,1, 1-Trichloroethane
Tetrachloroethylene
No
Samples
In This
Range
54-106
(yg/m3)
13.3
0.9
0.7
1.5
1.2
< 0.1
0.2
2.1
0.1
1.2
4.8
0.5
136.0
29.0
0.9
98.0
107-159
(yg/m3)
22.1
1.2
0.8
1.4
1.5
0.1
0.4
2.4
0.1
1.0
6.3
1.1
214.0
41.0
2.2
353.0
> 160
(yg/m3)
11.9
1.4
0.6
2.0
1.6
< 0.1
0.6
1.4
0.3
1.4
3.5
0.2
182.0
35.0
0.9
116.0
188
-------
The NO -0- relationship is apparent from aerometric data as one moves
X J
from Wolf Point to Creston (table 51). Although the concentrations of
NO measured were within the noise level of the instrumentation (i.e.,
x -
~ 10 yg/m ), it is a fair assumption that the reason ozone concentrations
at Wolf Point never reached the NAAQS was that the air at Wolf Point was
deficient in NO . Somewhere between Wolf Point and Creston, the air occa-
X
sionally began to pass the point of NO concentration in which 0_ generation
X J
was very sensitive to NO concentration. Figure 73 shows the relationship
X
between NO and 00 at Creston and Bradford; 00 concentration as indicated
x 3 3
by these data does not appear to be particularly sensitive to NO concentra-
X
tion. As stated above, however, the 0~ concentration attained at any given
point may be more sensitive to the mass of NO emitted to the air than to
the observed concentrations. Computer simulation results (table 44) show
how this can occur.
The above discussion for NO is a reminder that from 1960 to 1970 the
x
anthropogenic emissions of NO doubled in the United States. This is the
X
period in which high nonurban concentration of ozone came forcibly to the
attention of atmospheric chemists and control officials. Although the in-
creased NO output is not at this time offered as the explanation of the
X
increased rural 0_ concentration, it is, however, the only increase in an-
thropogenic activity that tends to correlate with the apparent spread of
high rural ozone concentration.
Table 51. Summary of NO data
3 x
NOX (average) NO (6 to 9 a.m. NO emissions
Station Location .3 X . . 3 X .-2 -1*
yg/m average; yg/m tons mi yr
Wolf Point, Montana
Creston, Iowa
Bradford, Pennsylvania
1.5
9.0
7.5
<1.0 0.5
9.2 5.4
9.6 15.0
*
Average of all counties wholly or partially within a 100 mile radius of the
station; Source - NEDS Files (1975).
189
-------
0:0:
CDO
10
to o»
CVJ
(O
Ul
o
a:
X
O
u.
o
o
o 111
o
to
o
cs
o
o
o
oo
10
( ui/67/) £0 39VH3AV
(0
g
V
CM
o
M-l •
T3 /-^
cd m
H r^
PQ CT»
rH
U
nj l-i
0)
0) ^3
60 B
d 0)
a)
C
o
N
O
co
r*.
a)
190
-------
As indicated above, there are limitations that can be proved by the
kind of data obtained in most aerometric studies. Two kinds of compounds
should be measured for better modeling and understanding of ozone behavior.
First, the ozone destructive capacity of the air needs to be known, especial-
ly at night. It is possible that no single ozone destructive agent is con-
centrated enough to produce the observed decrease in ozone at night, but
that a great number of compounds may be involved. (Heterogeneous destruc-
tion at the surface of the earth has been shown not to account for all dark
4/
phase ozone destruction.— Atmospheric gas phase destruction of ozone can
easily be detected and monitored.
Another group of compounds that is important is the stable intermedi-
ates of the smog reactions—those which lead readily to hydroxyl, hydro-
peroxy and organic peroxy radical formation. The cumulative burden of these
compounds may be important in the difference between Creston's and Bradford's
0., concentrations even though the concentration of a single individual com-
pound such as formaldehyde might be in the noise level of the analytical
technique at both sites. The study of this possibility should be initiated.
When the efficacy of a control program or the legal aspects of control
are being considered, the differences in ozone concentrations that exceed
the NAAQS from one year to another should be studied carefully in an effort
to determine whether they are due to differences in emissions, control meas-
ures , or meteorological variables. The difference in frequency between 1974
and 1975 of exceeding the NAAQS at various stations is explained as a mete-
orological phenomenon in section 7.2. In any event, this study adds to the
already almost-overwhelming evidence that high ozone concentrations in the
lower troposphere can be correlated positively with population density.
7.3.2 Qzonesonde Releases
To assess the role of the vertical distribution of ozone upon the
ozone measurements near the ground, a special program of serial ozonesonde
releases was initiated. These releases documented the vertical distribu-
tion of ozone and its changes. Special attention was given to any evidence
of the intrusion of stratospheric ozone into the troposphere. The behavior
of ozone in the planetary boundary layer was also investigated. Ozonesondes
were released during three program periods from Huron, South Dakota. These
soundings were made sequentially three times a day—near sunrise, in the
191
-------
early afternoon, and after sunset—and show the greatest contrasts of ground-
level ozone during the day. Ozone data were taken during vertical profile
flights, and a quality assurance program was conducted by the EPA.
7.3.2.1 Ozonesonde Releases at Huron, South Dakota, September 5-7, 1975
On September 4, a high pressure center moved southeastward through east-
ern Montana. It moved through the Black Hills and into northern Nebraska by
the evening of September 5. As the high pressure continued its southeastward
movement away from Huron, a trough of low pressure in western North Dakota
developed into a weak cold front. The front passed Huron at about 2100 CDT
on September 6, bringing a wind change from southerly to north and northeast-
erly at the ground, which persisted into the following day.
The RTI aircraft arrived in Huron at 1730 CDT on September 5. Just
prior to landing, an ascent to 3 km MSL was made. Ozone measurements were
made continuously but are reported only during level flight near 1.2, 1.5,
1.8, 2.4 and 3.0 km as the airplane ascended. On the next morning, another
vertical profile was made to 3 km with ozone measurements reported during
ascent and descent. During the same period, an EPA team performed a qual-
ity assurance of the ozonesoride system at Huron. Their mobile unit was set
up and operative by early morning of September 6.
In the first comparison, simultaneous ozonesonde and chemiluminescence
ozone analyzer measurements were taken on ambient' air. The ozonesonde gave
3
an ambient concentration of 76 yg/m while the ozone analyzer gave a concen-
3
tration of 73 yg/m . The RTI airplane recorded concentrations of 72 and 77
3
yg/m on two low passes at 50 ft before and after a vertical profile flight.
On the next day, the ozone analyzer was audited by the EPA team on the
0.100 ppm range. The Percent Audit Error generated was -9.0 percent. Short-
ly after noon, simultaneous measurements of ambient concentration were taken
using the ozonesonde and the Bendix ozone analyzer. The ozonesonde and the
Bendix analyzer were connected to the calibrator manifold, and several ozone
concentrations were generated. The ozonesonde data were computed without
prior knowledge of the analyzer measurements; the results are given in table
52. The ozonesonde was subsequently launched.
7.3.2.2 Stratospheric-Tropospheric Ozone Distribution
A time-altitude cross-section of ozone concentration and of potential
temperature from the ground to 100 mb was constructed. Potential tempera-
192
-------
Table 52. Comparison of ozone measuring techniques
Ozones onde
(yg/m3)
49
127*
370
559
Chemiluminescent
(yg/m3)
53
176
373
559
Analyzer
The ozonesonde fell over in high wind during the comparison
and was not transmitting properly at this time.
ture, 0, is the absolute temperature an air parcel would have if it were
brought adiabatically to a pressure of 1,000 mb. It is a particularly use-
ful descriptor of the state of the atmosphere. In the absence of heat input
or removal, 6 is conserved. Rising or descending adiabatic motion can be
deduced by the vertical displacements of the 6 isopleths. When ozone con-
centration is expressed in conservative units, i.e., units which are inde-
pendent of pressure or altitude, the ozone isopleths should be moved upward
and downward with the adiabats unless ozone is being generated, destroyed,
or horizontally advected into an area.
The static stability of the air is proportional to the vertical gradi-
ent of 9. The greater the stability, the more resistive the air becomes to
vertical motion or diffusion, often trapping materials in an inversion.
The analyzed cross-section is shown in figure 74. The ozonesonde re-
leases are indicated by the black triangles at the top and bottom of the
figure. The National Weather Service radiosondes from Huron are indicated
by the open triangles. The analysis of potential temperature uses both data
sets. The 6 contours are very continuous in time at all altitudes. The
continuity shows that the independent measuring systems function and cor-
related well and that the atmosphere did not undergo any major perturbations
in its structure, especially above 300 mb.
The tropopause separates the troposphere and stratosphere. It is
identified by the minimum temperature of the sounding and is very well de-
193
-------
CS> C3> Cl>
OZONE CONCENTRATION,
POTENTIAL TEMPERATURE, °K
TROPOPAUSE
c
'to
O5
5!
2
O
>
3)
O
>
H
H
D
rn
|
o>
$
T)
9
14_!__iJ.
06 12 18 00
SEPT 5
TIME ( COT }
HURON, SOUTH DAKOTA
Figure 74. Time-altitude cross section of ozone and potential
temperature.
194
-------
fined from sounding to sounding near the 200-tnb level. The static stability
and the ozone gradient increase rapidly just above the tropopause in the 0500
September 6 sounding.
The most outstanding feature of the cross-section is the apparent dra-
matic increase of ozone above 500 mb at 2100 CDT on September 6. According
3
to the data, ozone increased, 1,000 yg/m in 7 hours at 40,000 ft, which was
followed by an equally dramatic decrease over the next 9 hours. Below the
3
tropopause at 36,000 ft, the data show an increase from 160 yg/m to almost
3
800 yg/m , followed by a similar decline. This large impulse of ozone oc-
curs at night, without perceptably affecting the atmospheric structure as
defined by five soundings. Thunderstorms were not reported, so lightning-
generated ozone is discounted. Winds above 700 mb show no meaningful change
of wind speed or direction through the entire period, so a change in
advection characteristics or a "source region" is an unlikely explanation.
Since in the next sounding, the ozone has returned to concentrations similar
to those observed at the same altitude and potential temperature as before
thi event, it is highly unlikely that the apparent event occurred.
An intensive investigation of the data-reduction process, beginning
with the strip chart record, proved futile. Possible ozonesonde malfunc-
tions, such as a restricted air intake, a battery deterioration, a reduced
pump speed, or an erroneous sonde temperature indication, were investigated
without providing a feasible explanation for the high concentrations. The
sonde unit was not recovered. The temperature and pressure data agree ex-
tremely well with the data taken just 2 hours before by an unbiased party.
In the lower atmosphere, these soundings clearly show the destabiliza-
tion of the air as the high pressure system of September 5 moved and the
cold front passed near 2100 on the sixth. Analysis of wind? above Huron
showed the effect of the passing front only below 700 mb. It would be easy
3
to suggest that the changes from a relative minimum (< 80 yg/m ) in the
3
700-mb to 500-mb layer to a relative maximum (> 160 yg/m ) is a result of
the frontal passage. The change is supported by three different soundings
before and after the frontal passage. It is probably real; however, the
atmospheric structure does not give a clue to the reason for the change.
This change does not affect the ozone distribution at the ground.
195
-------
7.3.2.3 Aircraft and Ozonesonde Profiles
Vertical profiles of ozone as measured onboard the airplane, and the
preceding and following ozonesonde releases are shown in figures 75 and 76.
In figure 75, ozone concentrations are in an expected range of values. The
ozonesonde profiles suggest that ozone is increasing with time aloft while
it is being destroyed near the ground. The aircraft profile suggests that
the transition could be ongoing. Both the first ozonesonde and the aircraft
profile suggest a well-mixed afternoon convective layer, which the sounding
verified. The nocturnal ozonesonde shows a typical destruction of ozone
at the ground at night in the lowest part of a radiative inversion. Aloft
at night, the ozone is insulated from the ground and not destroyed.
The early ozonesonde of September 6 (figure 76) substantiates the con-
cept of a nocturnal decrease at the ground and persistence of ozone aloft.
It does suggest a return toward lower concentration at 3 km from the previ-
ous nocturnal sounding. The later ozone profile suggests a well-mixed layer
of ozone, probably locally synthesized. The aircraft profiles, going up and
coming down, agree well with each other. They show a shallow layer of higher
ozone near 1.2 km (4,000 ft). These profiles agree quite well with the one
made on September 5 in magnitude and shape. The low pass values 72 and 77
3
Ug/m are in excellent agreement with the concurrent sonde and analyzer
concentration reported in the; quality assurance program. These aircraft
soundings also agree quite well with early sonde measurements near 1.4 km.
At noon, the mixing depth was probably not to that altitude.
7.3.3 Hydrocarbons and Halocarbons
7.3.3.1 Variation of Selected Hydrocarbons
A. Acetylene
Figure 77 shows the average acetylene concentration for each day of
the week at the three ground stations. Wolf Point showed a fairly uniform
concentration of acetylene during the week. Creston tended to have higher
acetylene concentrations during the middle part of the week (Tuesday, Wed-
nesday, and Thursday). Thursday was the highest (9.0 ppb) and Sunday the
lowest (1.3 ppb). At Bradford, acetylene concentrations increased in an
orderly fashion from Sunday through Saturday. Acetylene concentrations on
Sunday were the lowest and were highest on Saturday.
196
-------
w
o
H
M
H
^m
I
1
SEPTEMBER 5
RTI AIRPLANE
1730 CDT
OZONESONDE
1530 CDT
OZONESONDE
2140 CDT
40 50 60 70 80 90 100 110
OZONE CONCENTRATION, pg/m3
120
Figure 75. Vertical profiles of ozone at
Huron, South Dakota.
197
-------
H
,-J
SEPTEMBER 6
RTI AIRPLANE
1156 CDT
A OZONESONDE
A 0451 CDT
OZONESONDE
' 1354 CDT
l
I
40 50 60 70 80 90 100 110 120 130
OZOME CONCENTRATION, yg/m3
140
150 160
Figure 76. Vertical profiles of ozone at
Huron, South Dakota.
198
-------
PROPANE
n
• 6
PROPANE
tl
PROPANE
M T W T F S S
M T W T F S S
M T W T F S S
ACETYLENE
ACETYLENE
M T W T F S S
WOLF POINT
M T W T F S S
CRESTON
M T W T F S S
BRADFORD
» 4
• 2
-0 ppb(V/V)
0 Ppb(V/V)
Figure 77. Bar graph showing propane and acetylene concentrations by
day of week at Bradford, Pennsylvania, Creston, Iowa,
and Wolf Point, Montana (July-September 1975).
199
-------
Average monthly acetylene concentrations at Wolf Point show a rapid
decrease for the months of July, August, and September (i.e., 4.0, 1.8,
and 1.5 ppb, respectively). Creston, Iowa, showed an increase in acetylene
from July to September, (1.5, 2.0, 6.5 ppb). Acetylene concentration at
Bradford, Pennsylvania, peaked in August (i.e., 2.9, 4.5, 2.9) (table
18, section 6.1).
B. Propane
Propane is a hydrocarbon not found in automobile exhaust. Its presence
is thus associated with an industrial user, petrochemical processes, natu-
ral gas, or other natural emissions sources. Figure 77 shows the average
levels of propane for each day of the week at the three ground stations
(table 18, section 6.1).
C. Butane
Butane concentrations at Wolf Point rapidly increased during the period
July to September (table 18, section 6.1). The increase can be attributed
to increased use of the compound as a fuel. Other closely related compounds
such as isobutane and isopentane also increase in concentration from July to
September while acetylene decreases. At the other two stations, butane con-
centration follows the acetylene concentration (i.e. , increase in acetylene
shows an increase in butane).
D. Toluene
As the DeRidder station, two of the northern stations showed a gradual
increase in toluene concentration from July to September. Bradford, Pennsyl-
vania, was the only station where this trend was not observed (table 18,
section 6.1).
E. Halocarbons
Selected halocarbons measured at the three rural sites were, in general,
at or near the accepted quasi-geochemical background reported for each com-
pound. For example, the following percent of samples were at the quasi-
geochemical background for Freon 11 at Wolf Point, Creston, and Bradford:
60, 51, and 50 percent, respectively. The highest percentage of samples
near these background levels occurred at Wolf Point. This is not surprising,
since the population density is much lower than at the other sampling sites.
Concentration gradients; for certain halocarbons were observed in the
data. For example, tetrachloroethylene showed a concentration gradient from
200
-------
west to east. Mean concentrations measured at the respective stations from
west to east were 234, 328, and 484 ppt (table 18, section 6.1).
7.3.3.2 Hydrocarbons: High Pressure System Flight of September 6-7,
1975
This series of flights began in Des Moines, Iowa, and ended at Portland,
Maine. The flight passed over parts of Iowa and Indiana, the Ohio Valley,
the upper part of Pennsylvania, through New York State (passing near Albany,
New York) and into Maine. The flight passed over some highly populated and
industrialized areas.
Concentrations of several of the hydrocarbons (from bag samples) are
plotted in figure 78 in a linearized fashion from west to east. The first
sample was taken about 50 miles south of the Des Moines, Iowa, area and may
show some influence from this city. Hydrocarbon concentrations in the sec-
ond and third samples which were taken in an unpopulated area are substan-
tially lower. On the next day, September 7, a sample was taken upwind of
Indianapolis. Hydrocarbon concentrations were generally higher than the
last two HC samples taken on the previous day.
An increasing trend is shown as the plane flies from west to east. For
example, propane concentration increases slowly as the plane approaches the
Ohio Valley, then increases rapidly in samples taken in this area. Propane
concentration falls off again in the sample taken between Columbus and Pitts-
burgh and remains low for the remainder of the flight. Acetylene and iso-
pentane also increase in air samples taken over the Ohio Valley area.
The last two samples were taken upwind and downwind of Albany, New York.
The concentrations of several of the hydrocarbons increase (but not propane)
in the downwind sample. This may reflect the influence of Albany, or con-
ceivably the influence of the metropolitan New York City-New Jersey area.
7.3.4 Particulates, Northern High Pressure Study
7.3.4.1 Ground Site Measurements
A. Total Suspended Particulates
Twenty-four-hour high-volume samples were collected daily at the three
ground sites at Wolf Point, Creston, and Bradford. Table 53 presents monthly
and overall averages for the three northern area sites.
TSP was by far the highest at Creston in July and August. This could
be a reflection of increased agricultural activity and/or an expression of
201
-------
UJ
Z LU
< Z
J— UJ UJ
UJ >- «
Q. h- O.
O UJ O
oo o o:
I-H 0
•H
fu
CM i—
(A/A)
0)
o •>- •>-
020
202
-------
Table 53. Average daily total suspended particulate (TSP)
by month at three sites (1975)
o
Average daily TSP, ug/m
riontn
July
August
September
October
Overall Avg.
Wolf Point
36.8
23.0
27.9
29.0
Creston
92.8
88.4
48.8
78.0
Bradford
40.9
38.1
25.2
34.0
the topography and degree and type of plant life in the area, i.e., how well
the soil is held against erosion and winds.
B. Sulfate as a Percentage of TSP
Table 54 shows that TSP at the Bradford site contained the highest per-
2-
centage of SO by weight, 29 percent on the average, for the study period.
Creston had 9.2 percent and Wolf Point, 6.2 percent. The highest monthly
2-
SO, percentage was at Bradford in August (32 percent); the lowest at Wolf
Point in August (5.6 percent).
2-
The high percentage of SO. at Bradford is attributed to the large
concentration of coal-burning installations in this region. Based on data
in table 55, Bradford would be classed as a nonurban station proximate to
Table 54. Sulfate as a percentage of TSP by month
Month
July
August
September
Wolf Point
6.3
5.6
6.4
Creston
9.8
7.2
6.9
Bradford
26.3
32.2
29.8
Overall 6.2 9.2 29.0
203
-------
MH
O
co
CU
00
cd
CU
0 *
(U r-
1=14 o\
CO rH
cd i
CO vO
4-1 O\
C rH
cu •^
3
4-1 CO
•H 0)
4-» 4J
CD Cd
0 '3
0 U
•H
CU 4J
cd cd
rH O,
O *T3
•H (U
4_l "U
cd a)
px, pi)
CO
*"O 3
cu cn
4-1
U CO
CU CO
rH O
CU \-t
C/> 00
in
in
cu
, — |
cd
H
e
r§
3
O
3
/-^s
0
rH
^CU
4J
O
CU
(2
6-S
cn
6
bO
-±
^
m
Jjj,
cu
4-1
cd
13
CU
p
cu
4-1
C
t— 1
to
*^>*
cu
4-1
cd
X
o
p-1
B-S
0
M
r^l
B-?
m
E
M
^3-
c
cd
M
s**^
cn
o
•H
•4—1
cd
4-1
P^
H
N-'
B-S
cn
-^^
60
00 rH CN rH O
>"lr^CVlaiCSlr^C^C:?'~l
u-iOCNrHOOOOO
rH
O in CM CN
in *»o rH vo m o co CN
O HrH-d-inOrHOOO
H HOOtNOOOOO
ri
rH °" IS 0 0 CN
.........
inocNcnooooo
rH
00 CN rH cn CD o G>
^OCNONOrHOOrH
cn r^
vo cn r-^ rH rH
f"} f^>» Q\ -*zf rH r~H •>^" ^^ C^ i~H
• •*»••••••
CM vOOcMOOrHOOrH
O rH
rH
CL1
cn rH
CU ,0
4-13 0
cd r-l OCC!
rH o cn -ri o o
13 3 O iH -H CU
CU O -H g CO
T3 -rl Q) C 3 CU -l
C 3
3 cd
O -D 6
V4 VJ 0
III •H <(
M-l
cd M
•U Cd M
cd a) cd
O »S p4
* cd ,0
204
-------
an urban area. In fact,, it was quite far from an urban center. This high
2-
value reflects the general high-level concentrations of SO, , which are
probably found throughout the western part of Pennsylvania. Both Creston
and Wolf Point are in line with other remote rural locations.
C. Nitrate as a Percentage of TSP
Table 56 shows that TSP at the Bradford site contained the highest
percentage of NO by weight, 2.7 percent on the average for the study peri-
od. Creston had 2.5 percent and Wolf Point 1.3 percent. The highest month-
ly NO. percentage was at Bradford in September (4 percent); the lowest at
Wolf Point in September (1.1 percent).
7.3.4.2 Aircraft Particulate Measurements, Northern Study
High-volume particulate samples were collected by aircraft on at least
nine of the northern high pressure study flights. Collection occurred during
the entire flight. These flights are listed in table 57 with the correspond-
ing analyses for nitrate and sulfate. Ammonium ion was below detectable
limits in all cases.
The overall average for nitrate and sulfate ion components of particu-
late matter was lower in samples from the nothern flights than in the Gulf
Coast flights over land surfaces. The nitrate and sulfate levels were
higher at lower flight altitudes 915 to 1220 m (3,000-4,000 ft msl) and
diminished at 1830 m (6,000 ft) and higher altitudes.
The highest nitrate and sulfate levels on the northern flights occurred
in a flight (086, 087) from Dayton, Ohio, to Pittsburgh, Pennsylvania, which
Table 56. Nitrate as a percentage of TSP by month
Wolf Point Creston Bradford
July
August
September
October
1.5 2.4 2.3
1.3 2.0 2.1
1.1 2.9 4.0
Overall 1.3 2.5 2.7
205
-------
Table 57. Nitrate and sulfate:
north high pressure flights
Date
9/4
9/6
9/7
9/11
9/12
9/15
9/28
9/29
9/30
Flight
RDU-Bismarck
(055, 056, 057)
Huron-Indianapolis
(060, 061, 062)
Indianapolis-Portland
(063, 064)
Raleigh-Huron (Pierre)
(067, 068)
Pierre-Springfield
(069, 070)
Springfield-RDU
(071, 072)
Dayton-Pittsburgh
(086, 089)
Pittsburgh, Bedford
(088, 089)
Bedford-Raleigh
(090, 091)
Altitude
(ft)
6000
4000
3500
8000
3000
9000
3500
3500
3500
Overall Average
Average of 3000, 3500, 4000
Average of 6000, 8000, 9000
NO 3 804"
(ug/m3)
2.7
2.4
3.9
1.3
1.1
0.6
9.8
1.5
5.6
3.2
4.1
1.5
6.5
1.4
14.5
0.7
BD
BD
22.2
0.2
18.6
7.1
9.5
2.4
BD - below detectable
206
-------
crossed the Ohio Valley. The lowest values were found in the flight (071,
072) from Springfield, Missouri, to Raleigh, North Carolina.
7.3.5 Summary of Ozone Generation (Northern High Pressure Oxidant
Study)
Under certain conditions of high pressure, ozone concentrations and
the frequency of ozone concentrations exceeding the NAAQS increases with
increasing population. As a high pressure system moves past a sampling sta-
tion, ozone concentration decreases behind the leading front and reaches a
minimum before the passage of the center of the high. In an eastward moving
high, the maximum ozone concentration in the system occurs on the back or
west side of the high pressure system. This behavior is explained in part
by the fact that air in the front half of a high has been there a day or less
whereas air on the back side has been in the high pressure system 2 or more
days. This extra time allows precursors to accumulate and provides time for
the generation and accumulation of high ozone concentrations.
In "spent" photochemical systems as they occur in rural areas, ozone
left over from the previous day is destroyed slowly so that frequently suf-
ficient ozone is left the next day to help attain and exceed the NAAQS with
a large net ozone generation for the day. It was judged that the reason the
air in the western part of the northern high pressure study and at Wolf Point
never reached the NAAQS was that it was deficient in NO and never contained
x
the requisite amount of the precursors to exceed the standard.
207
-------
8.0 INTERPRETATION OF RESULTS: GULF COAST OXIDANT STUDY
8.1 Examination of Ozone Measurements and Meteorological Conditions
Ground and aircraft measurements were made to characterize the ozone
concentrations in the northern gulf coast area.
The spatial variability on a given day among ground stations and
aloft along flight tracks, and the temporal variability at a given station
over the course of a given day or over the study period (e.g., diurnal
variation) were examined and documented. Once these characterizations were
made, the data could be interpreted and plausible explanations of observed
phenomena could be offered.
Ozone in the lowest 3 km of the atmosphere is postulated to be a
result of (1) generation/destruction from emitted precursors/destructive
agents—local or distant, natural or anthropogenic; (2) lateral transport
of ozone from one location to another; or (3) downward diffusion of ozone
having stratospheric origin. All of these processes continually affect
the ozone concentration, but they are not completely independent.
Under "normal" conditions (if there exists such a state), ozone concen-
3
trations are of the order of 40 to 120 yg/m . Principally, extreme concen-
trations of ozone (i.e., concentrations exceeding the NAAQS), large gradients
of ozone, or upper/lower deciles of concentration were examined in these
analyses. These extremes suggest an anomaly in one or more of the three
contributory processes.
8.1.1 Data Analysis Approach
To examine the role of horizontal transport of ozone or ozone precursors,
two sets of air parcel trajectories arriving in Austin, Houston, and Neder-
land, Texas, and in DeRidder, Louisiana, were computed at 0000 and 1200 GMT
for each day between July 2 and October 31. One set used winds vectorally
averaged over the lowest 200 m of the air; the other set, from the ground
to 2 km MSL. The averaging depths were chosen to agree with trajectory
30 /
analyses of Heffter.— The lower wind is far more sensitive to frictional
influences and diurnal changes than the other winds. The formation of a
nocturnal radiative temperature inversion stabilizes the layer of air above
the ground. The inversion inhibits the transfer of momentum to the
209
-------
ground, thereby reducing the near-ground wind speed at night. In the
absence of a well-defined pressure pattern, thermally driven local circu-
lations like land-sea breeze circulations or urban heat islands can
develop. (The trajectory analysis cannot be expected to reveal those cir-
culations, since the upper air data are available only at 12-hour intervals
and at widely spaced locations, 36 km (20 miles) or more inland from the
coast).
The surface to 2-km wind is a good indicator of the mesosynoptic
scale wind field of the lower atmosphere. Since wind speed usually in-
creases with altitude, the trajectories computed with these winds show
greater movement than at the lower level. During the summer afternoons,
the lowest 2 km of the atmosphere is usually well mixed, and there is little
change of wind direction with altitude.
Air parcel trajectories were also computed for morning (0700 CDT) and
evening (1900 CDT) arrivals using the upper level winds for four points
along each day's flight path. Usually these points were chosen at major
changes in course direction. Occasionally trajectories were computed for
locations where rapid changes in ozone concentrations were observed. These
trajectories show spatial variability of the flow patterns, enable examina-
tion of the history of the air prior to a measurement, and broaden the
assessment of the role of horizontal transport in the boundary layer. The
trajectories give a reasonable, although inexact indication of where the
air has been over the previous 48 hours.
Time-altitude cross sections of potential temperature, wind component
perpendicular to the gulf coast, inversion layers, and afternoon mixing
heights were computed for the first 3 km of the atmosphere above Houston
and Lakes Charles. The Houston data were taken at 1200 GMT (0700 CDT) each
morning at the Environmental Monitoring Support Unit (EMSU) at Houston
Intercontinental Airport on weekdays only, excepting holidays. The Lake
Charles cross sections were developed only for October. They were based
on the 0000 GMT (1900 CDT) and 1200 GMT (0700 CDT) soundings by the National
Weather Service at Lake Charles. Afternoon mixing depths at Houston were
defined as the altitude of the intersection of the adiabatic lapse rate
of the maximum daily temperature with the morning sounding. The same proce-
210
-------
dure was used at Lake Charles for both morning and afternoon soundings.
Only those cross sections relevant to cases in point are presented.
The analysis permits an examination of ground and airborne measure-
ments in the context of the past history of the air, of the possibilities
for generating ozone or injecting ozone precursors en route to sampling
points, and of the implications for ozone advection in the study area.
Ozone concentrations at 10-minute intervals along the flight path,
surface pressure distributions near the flight time, and appropriate air
parcel trajectories are incoporated into a single figure for each day's
flight.
8.1.2 High- and Low-Ozone Days
The maximum daily ozone concentration at each of the four stationary
monitoring locations (i.e., Nederland, Houston, Austin, and DeRidder) were
tabulated. The upper and lower deciles of the distribution were determined
at each location. Upper level trajectories arriving in the late afternoon
(1900 CDT or 0000 GMT the next date) were plotted for the days of the upper
or lower decile concentrations. The trajectories and the associated ozone
concentrations are shown in figures 79a through 79h.
Trajectories associated with high ozone concentration at DeRidder
characteristically have anticyclonic curvature indicative of flow in high
pressure systems. The curvature is greater than with the lower decile con-
centrations—indicating transitory, not stationary, pressure systems. The
location of parcels 12 hours prior to arrival are clustered within 100
miles of DeRidder. Some parcels show almost no movement. In most cases,
the high ozone can be identified with trajectories from a potential source
of ozone precursor material, i.e., Nederland, Houston, or Lafayette,
Louisiana, during the 24 hours prior to arrival.
The trajectories on low-ozone days show a preference for overwater
paths, with small curvature and faster air movement. The overland air
parcels move over potential anthropogenic precursor sources quickly, reducing
opportunities for injection of ozone precursors in large concentrations.
Trajectories on high-ozone days at Nederland also show the large
curvature and slow-moving character, with a tendency for the air to move
211
-------
O
•H
4-1
Ctf
Vl
4J
c
0)
u
C
O
O
•H
O
CD
01
s
O
T3 -H
OJ 05
4-1 -H
03 3
•H O
CO •>
CO S-l
ctf 0)
T)
CO T)
CD -H
•H pi
M 0)
O Q
4-1
O 4J
0) CO
•!->
cS CU
n rt
4J O
N
O
•H
tfl
r*
M-i
O
cu
J-i
oo
•H
212
-------
CO
§
•rl
4J
Cfl
M
4J
C
0)
o
CJ
u
cu
T)
M
01 •
P< CO
8-S
•H
J3 CO
•H 'a
U O
CO CU
•rl -O
U T3
O iH
CO P4
CO CU
CO Q
CO 4-1
CU CO
•H
)-i CU
O C
4J O
U N
CU O
?§
60 >,
C H
•H -H
> CO
•H T3
cu
1-1
00
•H
Pn
213
-------
214
-------
215
-------
216
-------
217
-------
218
-------
-------
into Nederland parallel to the coastline. High ozone may be a function
of the station location relative to the nearby sources of ozone precursor/
destruction agents. Easterly or westerly winds are infrequent and usually
have slower speeds than the predominant southerly winds. At the slower
speed, the precursors move away from the sampling site slowly, giving the
ozone a chance to develop locally high concentrations.
The lowest ozone concentrations in Nederland occurred with trajec-
tories arriving from the south-southwest after traveling exclusively over
a narrow portion of the Gulf of Mexico. Speeds averaged only about 3.5 m/s
(~8 mph) over the last 12 hours.
There is very little preference for trajectory paths between upper
and lower decile ozone concentrations at Houston. Low wind speeds in the
past 12 hours occurred with both extremes of ozone. Upper decile trajec-
tories with the long fetch over the Gulf of Mexico are difficult to under-
stand after examining the trajectories for low ozone at Nederland. However,
in these two instances, the air had spent the last 12 hours moving north-
31/
ward across the petrochemical complex at Texas City/LeMarque. MacKenzie—
indicates that a large variance among ozone concentrations has been
measured at stations within 2 to 4 miles of the Aldene (Houston) location.
The orientation of the receptors to the source and the reaction time to
produce ozone from precursors strongly control the day-to-day variation
of ozone maxima and minima.
Low ozone concentrations at Austin are primarily associated with a
strong southerly flow over areas of low anthropogenic emissions. The air
reaches Austin 12 to 15 hours after crossing the coastline. In only one
trajectory is there a suggestion of slow movement over a large area of
precursor emissions before arrival.
High ozone concentrations at Austin occur in an east-to-southeasterly
flow. In four instances the air apparently moved over Houston, which
suggests that horizontal transport of ozone or ozone precursors from Houston
could be responsible for the high ozone measured (section 8.1.4).
220
-------
8.1.3 Summary of Aircraft Ozone Measurements and Meteorological
Conditions
1) August 7, 1975, Case Study
The simple flight plan (fig. 80), flown in the afternoon, was
designed to determine if a plume could be found downwind—to the west—of
the Nederland refinery area. The arriving air parcels had moved slowly
over central Louisiana on a southwesterly heading. During the past 12
to 24 hours, the air parcels turned, arriving at the locations along the
flight track from the southeast.
3
Upwind of Nederland, ozone concentrations were 80 yg/m . On the
south-north leg of the flight across Nederland at 225 m above ground, ozone
concentrations remain low except immediately over the city, where concen-
trations exceeded the NAAQS. Along the north to south leg, downwind of
3
Nederland, ozone concentrations rose from approximately 90 yg/m to near
the NAAQS, suggesting that the plume may have a spread of approximately 24 km
at that distance downwind. Ozone concentrations continued to rise as the
aircraft turned eastward towards Nederland, and remained above the NAAQS
until the aircraft passed east of Nederland. Ozone concentrations then
3
decreased to approximately 90 yg/m .
The trajectories indicated a more southeasterly flow while the distribu-
tion of ozone suggested a more easterly flow. The low-altitude trajectories
also indicated a turn to more southerly flow, but wind speeds were very
light. The weak winds may have accounted for the high ozone concentrations
measured directly over Nederland. The Nederland ground station reported
3
a maximum of 237 yg/m .
Data from this flight clearly demonstrated that a plume existed down-
wind of the petrochemical complex of the Nederland area. Concentrations
3
within the plume were more than 100 yg/m greater than those found upwind
or cross wind.
2) August 8, 1975, Case Study
The weak circulation pattern of the previous days continued; therefore,
easterly to southeasterly flow was anticipated through the flight track
area, as shown in figure 81. This particular flight pattern was chosen to
survey the air flow onshore along the Texas gulf coast, to survey the
221
-------
CO
co
0)
M
0)
a>
OJ
co
13
a
ca
CO 4J
•H
Bj
CO 60
a 3
efl O
M 14-1
4-1
o a
0) O
O iH
fi 4J
O 3
a rd
•H
0) M
C 4J
o co
O
OO
00
•H
222
-------
o 3
o
C
O
N
O
oo
(U
J-l
60
•H
t-i
223
-------
air flow downwind of Houston, and to sample again for a plume downwind
of the Nederland area.
The ozone distribution encountered along the flight was complex.
After leaving DeRidder and heading southward toward the coast, high ozone
concentrations above the NAAQS occurred at an altitude of 600 m MSL while
passing over the Nederland area, and diminished to below the NAAQS upon
reaching the coast. On the scuthwestward leg, ozone concentrations
approached the NAAQS, then diminished rather rapidly after passing the
Galveston Bay area. From there to Port O'Connor on Matagorda Bay, ozone
concentrations decreased by 50 percent. On the northward leg, ozone
concentrations increased toward Navasota, rising rapidly within 2 minutes
33 3
from 40 yg/m to 160 yg/m , and then 2 minutes later reaching 178 yg/m .
3
Concentrations then decreased about 30 yg/n for the duration of the leg.
On the eastbound portion of the flight, ozone concentrations were 160 ±
3
15 yg/m . About 6 miles downwind of Nederland, ozone concentrations
decreased to approximately 140 yg/m and generally remained there except
over Lake Charles.
The trajectory pattern provides some insight into the distribution
of high ozone concentrations. In the southwestern corner where lower
ozone concentrations were observed, the air arrived from the southeast,
having traveled that way for approximately 36 hours. Air arriving in
the northwestern corner apparently had a 24-hour period over land and
passed through the Houston metropolitan area. The trajectories suggest
that the air with high ozone concentrations farther east had a recent
history over the precursor source regions of southeast Texas. Air flow
in the southeast was quite slow and offered the potential for local ozone
production. The largest concentrations were found directly upwind of the
Nederland area.
On this day the second highest maximum daily ozone concentration
3 3
observed were reported at Nederland (347 yg/m ) and at Houston (378 yg/m )
Concentrations at DeRidder, however, remained below 140 yg/m .
224
-------
3) August 9, 1975, Case Study
The flight pattern of August 9 duplicated the previous day's flight,
but ozone concentrations were quite different from those observed before
(fig. 82). The air flow was from the east southeasterly with a
long but slow fetch over water. On the southwestern heading of the flight,
along the coastline, ozone concentrations were generally between 60 and
3
80 yg/m , indicating that all the onshore flow had low ozone concentrations .
Immediately inland from Port O'Connor, ozone concentrations quickly
increased to 200 yg/m before slowly decreasing further northward. In the
flight leg eastward into the wind, ozone concentrations remained low,
even across the Nederland area. The trajectory of the air arriving along
that part of the flight may have come inland between Nederland and Houston;
the resolution of the trajectories, however, does not allow that distinction,
Nevertheless, it is difficult to understand the lowered ozone concentrations
of this day from those observed the day before along the same path at the
same time. Further discussions of the data obtained on this day are
included in section 8.1.4.
4) September 19, 1975, Case Study
A standard sea breeze flight was flown in the morning and afternoon
of September 19. Pressure gradients were generally weak all along the
gulf coast area. Low-level wind flow was onshore, as it was for the .past
several days (fig. 83). Throughout the entire flight, ozone concen-
trations were among the lowest found during the study.
At the ground stations, only a very few concentrations above 100
yg/m were reported. Maximum concentrations in Austin on this day were
33 3
80 yg/m ; in Nederland, 70 yg/m , and in DeRidder, 123 yg/m .
The vertical profile of ozone concentrations taken in the southeastern
corner of the flight path showed very well-mixed profiles with ozone
3
concentrations less than 100 yg/m .
5) September 21, 1975, Case Study
The frontal system that 2 days before was poised to the west of the
study area passed through and a high pressure system began pushing down
225
-------
CM
-------
oo
0)
n
(0
CO
0)
-------
from the northwest. The sea breeze flight pattern was flown for contrast
with the measurements made before the front passed (fig. 84).
On the outward bound portion of the morning flight at 1,300 m,
ozone concentrations decreased away from the coastline. After descending
to 165 m, and retracing the same path toward shore, ozone concentrations
3
on the order of 120 yg/m were observed all the way to the coastline. Just
3
at landfall, an ozone concentration of 164 yg/m was measured but quickly
3
dropped to below 100 yg/m as the flight progressed inland.
On the afternoon portion of the flight from DeRidder at 600 m, ozone
concentrations increased, and remained high throughout the eastbound leg
over the water. The vertical profile made in the southeastern corner of
the flight showed the high ozone concentrations at flight level; but at
3
approximately 1.0 km, ozone concentrations decreased to below 100 yg/m
and remained there through the next 1.0 km of altitude. Ozone concentra-
3 3
tions increased from 95 yg/in to 120 yg/m just below 2.5 km and continued
to the top of the profile near 3 km. The shift toward lower values below
2.5 km to approximately 1.0 km was seen in the descent phase of the profile.
3
Returning to flight altitude, ozone again increased to 164 yg/m . Neither
the aircraft temperature measurements nor the Boothville, Louisiana, rawin-
sonde indicate the presence of inversion layers. It is possible, however,
that a stable layer may have been present in the upper portion of the
profile. The surface frontal zone sloped toward the north, and with a
storm approaching from the south, the front could have been farther north
than indicated on the weather maps.
Proceeding northward from the vertical profile, ozone concentrations
again increased at flight level. After passing inland of Lafayette, ozone
3
concentrations never exceeded 125 yg/m .
Trajectory analysis suggests that the air which had been over a
relatively unpopulated region (e.g., Arkansas, Mississippi, Louisiana) was
associated with lower ozone concentrations. Higher concentrations appear
to be associated with northeast flow of air that had been in the Mobile or
New Orleans area over the previous 24 hours. Ozone concentrations at the
fixed monitoring locations in Texas and Louisiana were less than
130 yg/m3.
228
-------
0)
(-1
3
CO
CO
01
0)
2
CO
T) 4J
a) 00
•H
CO r-l
o
CJ -H
C *J
O 3
a> M
C w
o co
N -H
O T3
oo
cu
M
M
•H
ta
CJ
^
CD
229
-------
6) October 10, 1975 Case Study
A sea breeze flight pattern was initiated under conditions conducive
to establishing a land-sea breeze circulation. Very little difference
was observed in ozone concentrations during the sea breeze flight until
the very southern extent of the morning flight (fig. 85). During the
3
afternoon, ozone concentrations near Lake Charles were high (371 yg/m )
but decreased over the water. Along the eastern leg of the afternoon
flight, high ozone concentrations were encountered, with landfall south
of Lafayette.
Air flow into that region is very poorly defined. Lake Charles
soundings indicated a southerly flow at about 3 m/s. Boothville, Louisiana,
farther to the east, indicates a northeasterly wind about 5 m/s. Conse-
quently, the trajectories show very little air motion over the past 12
hours along the eastern side of the flight path, where some of the high
ozone concentrations were encountered.
Both portions of the vertical profile of ozone taken in the south-
eastern corner of the flight path are consistent above 2 km (above the
mixing height). A low-level inversion is apparent in the Boothville
sounding and in the aircraft temperature profile. Below the inversion
layer, ozone concentrations on the ascent and the descent of the vertical
3
profile differ in quantity by about 80 ug/m ; but they agree on the rate
increase of ozone between 1.0 and 1.3 km.
7) October 13. 1975. Case Study
As the high pressure system moved northeastward into the Carolinas,
southerly flow returned to the coastal areas of Louisiana and Texas. The
RTI aircraft flew approximately 200 miles out into the Gulf of Mexico to
sample the air being returned to the area on the southwestern side of the
high pressure system. The flight out was made at 650 m and the return
flight 165 m. At both altitudes ozone concentrations decreased with
increasing distance from the shore. No substantial difference in ozone
concentration between altitudes could be determined (fig. 86).
Air parcel trajectories were markedly uniform over the period
(possibly because of the influence of the Boothville, Louisiana, sounding
upon the trajectories; they remained over water during the 48-hour period.
230
-------
231
-------
CU
S-l
CO
01
CU
•a .
(0 4J
CO 00
Q) -H
•H H
M M-l
O
4J m
o r>
a) cy\
M »
•u en
•rl VJ
CO Q)
,0
« O
CO 4-1
C O
O O
•H
•U V4
CO O
C C
cu o
O -H
C -w
O 3
O ^2
•H
cu t-i
C -u
o co
N -H
O T3
vD
oo
C>0
232
-------
The data associated with this flight give no indication of advection
of high ozone into the northern gulf coast from potential source regions
in the Florida panhandle or peninsula.
8) October 14, 1975, Case Study
An east-west survey flight shown in figure 87 was flown during the
afternoon of October 14. A ridge of high pressure extended westward
from a high pressure center on the Carolinas coast, north of the gulf
coastal area. Winds aloft were southeasterly at about 10 m/s at most
locations along the gulf coast. Air parcel trajectories arriving at
locations along the flight path showed rapid movement. Ozone concentra-
tions measured over the water had only minor variability, although the
"source regions" of those trajectories were quite different. The air
parcels arriving in the southwestern corner of the flight path had been
exclusively over the Gulf of Mexico for the past 48 hours. Trajectories
arriving in the southeastern corner had come from the Florida peninsula
during the same period. There seemed to be no substantial difference in
the ozone concentrations associated with these two trajectories. In
south Alabama, air parcels spent a longer period over land areas closer
to the high pressure center. The air parcel arriving on the Mississippi-
Louisiana border was chosen to investigate the role of transport in the
3
high ozone concentration (174 yg/m ) found nearby. That air moved quite
rapidly, passing over the Baton Rouge, Louisiana, area prior to being
sampled.
In general, ozone concentrations over the water were lower than
those found over land. These were also typical of afternoon concentrations
found at DeRidder and Austin during the study period. The higher concen-
trations found over land may be due to low-volume emissions from anthropo-
genic sources near coastal areas, or may be due to some enhancement from
natural sources such as the pine forests of the coastal plain.
9) October 19 to 24, 1975, Case Study
From the afternoon of October 19 to the afternoon of October 24, daily
flights at low altitudes were conducted in east Texas, most of Louisiana,
and areas of the Gulf of Mexico adjacent to these States. During this
233
-------
0)
M
CO
CO
0)
M
P.
R)
-------
6-day period, high ozone concentrations were found over large portions of
the flight tracks in all the sectors investigated. High ground-level con-
centrations of ozone were observed at DeRidder, Nederland, Houston, and
Austin. A high pressure system developed and moved east-northeastward,
leaving the area under the influence of a ridge line. Time-altitude cross
sections of air properties at Houston and Lake Charles showed very unusual
conditions near 1.5 km.
On the evening of October 18, the atmosphere at Lake Charles was
well mixed to approximately 1.7 km (fig. 88). Above 1.7 km, large-
scale subsidence which began 12 hours before was developing a stable layer.
A complex pattern of warming and cooling in and above that layer indicated
an ongoing dynamic process. The subsidence continued, lowering of the
mixing height and inversion layer until the morning of October 21 (1200 GMT);
thereafter, the subsidence diminished. The stable layer began to rise,
but its intensity did not diminish until the afternoon of October 22.
By morning of October 23, the stable layer was effectively broken and
the concentration of ozone near the ground decreased.
On the morning of October 19, a weak high pressure center was
forming in southeastern Texas and western Louisiana. The western half
of the north-south survey flight was initiated to document the gradient
of ozone which might occur with the predominantly northerly flow of air.
By midafternoon, the high pressure center was located to the south-
southeast of Lake Charles, but the pressure gradient was weak. Low
altitude air flow into the four ground locations had first been northerly,
but had returned to a weak southerly flow the latter 12 hours.
Air flow aloft began northerly, turning westerly during the course
of the day (fig. 89). The predominance of a northerly flow even
in a developing anticyclone indicated an unusual situation. Anticycloni-
cally curved trajectories are expected with the transient high pressure
system. The curvature of these trajectories was cyclonic, suggesting
that the high pressure system was being formed and had only recently begun
to establish its influence in the flow regime.
Ozone concentrations meausured during the eastern half of the flight
3
were of the order of 125 yg/m , while in the western half, concentrations
235
-------
6
0)
«
f— I
O
O
O
O
O
4-1
CO
13
0)
4->
O
O
-H
-O
C
-H
0)
S-J
C
O
•H
4-1
a
Q)
cn
en
en
o
a)
T3
•H
4-J
0)
S
•H
oo
oo
0)
J-l
60
•iH
03
O
(0
XI
236
-------
01
1-1
co
co
CU
01
a;
ed
cu
to
CO
cu
•H
J-i
o •
4J tO
O r-s
cu CT\
•r-l rH
Cfl
•H M
cd cu
JQ
« O
W 4J
(3 O
.3°
•U M
cd O
>-i 4-1
4-1
a a
cu o
CJ -H
CS *J
O 3
O ,JQ
•H
CU M
C
O
N
O
co
(Tv
00
60
•H
P^
0>
237
-------
were mostly at or above the NAAQS. Clearly, the western half of the flight
shows an area-type distribution rather than a plume-type distribution of
ozone from a localized source.
3
Ozone concentrations of 180 yg/m , near the northwest corner of the
flight path could be interpreted to be a result of ozone transport from
the Dallas-Forth Worth metropolitan area. The afternoon trajectory indi-
cates that winds in north central Texas had turned to a westerly component
for the previous 12 hours. Assuming that a trajectory arriving at the
location of the high ozone reading paralleled the path of the more northerly
parcel, the air would have passed over the Dallas-Fort Worth metropolitan
area about 12 hours before. However, the crosswind width of the plume,
about 29 km at a distance of 96 km from Dallas, suggests no plume spreading
en route since the metropolitan area is about the same cross wind
width. The evidence of a plume downwind of Dallas was inconclusive.
The change of concentration from the northwest to the northeast
corner of the flight is not easily interpreted from the viewpoint of the
trajectories. The air arriving in the northwestern corner had a path
roughly over the Oklahoma City area 24 to 36 hours before the aircraft
reached this point. The air arriving in the northeastern corner passed
over the Tulsa, Oklahoma, area, also an oil refining area, 24 hours before.
On the southbound leg of the flight, the aircraft apparently passed through
a narrow plume southwest of the Shreveport, Louisiana, area and measured
3
ozone concentrations as high as 169 Ug/m for a brief period.
By the afternoon of October 20, the high pressure center had moved
some 300 miles to the east-northeast to a position just north of Mobile,
Alabama. Central pressures increased to 1,023 mb and the 1,020-mb isobar
extended from west of Austin, Texas, to central Georgia. A high pressure
ridge extended southeastward from the high center to south of the flight
area. Southerly winds had persisted for the previous 24 hours at all of
the ground monitoring stations. Farther away from the high pressure center,
more air movement had occurred for the air arriving at Austin and Houston
than for the air arriving at DeRidder and Nederland. The eastern portion
of the north-south survey blocks was flown at 650 m along the path shown
in figure 90. The area covered was reduced from the previous day for
238
-------
co
CO
a)
S-i
a.
0)
.-H
n)
QJ
CO
CO 60
(L) -H
•H rH
M <4-(
O
•u in
a r-.
a) c^i
•i-) i-H
n)
K -
4J O
« O
CO 4J
e o
n) o
t-l 14-4
4J
a c
-------
operational reasons. Ozone concentrations near 160 yg/m were encountered
on all legs of the flight and were persistent in the northern, western, and
southern portions of the flight. Again, eastern portions remained low.
3
Highest ozone concentrations were recorded near Lafayette (284 yg/m ).
3
Of the ground stations, only DeRidder (256 yg/m ) showed ozone concen-
trations above NAAQS. Low altitude trajectories showed that air arriving
at DeRidder in the late afternoon had been in transit from the Nederland
area for the previous 24 hours. Concentrations measured there were among
the highest along the flight path. The air which arrived at Nederland
3
had passed over the Gulf of Mexico. An ozone concentration of 224 yg/m
was noted along the northern perimeter of the flight path approximately
downwind of a refining complex at Monroe, Louisiana.
Persistent high ozone concentrations during a period of north-to-
westerly flow across areas of low emission densities are difficult to
understand or to explain. The air parcels arriving in the northeastern
corner of the flight suggests an increase of concentration with time.
Twenty-four hours previously, the parcel was near the Arkansas-Louisiana
border, roughly in the area where ozone concentrations on the order of
110 yg/m were sampled. At: flight altitude 24 hours later, concentrations
3
on the order of 145 yg/m were encountered. Air parcels arriving in the
northwestern corner on this evening came from areas of east Texas where
ozone concentrations were equal to the NAAQS. Upon arrival, these concen-
3
trations increased to about 190 yg/m .
The upper air trajectory in the southwestern corner of the flight
path showed air flow across Nederland to the sample point, where it had
remained nearly stagnant for about 12 hours. Ozone concentrations at
flight level were just below the NAAQS and were the lowest of those encoun-
tered. It may be improper to use winds integrated to 2 km in this situation
since the subsidence inversion was present at approximately 1.2 km. Wind
speeds below the inversion had a southerly component of about 6 m/s while
those above, about 2.5 m/s. Southerly flow as suggested by the lower
altitude trajectories are probably more appropriate in this situation.
The subsidence inversion covered the Gulf Coastal Plain from Victoria,
Texas, to Centerville, Alabama. Consequently, vertical motion probably
was restricted to 1.5 km or less.
240
-------
On October 22, the high pressure center had proceeded further east-
ward, leaving only a ridge line with a northeastward axis. Measurements
of high ozone concentrations on the previous day at DeRidder, and the
high concentrations already observed early in the afternoon indicated that
a flight was necessary. The RTI aircraft left Lake Charles for Nederland,
then went 160 km south out over the Gulf of Mexico before returning to
Lake Charles via Nederland and DeRidder. A vertical profile was flown at
DeRidder. This flight track is shown in figure 91.
Ozone concentrations exceeded the NAAQS throughout the flight except
3
at the southern tip of the flight, where ozone decreased by about 20 yg/m .
The over-water flight was made along the axis of the wind into Nederland
at both the lower and upper levels. Trajectories suggest that advection
of air having lower ozone concentrations into the Nederland area might
have been expected within about 12 hours. The ground-level concentrations
in Nederland decreased over the next 12 hours; but more probably in response
to local nocturnal ozone destruction, rather than advection. On the
3
following day, the maximum ozone concentration was 62 yg/m lower, indi-
cating a change of air mass.
Trajectories at the higher levels indicate that the air reaching
Nederland had been carried out to sea approximately 24 hours before a
reversal of wind initiated a return flow into the Nederland area. Though
initially it might appear that the high ozone concentrations were asso-
ciated with air flow from the Gulf of Mexico, the trajectory analysis
clearly shows that the air is only returning after having passed over
sources of ozone and ozone precusors.
The low-level trajectories into the DeRidder area show a slow move-
ment of air from the Gulf of Mexico northward across Lake Charles into
DeRidder over the past 24 hours. The upper air flow on October 21 indicates
the air had begun to return northward during the past 12 hours. Time
cross-sections of wind directions perpendicular to the coast at Lake Charles
showed that wind speed was in a transition period from northerly to southerly
on the morning of October 21, while it had had a southerly component for
the past 36 hours in the very lowest layers.
241
-------
242
-------
With the advent of the southerly wind flow, the mixing depth of Lake
Charles rose from 1.2 km in the morning to 1.45 km in the afternoon. The
potential temperature (298.3°K) associated with the mixing depth, however,
did not change. The mixing depth and the potential temperature at Houston
were only slightly higher and were increasing.
The airplane flew a vertical profile to approximately 3 km at DeRidder
(fig. 92). Ozone concentrations increased with altitude to 1.2 km.
At 1.55 km, ozone concentrations decreased and remained constant thereafter
to the top of the sounding. The ambient temperature decreased with altitude
to 1.55 km. Over the next 300 m, the temperature was isothermal and the
dew point temperature decreased by 11.2° C. On the way down, ozone
3
concentrations remained below 110 yg/m above 1.5 km. In the next 300 m
3 3
of descent, ozone doubled to 216 yg/m , and slowly decreased to 178 yg/m
in the lowest 300 m of the air. The aircraft temperatures are in reason-
ably good agreement with the late afternoon soundings taken at Lake
Charles, which showed a nearly isothermal layer extending from 1.0 km to
about 1.7 km. On the return trip from DeRidder to Lake Charles, ozone
concentrations also remained above the NAAQS between 300 and 600 m above
the ground.
On this day, trajectories arriving at Houston and Austin showed a
well-developed southerly flow, having been on the back side of the high
pressure system for a longer time. Afternoon concentrations in the
3
Houston area were 137 yg/m , a fairly low value for that station, and con-
3
centrations at Austin had decreased to 124 yg/m from a previous day of
156 yg/m .
By the afternoon of October 22, the influence of the high pressure
ridge had weakened through the area. The pressure gradient had increased
and a strong southerly flow into Lake Charles and Houston had developed
through all levels of the atmosphere. Ozone concentrations between 99 and
3
122 yg/m were recorded during the EPA aircraft flight on October 22 with
no particular pattern to the concentrations measured (fig. 93). Ground-
level ozone concentrations at the four stations were comparable to those
measured by the EPA aircraft. These measurements indicate a relatively
uniform distribution of ozone in the air mass.
243
-------
Temperature, (°C)
5 10 15
20
25
-I I I
I I I
I I
O Temp.
O Ozone
12
10
40 80 120 160
Ozone, (yg/m3)
200
240
Figure 92. Ozone concentrations and temperature from
vertical profile flight of October 21, 1975
at DeRidder, Louisiana.
244
-------
\
CVJ
245
-------
The air parcel trajectories indicate a long fetch over water preceded
by a turning from a northeasterly flow into a southeasterly to southerly
flow. The ozone concentrations were slightly higher than "background"
concentrations. The turning of the trajectories from the northeastern
into a southerly flow suggests that the air might have had an earlier
origin over the gulf coast region. (Caution is advised since air in the
central Gulf of Mexico is distant from the sounding stations). Mixing
depth increased slightly to about 1.75 km in the morning sounding at Lake
Charles, slightly higher than the day before and penetrated slightly into
the base of the inversion layer aloft at 1.6 km. By afternoon, the stable
layer had arisen to around 2.2 km, and the afternoon mixing depth reached
only about 1.8 km. Warm air apparently associated with the increased
onshore flow left a relatively unstable column of air within the mixed
layer. The ventilation of the near coastal area had increased substantially,
and ozone concentrations decreased.
For the next 2 days, October 23 and 24, strong onshore winds persisted.
Throughout the first 3 km of the air, mixing prevailed. Winds aloft exceeded
10 m/s through most of the first 2 km of the atmosphere. The lowest maximum
ozone concentrations reported during the measurement program occurred at
Nederland and Houston. The increased wind speeds have reduced the local
residence time of injected ozone precursors, increased the turbulence of
the atmosphere giving better dispersion, and probably have inhibited develop-
ment of the oxidant potential on the Texas gulf coast.
3
Ozone concentrations between 68 and 93 yg/m were measured during EPA
Flight Pattern No. 4, flown on October 24 (fig. 94). These values
were consitent with concentrations found onshore at ground measurement
stations. The strong southerly flow at this time gives no indication of
having had any recent history over any continental areas.
10) October 30, 1975 Case Study
A double-box pattern was shown in section 3.1 (fig. 6). It was
designed to give maximum coverage over an area close to a suspected source
of precursor pollutants. By enclosing a box within a box, at least two
measures of a downwind plume were possible during the course of the flight.
246
-------
\
CNJ
QJ
(-1
3
co
CO
Q)
.-I
0)
CU
cfl
-------
Measurements upwind of source regions and to the outer edges of the box
parallel to the wind help delineate the influence of the source region.
Across the diagonal, measurements were made at different times in the same
location to help identify changjes of concentrations during the elapsed time
and/or variances in the measurement techniques. A vertical profile was
flown in an area where the plume was identified on a previous pass through
the area.
On October 30, a ridge of high pressure extended southwestward to
the west of the gulf coast area. Ahead of the ridge line, large-scale
subsidence of the air established an inversion over the study area. The
morning sounding at Houston showed two stable layers, the upper one
slightly above 500 m, which defined the afternoon mixing for the station.
The mixing height at Lake Charles was below 1.0 km with an inversion
layer extending from 425 m (1,400 ft) to 760 m (2,500 ft). During the
course of the day, the inversion layer actually lowered to 370 m (~ 1,200
ft) with the top at 500 m (1,500 ft), but was not destroyed by heating.
Winds remained northeasterly to northerly in the mixed layer at about 6 m/s.
The double-box pattern was a joint RTI-EPA flight effort, with RT1
flying a double-box pattern, which was enclosed by a larger double-box
pattern flown by EPA. The boxes were oriented along the northeasterly
flow of air with the longer sides of the boxes to the southwest of Nederland.
The upper-level trajectories associated with this flight show a persistent
northeasterly flow for 24 to 36 hours before arrival (fig. 95). Below
200 m, air flow is not as strong, but it is persistent.
Combined analysis of ozone concentration using the RTI and EPA air-
craft are given in figure 96. Ozone concentrations of the order of 120
3
yg/m or less, prevail throughout much of the study area. Immediately
downwind of the Port Arthur area, a very distinct plume of ozone with con-
3
centrations as high as 280 yg/m and a width of 48 km at 48 km downwind
of Nederland. Gradients of ozone concentrations across the plume boundaries
3
are of the order of 40 yg/m per 10 km. Outside of the plume, concentra-
3
tions remained below 120 yg/m . A brief vertical profile flown by the
RTI aircraft near the center cf the plume showed ozone concentrations of
248
-------
249
-------
o\
T3 H
C
rt •>
O
w j-i
o>
C ,a
o o
4-1
T) O
0) O
J-l
3 "
CO CO
fl m
cu X
*£
co
•H 4J
•U M
3 .X
rH 60
cd -H
C rH
CU
M
00
•H
250
-------
236 yg/m3 at 305 m (1,000 ft), decreasing to 124 yg/m3 at 600 m (2,000 ft),
and remaining below 120 yg/m to 1.5 km (5,000 ft) (fig. 97).
The emergence of the plume immediately downwind of the Nederland area
suggests that the ozone was formed quite rapidly in the 24-km flight dis-
tance between Port Arthur and the 236 yg/m measurement. With a wind
speed of 6 m/s (~13 mph), the travel time is slightly over an hour. The
reactions continue even further downstream so that the 280 yg/m might be
representative of a 3.5 hour transport from the petrochemical complex near
Nederland. Plume dimensions stay farily uniform showing only moderate
spreading from a point downwind of Nederland.
There is no question that the petrochemical complexes surrounding
Nederland supplied the necessary precursor materials for the resulting
ozone plume. Background values of ozone are not high. There is no
apparent mixing with the air above the inversion. The plume can only be
generated by synthesis from the local emissions.
This regime emphasizes one of the difficulties in measuring ozone or
oxidants near sources of precursors. The maximum ozone concentration
3
recorded at Nederland on this day was 96 yg/m . In Houston, the Aldene
3
station attained the maximum value of 72 yg/m , while DeRidder, upwind of
both locations, had a maximum value of 107 yg/m , On the basis of the sur-
face ozone measurements, this day would probably have been classified as a
"low ozone case" for the gulf coast area. It is one of the lower decile
concentration days at Houston. It is one of the upper decile concentration
days at Austin, for reasons unrelated to the Nederland area.
11) October 31, 1975, Case Study
Area surveys were concluded on October 31 with the RTI and EPA air-
craft flying complimentary box patterns over the land (RTI) and over the
water (EPA) as shown in figure 98. The common leg of the flight from
Matagorda Bay to Sabine Pass was flown by the two aircraft at different
times as they returned to Lake Charles via DeRidder.
The high pressure system of the previous day had moved further
northeastward. Air flow was changing from a northeasterly flow into a
southeasterly flow on this morning. The ridge line had moved southeast-
251
-------
OZONE CONCENTRATION, yg/nf
TOO 200 300
6.0
UJ
o
20
30
TEMPERATURE °C
LAKE CHARLES, LOUISIANA
Figure 97. Vertical profile of ozone concentration, temperature,
and winds at Lake Charles, Louisiana, October 30, 1975.
252
-------
CO
CQ
CU
H
a.
a)
>
0)
CU
CO
-a .
03 4J
ft
CO 00
CU -H
O
4J in
o r-~
0) O>
•r-it-l
to
M •>
•U ^|
ct) CU
,0
« 0
CO 4-1
fi O
o o
•H
4-> M
ct) O
c c
-------
ward, permitting the return flow. Pressure gradients and, hence, wind
speeds were beginning to increase. No significant weather was encountered
during the flight at 305 m. Upper air data indicated that the subsidence
so prevalent from the day before, had broken and that vertical mixing was
less restricted, once the heating processes had begun.
On the landward box pattern, the RTI aircraft encountered a brief
period of high ozone concentrations along the Texas-Louisiana border. It
is not clear from the available trajectories exactly where that air came
from since in the very lowest layers morning sounding (the only sounding
for the day available from Lake Charles) showed an east-to-northeast flow
of air at flight altitudes.
Another brief encounter of high ozone (~15-minute duration and 72 km
or 45 mi wide) occurred over Bryan, Texas, midway along the northwestern
leg of the flight path. Morning winds at Houston showed a southeasterly
wind at 8 m/s at flight level. Upper air data from Victoria, Texas showed
south southeasterly winds at about 7.5 m/s at flight level. A simple
calculation taking Houston as an area source 20 km wide, suggests that the
air encountered over Bryan may be a downwind plume from the Houston area.
At the indicated wind speed, transport time from Houston to Bryan (~160 km)
would be on the order of 6 to 8 hours. The aircraft passed through the
plume at approximately 1400 GST. The measured ozone had ample time to be
synthesized in the air flow. Although these suppositions are quite reason-
able, it is difficult to establish that emissions from the Houston area are
responsible for the high ozone encountered.
Ozone concentrations decreased almost as rapidly as they increased and
3
remained below 140 yg/m until the flight path began paralleling the coast-
line. From Matagorda Bay northward to Calves ton, ozone concentrations
3
remained in 145 ug/m range and increased to almost the NAAQS on approach
to Galveston. During the remainder of the flight, ozone concentrations
remained around the NAAQS and exceeded it in western Louisiana. It is dif-
ficult to postulate any particular reason for the high ozone concentrations
in the eastern edge of the flight pattern. As suggested by the trajectories
and the wind observations, a southerly flow of air returned, but it seems
254
-------
unlikely that these high ozone concentrations resulted from local transport.
The eastern end of the flight pattern changes to southerly flow later than
other locations. Opportunities for recirculation of the previous day's
high ozone concentrations into the area are small.
8.1.4 High Ozone Occurrences at Austin, Texas
Austin is a city of 250,000 people in Travis County, Texas. It is
130 km northeast of San Antonio, the nearest population center in south
central Texas. Houston is approximately 200 km to the east southeast, and
Dallas-Fort Worth about 320 km north. The annual hydrocarbon emission
2
density (all hydrocarbons) for Travis County is approximately 22.5 tons/mi ,
placing the county in approximately the upper 94 percentile of manmade
hydrocarbon emissions within the United States.* A vast majority of
these emissions is associated with fuel useage for transportation. The
2
estimated annual NO emissions are between 13 and 24 tons/mi within
x
the upper 91 percentile of all counties within the United States. Emis-
sions of hydrocarbons and NO within surrounding counties are almost an
X
order of magnitude less. The Austin emissions are almost an order of
magnitude less than those found in Harris and Jefferson Counties (Houston
and Nederland).
The most frequently occuring values of daily maximum ozone were
between 80 and 90 yg/m during the study. The arithmetic mean is 103
3 3
yg/m with a standard deviation of 38 yg/m . The geometric mean is 96.5
3 3
yg/m with standard deviation values of 67 and 137 yg/m . The upper decile
3
of daily maximum ozone concentrations at Austin exceeded 160 yg/m", which
was achieved on 10 days. The dates and times of occurrences are summarized
in table 58. On four of those occurrences, the air parcel trajectories
indicate that the air had come from the Houston area during the past 24 to
48 hours. Those cases were examined for more evidence of transport from
the Houston area. These 4 days actually comprise two cases, August 9 and
10 and September 3 and 4. In these cases, the hour of maximum concentra-
tion occurred at 1200 Or 1300 CDT, before the sun reached its zenith. In
*From National Emissions Data Summary as of May 1975.
255
-------
OO
m
•3
H
e
3
§
•3
X
cu
c
0
N
O
0
0
4->
CO
^
O
03
co cu
>> M
CO O
O M-l
0)
CN pq
cu
r^*> ^
CO O
Q ^
CU
tH CQ
Vt (3
O O
4J e 4J
a o co
CU H 3
•i—) U-l O
CO 0=
H
Ozone
5
4J
CO
|
to
Q'
CO
O
o
V-i co
O
03
m
CO "^"^
IS C>0
CU
4->
CO
Q
OvOOOinvOiHOvDvD-J'
CslOOr^vtiHinoOO-3-tH
tHtHCOCMCNiHCMCOiHiH
inooomiHcsiooo^-cM
cor-^cTvininooOiHr^r^
CNCOiHiHtH CO-J-tH
co co
cu a;
>-> >~
^> ^>
tH rH
CO JO CO ,O
OOlcflOCUCOOOOO
etH^gtnjisaia^jz:
o o
H M
£•< PH
CMrHCMrHtHCOCOiHintH
A| A|
C'OOOOOOOOO
oooooooooo
vC)CNCMVtCNCOvOCs|COr~-
r-ltHtHtHtHtHrHiHtHtH
vD>*vOr-ir**«*OOCNCMCS
ooor — ^r^voi — -r^Ocs vo
CMiHiHiHtHrHtHCNCNiH
CTi O
CO -* CM CO
O
OCO>-lrJ}-lV-IVOCO
Oi tH CN CU CU 0) CU
O -O & ,£J JZ Vi »-i
co4J4J4jeegscucu
cocacacucuo)OirafC
>,3 334J4J4J4JOO
rH 60 60 60 p, PL^ Pi4 Pu 4J 4>J
3333Q)0)CU-3<2
-------
most of the other cases, the ozone concentration reached its maximum at a
later time of day. Although the maximum concentrations did not exceed the
standard by large amounts, the violations present a significant problem in
developing control strategies.
1) August 9 and 10 Case Study
The air parcel trajectories arriving in Austin in the morning and
afternoon of August 9 have been added to the analysis of figure 81 giving
figure 99. The parcel arriving in Austin on the afternoon of August 9
was within the plume identified downwind of Houston on the previous after-
noon. The air arriving in Austin late on the afternoon of the 10th, was
also sampled approximately 24 hours before in the area of high ozone
concentrations in the southwest corner of the flight pattern (fig. 100).
Trajectories of air parcels computed for the lower level of winds indicate
24-hour transport from the Houston area to Austin across the area where
the higher ozone concentrations were measured by the aircraft on the day
before high ozone concentrations were measured in Austin.
Temperature profiles taken at Victoria, Texas indicate a stable layer
aloft near flight level on the afternoon of August 8 when low concentrations
were reported at altitude and at the ground (fig. 101). On the morning
of August 9, an inversion layer at Victoria, extended to approximately 300 m
(1,000 ft) above the ground. With clear skies during the night, a temperature
inversion, comparable in intensity and depth for the Austin area is quite
reasonable. Beginning at the 0700 CDT surface temperature, a temperature
profile for Austin was extrapolated upward for 300 m to approximate the inve'"-
sion conditions. By 1000 CDT, the surface temperature had increased enough
to dissipate the inversion. The temperature continued to increase rapidly,
but by 1300 CDT adiabatic vertical mixing to a depth of approximately 1.5
km MSL had taken place. By 1600 CDT, mixing to near 2 km was indicated.
The hourly ozone concentration measured at Austin on August 9 and 10
are shown with the average diurnal curve for the 4-month period in figure
102. In the early morning hours of August 9, the ozone concentration at
Austin increased at a rate slightly greater than the diurnal curve between
0700 and 1000. After 1000, ozone concentrations more than doubled jn two
257
-------
CM
co
CO CO
rl 1-t
3 rl
co o
co 4-1 cu
a) u -d
rl CO 4-1
0) 4->
CO CO
rt ^
CO 13
co d oo
Cti C5
T3 -H
C 4-> C
Cfl ^3 rl
t>0 O
CO -rl 0
CU i-H
•rl <4-l 0)
H
O
4-1
4-1 r-
a ON C
(U i-l O
•I-)
c«
oo
CO
rl
•H .
CO 00
4-1
(0
4->
CO
00
4J
rt
4J MO
C C Cfl
CO O /-N
CJ i-l CO <]
(3 4J i-l v-'
O 3 CO
o 43 a oo
•H H C
a> t-i a) iH
c 4-1 o. a
o co
N -ri m
O T3 O
0)
00
•H
258
-------
CU
p
co co
CO 0) CU
CU -H JS
M Vl 4J
a o
4J T3
HOC
cu cu n
> '!->
CU cfl ,-N
CU T)
CO C 60
n) a
T3 iH
c 4J a
C8 J3 M
60 O
CO -rl S
CU H
•H «4-i 0)
!-i JS
O "^ 4J
AJ r^.
o ai c
CU i-H O
•I-)
td « c
in
W
C
•P CO
^ to 3 «
•H 3 < O
« So H
3 C
•< -H 4J
CO
<)
tfl «4-l -H
f-l O M <4-l
•U MO
C C CO
CU O ^
O -H co <]
C3 4J H "-^
O 3 CU
O & (J 60
Oi H
fl 4J
O CO
N
cd -
Q.
O T3 O CU
O
O
CU
M
bO
•H
259
-------
w
Q
H
1000
Figure 101. Temperature profiles at Victoria, Texas, August 9,
0000 GMT to August 10, 0000 GMT, Isothermal (C) are
skewed. Dashed lines are the dry adiabats for the
0900, 1200 and 1500 CST temperatures at Austin, Texas
on August 10, 1975.
260
-------
00
o
•H
4J
cd
n
4J
a
-------
hours, exceeding the standard at 1200. By the next hour, concentrations
had decreased. For the remainder of the afternoon, ozone slowly decreased
3
until nightfall and returned to concentrations below 50 yg/m briefly. The
rapid rise of ozone coincides with the time the low-level inversion layer
was dissipated by solar heating and when the mixing depth increases rapidly.
The air arriving in the mixed layer in the morning was sampled the previous
3
day and showed ozone concentrations of approximately 170 yg/m . Comparable
concentrations were observed on the morning of the August 9 in Austin. The
maximum concentration was more than twice the diurnal average for 1200 CDT.
The rapid rate of increase of ozone during the morning is certainly not
characteristic of the Austin area and implies that an abnormal source of
ozone may be responsible. Since it was demonstrated that the air came
from Houston over the past 48 hours, either ozone or ozone precursor
materials from Houston have been injected into the Austin atmosphere.
The evidence suggests ozone, rather than ozone precursors, is mixed
downward from just above the nocturnal radiation inversion. If it were
the ozone precursors, then ozone generation processes would continue into
the afternoon after the precursor materials have been mixed with the low-
level emissions. The rapid increase of ozone with the breakup of the
inversion suggests that the ozone is already present. As mixing continues
the mixing volume apparently increases more rapidly than ozone can be
synthesized, diluting the ozone concentration at the ground.
On the evening of August 9, ozone concentrations decreased rapidly
with sunset but rose again to 100 yg/m around 2300 before returning to
3
values of 60 to 70 yg/m on the morning of August 10. The early morning
sounding from Victoria did not indicate a low-level stable layer; however,
inland a very strong, stable layer was indicated in the first 300 m of the
atmosphere at Dallas-Fort Worth. Clear skies and light winds during the
night were conducive to reforming radiative inversion. Ground fog and
haze, and the lowest temperature of the month, were reported in the early
morning of the August 10. Temperatures increased rapidly, dissipating the
inversion probably more rapidly than on the day before. Ozone concentra-
tions began a steady rise shortly after sunrise and reached their peak at
262
-------
1200 CDT. Concentrations again receded in the afternoon and decreased
with nightfall. The rise of ozone during the morning suggests that verti-
cal mixing again brought air down from above the radiation inversion to the
ground giving the increased ozone concentrations. Previous samples of the
3
air showed concentrations of about 160 yg/m . The air which had been
sampled arrived in the Austin area after having had a transect over the
metropolitan Houston area approximately 48 hours before.
2) September 3 and 4, 1975, Case Study
On September 2 and 3, a weak high pressure center was stationary in
southeastern Texas. The winds were poorly organized in a weak anticyclonic
circulation. Clear skies and very warm temperatures characterized the
daytime hours. A subsidence temperature inversion near 2 km was over
much of the area. Morning radiation inversions extended several hundred
meters above ground level. The air flow arriving in Austin on these days
are indicated in figure 103.
The winds shifted northerly and easterly during the day of the 3rd,
remained there overnight before shifting southerly again by late afternoon
of September 4. The first and last of these trajectories show air passing
southwest of Houston. The middle two trajectories show air passing directly
over Houston and near the Nederland petrochemical complex during the 24
hours prior to arrival at Austin. Ozone concentrations had been high at
the Houston station on September 1, but had diminished on September 2 and 3.
The lower values may have been a result of the more easterly flow changing
the source-receptor orientation of the Aldene location. Other data from
the Houston area were not available.
Hourly averaged concentrations of ozone in Austin on September 3 and 4
are shown in figure 104. In the early morning of September 3, ozone was
very near the average values through 0900. Between 0900 and 1000, a large
increase of ozone occurred and continued through 1200. By 1400, ozone
3
concentrations had decreased to below 150 yg/m and decreased during the
late afternoon. Clear skies were reported all day, and maximum temperatures
reached 96° F, giving favorable opportunity for ozone generation.
263
-------
CO
§
4-1
{X
0)
en
c
o
CO
a
0)
H
C
•H
4J •
W ^^
3 60
c c
•H 0)
t>0 Q)
•S ,
t-l ••
n) 60
C2
03 -H
0) d
•H K
M O
O B
1
•i-> n
03 U
CO
O
OJ
t-l
60
•H
264
-------
200 rri
j .a a _i a i- i
September 3^ 1375
0
1234567
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
•~iiiaB Hourly ozone concentrations
^••» Four month diurnal ozone averages
Figure 104. Hourly average ozone concentrations, September 3, 4, 1975
at Austin, Texas and the mean hourly average concentra-
tion from July 1 to October 31, 1975.
265
-------
Shortly after sunrise on September 4, ozone concentrations increased
3
at a more rapid than normal rate to a maximum value of 175 yg/m at 1300.
Ozone concentrations then began to decrease. A thundershower occurred at
1600 CDT, and ozone concentrations decreased and remained at very low values.
Skies were clear on the mornLng of the 4th. Cloudiness increased during the
afternoon as the thunderstorn activity developed.
Aircraft measurements were not available during this period, inhibiting
identification of an ozone plume downwind of Houston or Nederland. There-
fore, the case for transport is weaker and more tentative than in the
previous case.
In both cases, however, ozone concentrations increased rapidly during
the morning hours, reached their maxima at 1200 to 1300 hours, remained
over the standard for less than 3 hours and had a trajectory from the
Houston area where high ozor.e concentrations had been measured on one to
two days previously. Rapid increases of ozone in midmorning apparently
resulted from the same processes occurring in the August 9 and 10 cases,
i.e., vertical mixing of boundary layer ozone downward to the ground after
it had been transported aloft, insulated from the ground by the nocturnal
radiation inversion. These sets of circumstances were encountered in the
Ohio-Pennsylvania regions in an earlier investigation.—
The meteorological conditions of these two cases are similar and are
common occurrences during this time of year. The diurnal variations of
ozone in these two cases are very similar but atypical to Austin. The
trajectory from Houston is atypical. Transport of ozone and/or ozone pre-
cursors from the Houston area is strongly implicated as one cause of high
ozone at Austin.
8.1.5 Summary of Gulf Coast Aerial Survey
Air moving slowly over areas of large hydrocarbon emissions was asso-
ciated with upper decile ozone concentrations at urban and rural locations
in the gulf coast area. Ir. most cases, trajectory analysis showed air
with high ozone arriving from nonprevailing directions. Air—which moved
rapidly, had long overwater fetches, and showed weak anticyclonic curvature—
266
-------
was associated with lower decile ozone concentrations at all of the ground
monitoring locations.
In most cases, the highest ozone concentrations were attributed
through trajectory analysis to the principal cities or areas of high pre-
cursor emissions. These observations suggested that ozone plumes commonly
develop downwind of large precursor emission areas. The separation of urban
and industrial centers of the area against a background of low emissions
density facilitated identification of urban influence.
The double-box flight pattern very convincingly showed the develop-
ment of an ozone plume downwind of a petrochemical complex. The plume
apparently began downwind of the complex. The downwind distance is probably
proportional to the product of time of travel (e.g., a time to generate
ozone from emitted precursor material) and the wind speed, although that
relationship cannot be established with the available data. Data from
other flights also suggested that the maximum ozone will be found down-
wind of the emission data.
The location of the ozone monitor with respect to the precursor
sources and the existing wind velocity (speed and direction) showed a sub-
stantial effect upon the ozone concentrations measured in the immediate
vicinity of a large emission area. Winds from the prevailing direction
tend to have higher speeds. Measurements of ozone along the axis of the
prevailing wind direction downwind from a major source area may be biased
because of the source-receptor distance and the increased ventilation
normally associated with the prevailing wind speed.
Intercity transport of ozone or ozone precursor materials was asso-
ciated with an urban plume and was shown as a possible cause of some
violations of the NAAQS at Austin.
In the area survey flights, the mean ozone concentrations over water
were usually less than those over the land, regardless of the level of
ozone encountered. When elevated ozone concentrations were measured over
the water, the trajectory analysis usually showed the air parcel had a
recent (< 24 hr) history over continental areas, usually over high hydro-
carbon emission areas.
267
-------
When areawide ozone concentrations exceeded the NAAQS, vertical mixing
was usually restricted by a stable layer below 2 km. That layer was
usually associated with the subsidence within a high pressure system.
Although the afternoon mixing height seldom exceeded 2 km, further vertical
mixing was possible because stable layers occurred infrequently at higher
levels. Vertical profiles of ozone concentration showed sharp transitions
across the subsidence layer. The ozone below the stable layer was about
twice the concentrations above.
8.2 Chemistry of Ozone Generation
The main purpose of the study in the gulf coast area was to document
ozone behavior, especially the generation of high ozone concentrations.
Much of the area under inves tigation has a large concentration of petroleum
and petrochemical installations and a much smaller population density than
the northeastern United States.
The basic reactions of the mechanism of ozone generation in the
troposphere are assumed to be the photolysis of NO to NO and 0, the
reaction of 0 with 0 to form 0« and the oxidation of some of the NO to
NO by a species other than 0_. The details of NO oxidation will vary
mostly due to differences in identity of organic compounds present. Oil
refineries are sources of both hydrocarbons and oxides of nitrogen. When
the refineries provide a major portion of the precursor pollutants, it is
anticipated that more propane will be observed than in an urban pollution
system consisting mainly of automobile exhaust. On the other hand, acety-
lene should not provide a good index to automobile exhaust, since about
47 percent of the commercial acetylene production in the United States is
located between Houston and New Orleans. Halocarbons are also manufactured
in the gulf coast area.
Due to the proximity of the Gulf of Mexico, complex thermal layering
of the air is a common occurrence. Photographs of the area taken from
satellites have also shown that pollution from a point source can be
transported great distances without losing the identity of thplume.
Major differences between the northern area of study and the gulf coast
area as far as ozone generation is concerned are:
268
-------
1. Source areas are discrete entities in the gulf coast area
whereas source areas in the north tend to sprawl over great
distances with considerable population in any apparent
interstices.
2. The gulf coast is an oil-producing area, and refinery
emissions tend to have a smaller alkene-to-alkane ratio
than city hydrocarbons derived chiefly from the automobile.
One effect of the "clustering" of sources in the gulf coast area and
the thermal layering of the maritime air is that a plume effect can occur
in stable atmospheric conditions before a regionwide system of high ozone
can be observed. Figures 105, 106, 107, and 108 show the different aspects
of the ozone concentration picture. Figure 105 represents an aircraft
flight on a day where low ozone concentrations were observed. Figure 106
represents a flight when conditions allowed only a few areas to generate
high ozone concentration. Figure 107 depicts a flight where considerable
stretches of countryside were covered by air with high ozone concentrations
in the mixing layer. In figure 108 high ozone is evident in almost all
areas covered by the flight path.
The average ozone concentration from June through October at DeRidder,
3
the only fixed rural station in the gulf coast area, was 61 yg/m (2,136
3
hours); nitric oxide was 2 and NO 5 yg/m (2,444 cases). Seventeen of the
2,136 hours (0.8 percent) equaled or exceeded the NAAQS for photochemical
oxidants. The frequency of ozone exceeding the standard (table 13) and the
hydrocarbon data (tables 59, 60, and 61) are similar to that of the Midwest
rural areas where high ozone concentration and low concentrations of NO
X
occur. Probably the main difference is due to the difference in identity
and the proximity to large numbers of all refineries and petrochemical
installations.
The high ozone concentrations at DeRidder appear to be due both to
ozone associated with area plumes (e.g., city plumes) and to air so generally
polluted as to present high ozone concentrations as an air mass characteris-
tic. The aircraft data presented in figures 109 and 110 show two cases
3
where ozone was low everywhere in one case (<_ 19 yg/m ) and the other
3
where ozone was high on the transition flight 0> 160 yg/m ).
269
-------
0)
i
0)
ID
C/3
4-J
M
•H
t-l
-------
HENDERSON
17:29
(0-119
yg/m3)
(120-159 yg/m3)
03 (160-199 yg/m3)
0. (.>200 yg/m3)
Figure 106. Ozone concentrations observed on aerial survey
flight on June 27, 1975.
271
-------
2006Z
96°22'W
33°50'N
D1940Z
A1857Z
03 (0-119
03 (120-159 vg/m3)
03 (160-199 vg/m3)
°3 (l200
1732Z
96°19'W
26°00'N
2058Z
93°18'W
33°50'N
LOUISIANA
DeRidder (099)
A2211Z
(LP-2)
D1556Z
1638Z
93fc22'W
29'00'N
GULF OF MEXICO
Flight 107. Ozone concentrations observed on box flight pattern
on October 19, 1975.
272
-------
:44
HENDERSON
:42
HEAVY RAIN
FLIGHT 005
6/26/75
21:06
RAIN DECREASES
17:52
LIFTOFF
DRI
:44
SBI
°3 (0-119
03 (120-159 pg/m3)
03 (160-199 yg/m3)
(>200
ug/m )
Figure 108. Ozone concentrations observed on aerial survey
flight on June 26, 1975.
273
-------
Table 59. Mean hydrocarbon and halocarbon concentrations
for ozone concentration ranges at DeRidder,
Louisiana (July-October 1975)
Compound
Ethylene/ Ethane (ppbV)
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1, 1-Trichloroethane
Tetrachloroethylene
CL Maximum
0-53 54-107
(ug/m3) (yg/m3)
No Samples 36.8
Correspond-
ing to 6.7
Hourly Max.
<54 yg/m3 1.6
3.2
2.4
0.1
2.0
1.1
0.2
3.2
5.7
0.9
376.0
85.0
1.5
386.0
Hourly Average
105-159
(yg/m3)
22.2
5.1
2.4
3.1
3.0
0.1
2.1
1.9
0.2
2.8
10.8
0.9
371.0
71.0
1.3
358.0
Range
^ 160
(yg/m3)
26.1
8.2
1.5
2.7
3.5
0.2
2.6
2.1
<0.1
1.4
4.4
1.0
291.0
51.0
1.1
298.0
274
-------
Table 60. Mean hydrocarbon and halocarbon concentrations
for ozone concentration ranges, gulf coast -
flights over land (July-October 1975)
Compound
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1, 1-Trichloroethane
Tetrachloroethylene
°3
0-53
(pg/m.3)
5.9
0.3
0.3
0.9
0.3
0.1
0.1
Not
Detected
Not
Detected
0.2
3.4
Not
Detected
86.0
54.0
1.0
78.0
Maximum
54-107
(pg/m3)
46.0
3.6
1.8
2.0
2.1
0.1
1.0
3.9
0.1
2.2
11.4
4.5
201.0
46.0
0.9
141.0
Hourly Average
105-159
(yg/m3)
19.2
4.4
1.5
2.3
1.7
0.1
0.9
4.5
0.5
1.5
12.4
0.4
325.0
33.0
1.1
158.0
Range
>" 160
(yg/m3)
18.7
7.2
2.2
2.4
2.2
<0.1
2.1
1.6
Not
Detected
1.5
10.6
0.9
Not
Detected
33.0
1.3
129.0
275
-------
Table 61. Average hydrocarbon and halocarbon concentrations
for ozone concentration ranges, gulf coast -
flights over water (July-October 1975)
Compound
Ethylene/Ethane (ppbV)
Propane
Propylene
Acetylene
n-Butane
1-Butene
Isobutane
Isopentane
Cyclopentane
n-Pentane
Toluene
o-Xylene
Freon-11 (pptV)
Carbon Tetrachloride
1,1, 1-Trichloroethane
Tetrachloroethylene
°3
0-53
(yg/m3)
21.0
0.3
0.8
0.5
1.3
Not
Detected
0.2
2.7
Not
Detected
0.6
22.5
0.7
172.0
46.0
0.9
75.0
Maximum
54-107
(yg/m3)
.11.0
3.2
2.4
1.2
7.3
0.9
0.6
5.8
0.4
1.7
10.6
2.2
211.0
56.0
1.0
119.0
Hourly Average
105-159
(yg/m3)
9.6
1.7
0.7
1.2
0.8
Not
Detected
0.4
0.8
Not
Detected
1.2
7.3
0.8
179.0
42.0
1.2
(12)
138.0
(12)
Range
~ 160
(yg/m3)
21.4
9.2
2.7
2.2
4.2
0.1
3.4
4.0
1.4
3.3
10.9
Not
Detected
310.0
35.0
1.1
167.0
276
-------
CO CO CO CO
O O O O
II*
ON
3,
•H
PM
-------
m
r-.
0)
§
I-J
4-1
•a
cd
H
4J
CO
g
•rl
c
0)
O
c
o
o
§
N
O
(U
H
3
60
278
-------
The population density of the Southeastern States over which the air-
_2
craft flights took place was ~63 persons mi (1970 census) (table 62).
Comparison of the ozone/population relationship with the northern stations
is probably not valid largely because of the presence of the large oil
fields, the refineries, the petrochemical installations, and longer smog
seasons.
The average ozone concentration, as well as the average maximum,
increased month by month in DeRidder from July through October. The general
chemical system at DeRidder appears to be: low NO , high NMHC/NO ratio
X X
with opportunity for reactant concentrations and reaction times to achieve
concentration of ozone above the NAAQS. Ozone and hydrocarbon relation-
ships are shown in tables 59, 60, and 61. Because of shifting relation-
ships, discussed in more detail in section 7.2, there is little evidence
of a one-to-one relationship between the hydrocarbon and ozone concentra-
tions.
Table 62. Population density for southeastern States*
States
Texas
Oklahoma
Arkansas
Louisiana
Tennessee
Mississippi
Alabama
North Carolina
South Carolina
Georgia
Florida
Total
Average Density
Population
11,196,730
2,559,253
1,923,295
3,643,180
3,924,164
2,216,912
3,444,165
5,082,059
2,590,516
4,589,575
6,789,443
47,959,292
63.2
Density of Population
(By Square Mile, 1970)
42.7
37.2
37.0
81.0
94.9
46.9
67.9
104.1
85.7
79.0
125.5
Area (Sq. Miles)
262,134
68,782
51,945
44,930
41,328
47,296
50,708
48,798
30,225
58,073
54,090
U.S. Bureau of the Census, Census of Population and Housing; 1970,
United States Summary.
279
-------
8.2.1 Bag Irradiation Experiments
In October in DeRidder, a number of ambient air samples were taken in
Teflon bags. One bag was irradiated with sunlight without additional
reactants; a second bag, taken at the same time, was spiked with about 10
ppb of NO and irradiated at the same time as the first. Data obtained
for October 18 and 19 experiments are shown in table 63. All data for
both spiked- and unspiked-bag irradiation experiments are shown in figures
111 and 112. Obviously, in the ambient system, the added 10 ppb of NO
increased the ozone-generative capability of the system, but none of the
systems was sensitive enough to added NO that its addition would have
generated ozone concentrations over the NAAQS.
8.2.2 Ozonesonde Releases
8.2.2.1 Introduction
Recent field measurements have shown frequent occurrences of ambient
ozone concentrations at nonurban locations in excess of the NAAQS of 160
3 5 26 /
Vlg/m for more than 1 hour per year.—!— These may arise from a combina-
tion of local generation, horizontal transport of ozone or ozone precursors,
and from downward diffusion from the stratosphere. The role of the strato-
spheric ozone in the nonurban ozone problem is not fully understood. Analyses
32/
of the ozonesonde soundings reported by Bering and Borden— were concerned
primarily with documentation oJ: the stratospheric ozone. The analyses
showed high ozone normally confined to the stratosphere, with a major
33 /
increase just above the tropopause. Reiter's discussions— indicate occa-
sions when large concentrations? are brought from the stratosphere into the
midtroposphere through folds or breaks in the tropopause. The implication
is that the ozone could reach the ground in sufficient quantity to cause
high concentrations. The conditions for these occurrences are usually
found in late winter and early spring, when frontal systems are strong and
deep and when the stratosphere is relatively ozone, rich, rather than the
summer and fall periods when most of the ozone concentrations exceeds the
NAAQS at nonurban stations. Subsequent analysis of Reiter has indicated
that the maximum 24-hour average of 0« attributable to the stratosphere
3 347
is not likely to be greater than 60-80 yg/m — .
280
-------
A
IT)
O
rH
^
CO
4-1
d
CU
0
•H
M
0)
ft d
X o
0) -H
rt Cd
0 -H
*H 13
4-J Cd
Cd rl
13 *H
cd
M 00
rl Cd
•rH XI
00 0
cd o
XI rl
cfl
p5 cO
Cd 4->
*H1 flj
CO 13
•H
O cd
C
0) 0
13 iH
13 M
•H CU
P3 ft
CU X
Q cu
.
en
CU
1
H
U
CO XI
d ft <*H oo
o ft o cd
Xi pr.
M " 0)
cd -O 0 00
O 01 vH d
O M H -H
M d rH
13 CO 4-1 rH
>, Cd cfl -H
W CU fii
S
0)
en 4J
O cfl 0
M Pi
4J CU ft
in
CN
o
d
O
d
CTi
m
CN
en
o
•
o
en
o
d
oo
o
o
o
o
o
H
o
o
r-l
o
o
en
rH
o o
d d
CN CM
H rH
O O
m
in
o
rH
cu
JS
4-1
cd
0)
rl
CO
E>">
cfl
T)
4-1
o
pq
CO
C
o
•H
co
d
rH
U
C
o
CJ
rt
0)
cu
cu
00
cd
13
CU
ft <«H
CO O
O
CU
d
t-l
co
•H
CO
00
0)
X!
cd
Xi
4-1
CO
cd
en O
O N
o
MH
O XI
U
c a
o p
o
d
13
o
rl
ft
01
C
CU
rl
(U
X!
co
4-1
C
§
•H
0)
ft >4H
x o
0)
CO
00
CO Cfl
CU
CO
cu
X!
4-1
X!
4-1
O
X)
CO
cd
CO
CU
a
M
cu
o
o
a
o
rl
ft
13
CU
^i
•H
ft
co
M
O
a
cu
fi
oo
cu
d
o
1
01
4J
t-l
1
01
§
N
o
•H
cu
X!
4-1
00
d
o
d
0)
>
0)
M
a
CU
CO
cd
o
d
T-l
0)
O
d
I
j-t
•H
cd
13
CU
cu
§
4-1
O
CJ
O
XI
rl
cfl
U
o
>.
X!
0)
X!
cu
O
iH
in
•H
O
01
ft
CO
CU
Xi
rl
01
X!
X!
OO
0)
X!
CO
Cfl
4-J
CT\
CU
XI
4J
§
d
o
o
§
Xi
}-t
cfl
o
XI
a
i
CO
cd
01
a
•H
d
o 3
4-J 4J
CO
rt
o
Xi
M
cd
CJ
o
rl
d
o
co
cd
i
13
CU
13
13
cd
01
rt
o
o
CO
cu
4-1
rl
cd
4-1
CO
rt
o
•H
cu
X!
E*.
XI
01
d
rH
cd
M
0)
00
•H
X!
01
cfl
4-1
4-1
cd
00 4-1
d o
o d
X! -O
4J O
co
X!
d
o
o
CO
CO
o
CO
V.
cd
281
-------
(^
•H
Q
)^
1-1
^
4->
c
0)
43
1
•
CO
CO
*
1
O
CM
H
O
Z co
IU
a. a
a 1
w
CO t/5
•a
•3 CU
C )-i
« -H
CQ
I-l Ps
•rH
•H U
4-1 a)
•rH *^3
A. -H -H
ft pel
<4-l 0)
0 O
X
O C •
Z o co
•H W
•
-------
!Z CO
CD
•a *
a 9
CO W
iH
>> TJ
iH -O *
o
CO
o
CM
O
iH
0 0 0 O 0 0
CM O 00 vO -* CM
rH r-l
(oosi-0060) add 'asvaHDNi aNozo
'c
O
•rl
4-1
cd
^
4J
0)
a
c
o
o
0)
a
o
N
o
PQ iH
CL, cd
PLI -H
4J
" *rl
a! C
M -H
<:
<4-i
H 0
W fl
M O
CQ »H
S 4J
3 cj
c
« 3
s "^
o td
N
O CO
cd
^
00
irl
cd
cd
•H
ca
•H
o
rJ
*l
M
0)
T3
T3
•H
PS
01
n
9i
CO
4J
C
0)
a
•H
M
0)
cx
X
0)
c
o
•rl
1 1
cd
•H
XI
cd
)-l
i I
M
•rl
M
•rl
td
T3
0)
M
4J
cx
cd
o
s~*
U1
r~
CT>
r-l
•t
rH
CO
1
r-l
M
-------
Thunderstorms or other mechanisms of strong vertical mixing have
35 /
been suggested as avenues for bringing stratospheric ozone to the ground.—
Lightning discharges associated with the thunderstorms also produce ozone.
Few occurrences of high ozone at nonurban locations have been reported to
be associated with thunderstorms.
8.2.2.2 Ozonesonde Releases at DeRidder, Louisiana, October 24-27, 1975
Ozonesonde releases were conducted at DeRidder, Louisiana, during the
period October 24-27, 1975. On October 24, a fairly strong front was moving
steadily southeastward across Oklahoma and central Texas into central
Louisiana. About 0100 EDT on October 25, the front passed DeRidder and
stalled before passing Lake Charles. Heavy rains ensued over the next 24
hours, before the front began to move again. Skies slowly cleared as the
front moved to northwest Florida by late afternoon of October 25. A ridge
of high pressure established itself along a northeastward axis lying slightly
aorth of the DeRidder area.
One ozonesonde was released at 2300 CDT on October 24, several hours
before the front passed. The sounding operations were suspended on October
25 because of intense rain. Three ozonesondes were released on October 26,
and two more ozonesondes were released on October 27.
8.2.2.3 Stratosphere-Troposphere Ozone Distribution
The time-altitude cross-Election of potential temperature and ozone
from the ground to 100 mb is given in figure 113. Only the first four
postfrontal ozone concentrations were used in this analysis.
The changes of atmospheric structure as the front passed Lake Charles
were quite dramatic. Potential temperature from the Lake Charles,
Louisiana, rawinsondes augmented the DeRidder measurements.
The ozone increases with altitude above 700 mb in all of the soundings,
but the ozone isopleths cross the adiabats rather than paralleling them.
It then appears that ozone is being produced in the upper troposphere.
The major increase occurs after 0600 on October 26 and continues until
night (1800) before decreasing in the night.
The time-altitude cross-section for higher altitudes is shown in
figure 114. The tropopause is near 85 mb (~57,000 ft) where minimum
temperature of -70° C were encountered. The atmosphere began to stabilize
284
-------
OZONE CONCENTRATION,
POTENTIAL TEMPERATURE, °K
100
___,_--T-/ J/0
1000
06 12
OCT. 26
06 12
OCT. 27
DE RIDDER, LOUISIANA
Figure 113.
Time-altitude cross section of ozone and potential
temperature, lower portion.
285
-------
ALTITUDE, THC ISANDS OF FEET
SO if)
U> if)
10
£ *
4. .
UJ
. QC
fZ Q;
< LU
CC 0.
y H UJ
«p C/)
— OZONE CO
-- POTENTIA
_ TROPOPAL
1
1
1
1 1 1
» ' V" "^~ll
— l-
• i •
1 ! 8
i 8'g|
Jo PJlcxj ^
-A;\\;\ -*
\\\ Vf-
\V
A!\
?\
*!\
i
H/
w/i
t
!
U1
\l
1
:
/li
. i ' ii
vr ~V' P
V-r-Vi— \
^x"1
1 Y i
; \ i
i
, I ! ;
i i ii
i i
* i i
-( , -r -r
\ '. i
i i
• i ' i '
,' '
,-
1
1
\
1
&
I
V
5
i
f-
i i / '
. ! />
r ——
T-",—
1 O
to ' *
I
1
j
i
t
i i
\ i
-J-'r'.-l-
d. oo i
* ' Ci 1 i '
* 51 ' ,' '
' It 1 1 /
: • : P :
: i J i!
ilk 1 1
1 ' •
1 1 ' ™ '
1 III
I M 1
-*>•
i
' *
> • ^ !
1 ' i$
1 CS
f?
» ' • !
o 10
in 1
\ ' ' ' 1
' \ / / ' / —
1 ' \ / / ' /
1, 111 1 1 1 •*
1 1 1 ' I *
1 t '
1 ' // 1
\ /' ' i; '/
• 1 1 I I n r^
- \ S7
\ /'
/
1 ^
• i ^
1 1 '
i ' '
! • .'
8 !/ -
•? 'o
J8 \
-T-TiV^-tr--^-^
oloo \i ; ; b o • ^
Oi 00 't A> o 0 §r 2
^•|040J§\ 1 l( CVJ 00 »'-O
li! ' . i \Oi \ i ' i I lii
Tl^i »» 1 l ''| "I |ir
: i ' 1 '
o o
IO ^~
' '
*
1 1 l
1 l
ill, 1 > ^
1 1 ^
« v
1 h 1 *
•I '1 1 1 1 1 IM 1 .
go
cu
~ H
o
O
•D
0
O
O
« <
V ^
~ <
h- ^ —
O H* ^
£s! o w ^
f X «H^
^-. O
_J
UJ
8 P.?
UJ
o
o 9
0 (£
UJ
a
CO
cvi
04 .
0
o
(0
o
0
0
O o
O 10
g
•H
4J
(-1
O
3
4J
CO
M
0)
t
0)
4J
§
4J
O
&
T3
0)
O
N
O
M-l
O
g
o
0)
CO
en
CO
o
4-1
•rl
4-1
H
cd
I
H
0)
t-i
00
save mi w '
286
-------
just below 100 mb. Because of the strong stratification, the prefrontal
sounding was included in this analysis. Within the stabilized layer, the
ozone is very stratified, following the adiabats, giving no indication of
transport across the tropopause.
8.2.2.4 Low Altitude Distribution of Ozone
The analysis of the five postfrontal ozonesonde data show (figure 85)
that the ozone concentration, [0_], increased almost linearly when plotted
against the logarithm of pressure, P. A least-squares regression fit to
the equation
[03] = a + b log (P/1000) P >_ 100 mb
gave the dashed line of figure 115 and the regression parameters, a and b,
and correlation coefficients of table 64.
The slope of the regression, b, changes slightly from one sounding
to another. The slope of the last sounding is two to three times the
slope of the other soundings, while the linear correlation remains high.
If the regression trend is removed from the sounding, then departures from
Table 64. Average daily total suspended particulate (TSP)
by month at DeRidder, Louisiana (1975)
3
Month Average daily TSP, pg/m
DeRidder
July 55.0
August 31.0
September 38.5
October 42.8
Overall Average 42.3
287
-------
H
CO
O
4J
•H
M-l
CO
CU
1
TJ
•H
O
co
CO
•u
cs
T3 •
il
O O
CO .-I
0)
(3 O
O 4-1
N
O O
O
M-l O
O H
CD ,-N
rH T3
•H 0)
M-l fi
O CO
M cd
P.T)
cd o
O H
•H -H
4J MH
H O
•H
PM
288
-------
the trend, [0 '], show a close relationship with stable layers in all five
soundings. In figure 116 the stable layer between the 308° K and 312° K
potential temperatures persists during the analysis r>eriod. In each
sounding, the [0 '] increases in that layer. A stable layer develqps
between the 292° K and 295° K values and [0 '] increases through it.
Obviously, persistent stable layers appear to maintain an excess of ozone.
While these stable layers do not appear to influence surface ozone concen-
tration, a mechanism for transport of ozone, near the surface is verified.
This analysis suggests that the ozonesondes probably provide a good
measure of relative change from one sounding point to another. Absolute
concentrations are harder to establish.
8.2.3 Hydrocarbons and Halocarbons
8.2.3.1 Theoretical Considerations
The major purpose in making hydrocarbon measurements for both the
gulf coast and the northern high pressure studies was to verify that
anthropogenic pollution was present in the atmosphere when samples were
collected. All samples analyzed contained acetylene and selected halocar-
bons. Since these samples were collected in a spatial and temporal manner,
the conclusion could be drawn that all the air with high ozone concentra-
tions contained anthropogenic pollution.
Difficulties are encountered in trying to relate hydrocarbons to
ozone on a one-to-one basis. The fact that alkene to alkane ratios tend
to be lower in rural air than in city air is a characteristic of a "spent"
photochemical system. The fact that there is always some alkene material
and that specific alkene/alkane ratios vary from place to place and time
to time is an indication that some new pollutant material is continually
being injected into the air.
Another indication of the continual injection of new material is that
sequential hydrocarbon samples taken along the flight path show no progres-
sion at all in the concentrations of individual hydrocarbons. In other
words each sample appears to come from a different microsystem with a dif-
ferent history of injection and atmospheric residence time.
289
-------
a
4J 3
•H 4-1
IH Ct)
M
4-1 0)
co a.
a) g
IH
M rt
<+-! -H
4-1
m C
cS H
u rt
•H I
4J j)
> 4J
bO
•rl
290
-------
There are no good figures for natural organic emissions from the areas
covered by this study, but there are estimates for primary production.
Primary production is the amount is the amount of carbon fixed by plants per
unit land area per unit time. Since, in the long run, the amount of gaseous
organic emissions is a fraction of the new plant tissue produced, a compari-
son of primary production should fairly accurately represent the relative
values of the natural organic emissions.
The primary production along the gulf shore where the study was con-
-2 -1
ducted was 1.5 to 2.0 kg m yr , with a change to 1.0 to 1.5 not far
36 /
inland.— Natural organic emissions on a yearly basis should be more
than twice the emissions at Wolf Point, and somewhat larger than the
emissions at Creston and Bradford. An absolute number cannot, however, be
assigned at this time. Aside from the difficulties in determining rates
of primary production, the nature of the plant cover might cause a differ-
ence in seasonal released of organics from the plants. The Gulf Coast
area around DeRidder is characterized by grassy land and mixed decidious
and coniferous trees.
8.2.3.2 Variation of Selected Hydrocarbons
A. Acetylene
Acetylene occurred in all ground-level samples. The averages for
Saturday and Sunday were lower than for weekdays, but since a large portion
of the United States production and commercial utilization of acetylene
occurs in this area, acetylene is not a good index to the proportion of
automobile exhaust emitted. Figure 117 shows the average acetylene concen-
tration by day of the week.
B. Propane
Propane is a hydrocarbon not found in automobile exhaust. Its presence
can, thus, be associated with industry, petrochemical processes, natural
gas, or other natural emissions sources. Before the study, it was predicted
that, due to the proximity of oil-producing and -processing installations,
the propane concentrations would be higher or in higher ratio to other com-
pounds at DeRidder than at the northern sites. This proved to be the case.
Higher values of propane were expected in this area of the country since
291
-------
4-
0»ppb(V/V)
ACETYLENE
M T W T F S S
ALL SAMPLES
12.
8-
4-
0-ppb(V/W
PROPANE
M T W T F S S
MORNING
SAMPLES
M T W T F S S
EVENING
SAMPLES
M T W T F S S
ALL SAMPLES
Figure 117. Bar graph showing acetylene and propane concentrations
by day of week at D€;Ridder, Louisiana (July-October 1975).
292
-------
there are many natural gas seeps as well as heavy concentration of refineries
and hydrocarbon users here. Figure 117 shows the average propane concentra-
tion for each day of the week. These data suggest that propane is not due
entirely to natural emissions or evaporative losses (which would be expected
to be constant 7 days per week) but is also influenced by weekday activities
unrelated to automobile exhaust.
C. Isopentane
Isopentane is one of the most abundant hydrocarbons in automobile
exhaust. The acetylene-to-isopentane ratio in auto exhaust has been re-
ported to be approximately 2. Due to the photochemical activity of isopen-
tane, the ratio in ambient air samples should be slightly higher. The
acotylene/isopentane ratio was calculated for each sample and averaged
over the month for all samples. The acetylene/isopentane ratio at the
DeRidder site for July, August, September, and October were 1.4, 2.4, 1.1,
and 1.4, respectively. Only during the month of August did the ratio
exceed 2. These data suggest that both compounds are not necessarily
related to automobile emissions (table 18, section 6.1).
D. Toluene
A gradual increase in toluene concentration was observed from July to
October. This trend does not follow the trends observed by other compounds,
such as acetylene and isopentane. The increase cannot be associated with
the increase of aromatic production, since o-xylene concentrations do not
increase during the same time period (table 18, section 6.1).
E. Halocarbons
Selected halocarbons were analyzed in grab samples collected in
DeRidder. There was considerable variance in their concentrations, which
indicates that the mean concentration of short integrated samples may not
be representative of the air mass. Halocarbon concentrations changed
rapidly by several orders of magnitude during the study period (table
18, section 6.1). Even though the DeRidder station was located in a rural
area, it is surrounded from the east to the south and southwest by large
petrochemical centers and urban areas.
293
-------
8.2.4 Particulates
8.2.4.1 Total Suspended Particulates; Sulfate, Nitrate, Ammonium
Total suspended particulate (TSP) samples were collected on a daily
basis at the DeRidder site from July through October 1975 by the 24-hour,
integrated high-volume method. Selective analyses for sulfate, nitrate,
and ammonium ions were also made. Table 65 summarizes the average daily
TSP levels by month. The highest TSP concentration occurred in July.
TSP averages by day of the week show the highest concentration on
3 3
Thursday (51.2 yg/m ); the lowest value on Saturday (35.4 yg/m ). Chemical
analysis of the particulate samples gave these average concentrations:
3 3
sulfate, 5.6 yg/m (13 percent of TSP); nitrate, 1.4 yg/m (3.3 percent of
3
TSP); and, ammonium, 0.03 yg/m (0.07 percent of TSP).
Sulfate and nitrate data from the DeRidder site were also expressed
as percent of TSP by month as follows (month, percent sulfate, percent
nitrate): July, 10.6, 3.5; August, 17.6, 3.1; September, 13.2, 3.1; and
October, 13.3, 3.7. Sulfate was also averaged by day of the week as a
percentage of TSP. The sulfate percentage of TSP was consistent during
the week.
Comparison of these percentages with data taken from the National
Air Sampling Network (table 55, section 7.3) classifies the DeRidder site
as a nonurban site intermediate in distance from sulfate and nitrate
sources.
8.2.4.2 Gulf Coast Aircraft Particulate Samples
High-volume particulate samples were collected by the aircraft on
nine of the Gulf Coast area flights. Sample collection occurred during
the entire flight. These flights are listed in table 66 with the corre-
sponding analyses for nitrate and sulfate.
The flights from DeRidder, Louisiana, which extended over the Gulf of
Mexico, were much lower in nitrate and sulfate than the corresponding
southern flights over land surfaces. The highest concentrations of nitrate
and sulfate in the gulf coast flights occurred on the flight which surveyed
the Port Arthur, Texas, area (fig. 66). The lowest concentrations
occurred on flights over the Gulf of Mexico.
294
-------
o
CO
01
60
Cfl
4-J
c
0)
CJ *
Q) f^
O-, VO
CTi
CO rH
cd I
vO
CO VO
4-1 CTl
C H
0) ^
3
4-1 CO
•H 01
4-J 4-1
CO Cd
C rH
0 3
CJ CJ
•H
CU 4-1
Cfl Cd
T— I &t
Jj
CJ 13
•H 0)
4J '•rj
rl C
Cfl 01
CO
T) 3
01 CO
4-1
O CO
CU CO
rH O
CU to
CO 60
•
m
vO
cu
rH
43
cd
H
43
to
3
|
,— s
O
rH
01
4J
O
S
01
PA
/•""S
LO
rH
s«x
0)
Cfl
•H
CU
g
QJ
4J
£
M
**
CO
e
^^^
60
3.
&*S
1
00
^
x"s
i-O
N*-'
cd
01
4J
Q
•H
^
O
M
P-i
gsjj
fO
0
60
a.
8
Q
M
}~*
x*^
CO
c
o
•H
4-1
cd
4-1
CO
p^
H
CN
v*^
&^
CO
3
60
3.
00 i-H CN rH O
f^H r*^ ^^ oo ^^ r*^ c^ C3 ^H
• ••••••••
mOCNrHOOOOO
rH
o m CM CN
in vo rH vo in CD co CN
O rHrH-sfinOrHOOO
• •••••••**
rH rHOOCNOOOOO
CN
O\ f*^ CO rH *^
^^ f*^ i-H i-H rH vO CO CD CN
• ••••••••
mocNcoooooo
rH
00 CN >3- VD
O CNCNOOCNOCNOOO
• *••••••••
o (Nooinooooo
0) 01
StiSSl'ccflcflcI) cf'o)
P<4 Cd N M O r4 MH P^ ti 60 pM 'O
co &H CO g 4-J t-H a, o fi O cfl
3 0) J-rl3OM5-HCU
•
p^
VO
rH
vO
vO
CTi
rH
A
*Jt
^
o
^J
CU
2
60
•S
H
a.
a
cd
CO
t—J
rt
rH
cd
C
O
•H
4J
Cfl
2
CO
Ol
4J
cd co
4-1 cd
co o)
K
T3 CO Cd
CU cfl
4-J cu a
•H M Cd
(3 cd 43
s c3
0 43 &
to M O
4-1 3 M
>4H
Cfl M
4J Cfl to
Cfl CU Cfl
O JZ fe
•K tfl 43
295
-------
Table 66. Nitrate and sulfate concentrations for samples
collected during gulf coast flights
Date
9/19
9/21
10/10
10/13
10/14
10/19
10/20
10/30
10/31
Overall
Average
Average
Flight Altitude
JJiight (ft)
DeRidder to Sea 1000/500
(075, 076)
DeRidder to Sea 4000/500
(078, 079)
DeRidder to Sea 500
(094, 095)
DeRidder to Sea 2000/500
(096)
Alabama 2000
(097, 098)
Texas 2000
(099, 100)
Louisiana 2000
(101, 102)
Port Arthur, Texas 2000
(109)
Texas 2000
(HO)
Average
of 2000 ft. Flights
of Sea Flights
NO 3- S04~
(yg/m3)
1.9
3.2
2.4
2.1
2.7
4.4
5.1
10.6
5.7
4.2
5.7
2.4
0.7
9.7
2.8
2.9
8.6
8.4
11.4
25.5
11.2
9.0
13.0
4.0
296
-------
8.2.5 Summary of Ozone Generation (Gulf Coast Oxidant Study)
The gulf coast area produces all the necessary precursors for tropo-
spheric ozone generation. These come from the urban areas and from the
petroleum and petrochemical industry, which is concentrated in the east
Texas and west Louisiana regions. Oil refineries are sources not only of
hydrocarbons but of NO also.
X
The cities and industries in this region represent discrete sources
and are not sprawling megalopolises nor thickly settled rural areas, and
certain aspects of the high ozone generation show this. With good atmos-
pheric ventilation the whole southeast exhibits ozone concentrations which
3
can be considered at or near natural background levels (i.e., <^ 70 yg/m ).
With less favorable conditions, a gradient of high ozone concentrations
can be observed, progressing from plumes of high concentration and
covering more and more areas until the high ozone concentrations become
virtually an air mass characteristic.
297
-------
9.0 CONCLUSIONS
The following conclusions derived from the data are listed separately
for each of the two study areas. Section numbers are provided to refer to
the section of the report that pertains to each conclusion.
A. Conclusions: Northern High Pressure Oxidant Study
3
1. In the summer, high concentrations of ozone (i.e., >_ 160 yg/m )
in the rural boundary layer and in the eastern portions of
the United States are most often found within high pressure
systems. Sustained periods of high ozone are associated with
macroscale high pressure systems that persist for more than
20 days. (Section 7.2.2)
2. Highest concentrations of ozone were found in the back side
of a high pressure system. A relative minimum is observed
in the front side or near the center. (Section 7.2.3)
3. Locations of maximum and minimum ozone concentrations in a
moving high pressure system correlate with the location of
air having maximum and minimum residence time in that
system. (Section 7.2.3)
4. The air initially in the northeastern quadrant of an east-
ward moving high pressure system has the longest residence
time in that system. (Section 7.2.4)
5. Oxides of nitrogen concentrations in rural areas in the
western section of the study area were apparently too low
to promote the generation of ozone concentration equal to
or greater than the NAAQS. (Section 7.3.1)
6. High ozone concentrations and the frequency of exceeding the
NAAQS for photochemical oxidants are associated with increased
population density (i.e., both increased from west to east).
(Section 7.3.1)
B. Conclusions: Gulf Coast Oxidant Study
1. Ozone concentrations over the Gulf of Mexico usually were
less than those over land. High ozone concentrations (i.e.,
>_ 160 yg/rn-^) that were measured over water or in air flowing
off the Gulf of Mexico were associated with air that had
previously passed over continental sources of pollution.
(Section 8.1.3)
2. Changes in the vertical structure of ozone concentrations
below 3 km are primarily controlled by boundary layer
processes. (Section 8.1.2)
299
-------
3. Elevated ozone concentrations (i.e., > 160 yg/m ) are
frequently measured in plumes downwind of potential
ground sources of precursors, i.e., cities, major refineries,
and petrochemical installations. (Section 8.1.3)
4. Upper decile concentrations of ozone are associated with
slow moving air that had passed over high precursor emis-
sion areas and arrived from a nonprevailing wind direction;
lower decile concentrations are associated with faster
moving air, having, a long over-water fetch with a weak
anticyclonic trajectory. (Sections 8.1.3 and 8.1.4)
300
-------
10.0 REFERENCES
1. Federal Register, "National Primary and Secondary Ambient Air Quality
Standards," April 30, 1971.
2. L. A. Ripperton, H. E. Jeffries, and J. J. B. Worth, "Relationship of
Measurements in Non-Urban Air to Air Pollution: Ozone and Oxides of
Nitrogen," Proceedings, Second International Clean Air Congress,
Academic Press, N. Y., pp. 386-390 (1971).
3. Research Triangle Institute, "Investigation of High Ozone Concentration
in the Vicinity of Garrett County, Maryland, and Preston County, West
Virginia," issued as Environmental Protection Agency Report No. EPA-R4-
73-019.
4. Research Triangle Institute, "Investigation of Ozone and Ozone Precursor
Concentrations at Non-Urban Locations in the Eastern United States,
Phase I," issued as Environmental Protection Agency Report No. EPA-450/3-
74-034 (May 1974).
5. Research Triangle Institute, "Investigation of Rural Oxidant Levels as
Related to Urban Hydrocarbon Control Strategies," issued as Environmental
Protection Agency Report No. EPA-450/3-75-035 (March 1975).
6. D. A. Craven and D. J. Johnson, Yellow Pine Study, Texas Air Control
Board, Technical Support Program, Air Quality Evaluation Division (1975).
7. W. N. Stasiuk and P. E. Coffey, "Rural and Urban Ozone Relationships
in New York State," J.A.P.C.A., _24:564-568 (1974).
8. California Air Resources Board, Air Quality Data, Indio, July 1972.
9. P. R. Muller, M. H. McCutchan, and H. P. Milligan, "Oxidant Air
Pollution in the Central Valley, Sierra Nevada Foothills and Mineral
King Valley of California," Atmospheric Environment, 6^:603-633 (1972).
10. C. E. Junge, "Air Chemistry and Radioactivity," N. Y. Academic Press,
pp. 37-59 (1963).
11. H. G. Richter, Special Ozone and Oxidant Measurement in the Vicinity
of Mt. Storm, West Virginia, Research Triangle Institute, Task Report,
Task No. 3, NAPCA Contract No. 70-147 (1970).
12. T. Y. Canby, "Skylab, Outpost on the Frontier of Space," National
Geographic. .146:441-493 (1974).
13. Texas Air Control Board, "Background Concentrations of Hydrocarbons in
the Atmosphere of the Northwest Gulf of Mexico, Air Quality Evaluation
(1973).
301
-------
14. E. F. Gibich, D. J. Johnson, and R. Wallis, "Ambient Air Quality Survey,"
Corpus Christi, Texas, Texas Air Control Board, Technical Support Pro-
gram, Air Quality Evaluation Division (1973).
15. D. J. Johnson, "Texas Ambient Air Quality Continuous Monitoring Network,"
Texas Air Control Board, Technical Support Program, Air Quality Evalua-
tion Division (1973).
16. R. Wallis, J. H. Price, G. K. Tannahill, and J. P. Gise, "Ozone Concen-
trations in Rural and Industrial-Urban Cities in Texas," Texas Air
Control Board, Air Quality Evaluation Division (1975).
17. C. E. Decker, T. M. Royal, and J. B. Tommerdahl, Development and Testing
of an Air Monitoring System, Final Report, Research Triangle Institute,
Research Triangle Park, North Carolina, issued as Environmental
Protection Agency Report No. EPA-650/2-74-019.
18. EPA-650/2-74-056, "Development of an Acetylene Monitor at the PPB Level,"
Final Report, Beckman Instrument Company (August 1974).
19. Federal Register, "Ambient Air Quality Standards: Reference Method for
Determination of Nitrogen Dioxide" (June 8, 1973).
20. W. D. Komhyr and T. B. Harris, "Development of an ECC Ozonesonde,"
NOAA TR ERL 200-APCL 18, Boulder, Colorado, February (1971).
21. A. J. Wagner, "Weather and Circulation of June 1975," Mon. Wea. Rev.,
103, 932-939 (1975).
22. R. E. Taubensee, "Weather and Circulation of September 1975," Mon. Wea.
Rev., 103. 1143-1148 (1975).
23. R. R. Dickson, "Weather and Circulation of August 1975," Mon. Wea. Rev.,
103, 1027-1031 (1975).
24. A. J. Wagner, "Weather and Circulation of October 1975," Mon. Wea. Rev.,
104, 107-113 (1976).
25. S. M. Brunty, W. S. Cleveland, T. E. Graedel, B. Kleiner, and J. L.
Warner, "Ozone Concentrations in New Jersey and New York: Statistical
Association with Related Variables," Science, 186. 257-258, 1974.
26. W. D. Bach, Jr., Investigation of Ozone and Ozone Precursor Concentra-
tions at Nonurban Locations in the Eastern United States, Phase II,
Meteorological Analysis, Research Triangle Institute, Research Triangle
Park, N. C., Report No. EPA-450/3-74-034a (February 1975).
27. W. S. Hering and T. R. Borden, Jr., "Ozonesonde Observations over North
America," 4, Air Force Cambridge Research Laboratories, Office of Aero-
space Research, 1967.
302
-------
28. H. E. Jeffries, An Experimental Method for Measuring the Rate of
Synthesis, Destruction and Transport of Ozone in the Lower Atmosphere,
E.S.E. Publication No. 285, Ph.D. Thesis, Department of Environmental
Science and Engineering, University of North Carolina, Chapel Hill,
North Carolina, 1971.
29. F. M. Vukovich, "Some Observations of the Variations of Ozone
Concentrations at Night in the North Carolina Piedmont Boundary
Layer," J. E. R. . ^78, 4458-4462, 1973.
30. J. L. Heffter and A. D. Taylor, "A Regional-Continental Scale Transport
Diffusion and Deposition Model, Part I, Trajectory Model," NOAA TM
ERL-ARL-50, Silver Springs, Maryland, 1975, 16 pp.
31. K. MacKenzie, Personal Communication, Health Department, Houston, Texas
32. W. S. Hering and T. R. Borden, Jr., "Ozonesonde Observations over
North America," 2, 3, Air Force Cambridge Research Laboratories
Reports AFCRL-64-30 (II, III, IV), (1964, 65).
33. E. R. Reiter, Atmospheric Transport Processes, Part 2: Chemical
Tracers, U. S. Atomic Energy Commission, Division of Technical
Information Extension, Oak Ridge, Tennessee, pp. 113-154 (1971).
34. E. R. Reiter, "Significance of Stratospheric Ozone for Ground Level
Ozone for Ground Level Ozone Concentrations." Report to Stanford
Research Institute, April 7, 1976. Prepared pursuant to EPA contract
no. 68-02-2084.
35. D. R. Davis and C. E. Dean, "Meteorological Aspects of Atmospheric
Ozone as a Potential Threat to the Forest Industry of North Florida,"
J. Environ. Qual., 1, pp. 438-441 (1972).
36. H. Leith, "Primary Production: Terrestrial Ecosystems," Human Ecology,
1:303-332 (1973).
303
-------
APPENDIX A
CALIBRATION SYSTEMS/PROCEDURES
Dynamic calibration procedures were used to calibrate all analyzers
used during the field measurement program. Monthly calibrations were per-
formed on each instrument using the procedures described below.
A-l. General
Because the five stations were at elevations above mean sea level,
adjustments to the data were necessary to reduce values to reference con-
dition of 25° C (298° K) and 760 mmHg. Adjustments to volume measurements
were made using the following equation:
298
60
273
where
V_ = volume of air at reference conditions, liters,
V = volume of air at sampling conditions, liters,
P = barometric pressure at sampling conditions, mmHg, and
t = temperature at sampling conditions, °C.
Table A-l summarizes the sampling conditions used for each site.
Table A-l. Altitude-pressure relationship for sampling sites
Altitude above
mean sea level
Station
meters
feet
Baro-
Room metric
temperature, pressure
°C mmHg
Volume of 1 liter at
reference conditions
(25°C, 760 mmHg)-
liters
Bradford, Pa.
Lewisburg, W. Va.
Creston, la.
Wolf Point, Mont.
DeRidder, La.
653
702
394
605
62
2,143
2.301
1,293
1,985
203
25
25
25
25
25
± 2°
± 2°
± 2°
± 2°
± 3°
701.9
697.7
724.2
706.1
751.3
0.92
0.92
0.95
0.93
0.99
Derived from table p. 9-4, Handbook of Air Pollution, PHS Publication No.
999-AP-44, "Barometric Pressure at Various Altitudes."
306
-------
The same adjusted volume was used each time a calibration was per-
formed. A basic step-wise procedure was employed for dynamic calibration
of all air quality analyzers 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, directly to cylin-
ders containing calibration gas.
3. Allow instrument to sample zero air (i.e., air minus the pol-
lutant of concern) for a period of time sufficient to establish
a valid zero output. Average the instrument output for zero
input concentration for at least 15 minutes. Record voltage
and/or chart readings as dictated by the type of data acqui-
sition system used.
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. Omit step 4, except
on initial setup of analyzer.
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 estab-
lish a valid instrument output for each calibration concentra-
tion, and average the instrument output for that input cali-
bration coiicentration for at least 15 minutes. Record the
voltage and/or chart reading for that calibration point and
proceed to the next higher calibration concentration and repeat
the sequence of events.
6. Return the instrument inlet line to the ambient air sampling
manifold and compute a transfer equation, which relates
307
-------
pollutant concentration input to instrument voltage output,
for each instrument.
7. Document thoroughly all parameters required to complete the
calibration record sheet designed for each analyzer.
A-2. Ozone Analyzers
A dynamic calibration system producing ozone by ultraviolet irradia-
A-l/
tion of oxygen was used to calibrate the gas phase chemiluminescent
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 approximately 1 ppm
(1960 yg/m3) 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 dia-
gram of the ozone calibration system is shown in figure A-l.
In order to obtain a reference measure of the ozone output of the
A-2/
calibration unit, the neutral-buffered KI analysis method was used
for each calibration point. A diagram of the calibration system is
shown in figure A-l. The analyzer and the bubbler train sampled simul-
taneously from the glass manifold. A diagram of the KI sampling train
and apparatus used to calibrate the sample flow rate is shown in figure
A-2.
A-3. Nitrogen Oxides Analyser
A-3/
The N0-N0_-N0 analyzer was calibrated by gas phase titration.
<£• X
The technique utilizes the rapid gas phase reaction between NO and
0_ to produce a stoichiometric quantity of NO . A schematic diagram
showing components of the parts of the calibration system is presented
in figure A-3.
308
-------
Pen-ray
lamp
Adjustable sleeve
\
Collar
C i ( LI
Quartz
W
tube
Needle
valve
I Flow controller
Silica gel
Molecular sieve
Charcoal
Cylinder
air
Manifold
Vent
Figure A-l. Ozone calibration system.
309
-------
To vacuum
pump
Manometer
Absorbers
Figure A-2. KI sampling train and apparatus for calibration
of flow rate under operational conditions.
310
-------
Calibration system
Pen-ray
lamp
Capillary
restrictor
Adjustable sleeve .,
v ,-, -I -I Mass flowmeter
Collar , 3. .
/ (0-50 cm-Ymin)
NO/N2
"(100 ppm)
[Quartz tub
eaction
chamber
.xxing
bulb Manifold
-AFlowmeter
Silica gel
Molecular sieve
Charcoal
Cylinder
air
Figure A-3. Gas phase titration system.
311
-------
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, N0~, and NO channels.
^ X
Before transporting the NO calibration gas cylinders to the field,
the NO concentration of the contained calibration gases was verified
A-3/
using the technique of Hodgeson and associates. - The procedure con-
sisted of titrating an NO concentration of 1.0 ppm with successive con-
centrations of ozone (0-0. 8 ppm) produced by an ozone generator refer-
enced 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 concentration 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 of the x-axis intercept, C', was the 0_ con-
centration equivalent to the initial diluted NO concentration. An example
of a typical gas phase titration. curve is presented in figure A-4. The
cylinder NO concentration was then calculated as follows:
c
N0 " F
NO
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 dilu-
tion of the cylinder gas. This was accomplished by metering the NO from
the cylinder through a calibrated mass flowmeter and then into the dilu-
tion system of the ozone generator. To calibrate the NO- portion of the
analyzer, a constant NO concentration of approximately 940 yg/m3 (0.5 ppm)
was produced by dilution. Ozone was added in increments from the generator.
312
-------
ex
fx
c
o
rt
M
4-1
s
a
a
o
a
0.1 0.2 0.3 0.4 05 0.6 0.7 OB 0.9 1.0
Oo Concentration (ppm) (0_ generator)
Figure A-4. Gas phase titration of NO with 0.,.
313
-------
Decrements observed on the spanned NO detector are then equivalent to the
N0~ concentration 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 re-
corded.
The NO- concentration was deduced from the decrease of the NO signal,
and a calibration curve relating NO- concentration and analyzer mV output
was constructed.
REFERENCES
A-l. J. A. Hodgeson, 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. J. A. Hodgeson, R. E. Baumgardner, B. E. Martin, and K. A. Rehme,
"Stoichiometry arid Neutral lodometric Procedure for Ozone by
Gas-Phase Titration with Nitric Oxide," Anal. Chem. 43. 1123-
1126, 1971.
314
-------
APPENDIX P
PERFORMANCE CHARACTERISTICS AND
OPERATIONAL SUMMARIES FOR INSTRUMENTS
315
-------
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 by Station.
Operational data are summarized for each analyzer by station during
this study in table B-2. For this program, operational time is divided
into the following categories:
(1) Percent operational time, and
(2) Percent downtime.
Percent operational time includes all categories or operational status
other than downtime, which includes routine maintenance, awaiting repair,
or repair. The data presented in table B-2 include percent operational
time—which is synonymous with percent valid data—percent downtime, and
number and nature of failures. Urban station ozone data provided to EPA
by state/local agencies did not always include reasons for or the nature
of instrument failures. Therefore, the data summary in table B-2 may not
reflect anything other than percent valid data reported"for each station.
316
-------
Table B-l. Instrument performance characteristics
Minimum
detectable Range
concentration Precision
(% of indicated
3 3
Instrument Parameter yg/m ppm yg/m ppm concentration)
Bendix Model 8002
Chemiluminescent Ozone 0_ 4.0 0.002 0-392 0-0.2 + 2%
Analyzer
Bendix Model 8101-B
Chemiluminescent NO, NO , NO, 9.4 0.005 0-940 0-0.5 + 2%
N0-N0x-N02 Analyzer
Perkin-Elmer _ 1 ,
Model 900 Gas Hydrocarbons 8ub-PPb
Chromatograph
317
-------
0
.3
4-1
cd
1 1
co
^
f\
CO
4J
(3
0)
1
4J
co
C
•H
o
MH
^
rl
S
rH
§
O
•H
4-1
cd
M
0)
&
CN
PQ
CU
rH
cd
H
MH
O CU
rl
0) 3
M rH
3- iH
4J Cd
cd ,
M rl Cd
2 01 TJ
cy ^* ^^
p.
o
4-1
rj
1
3
4-1
CO
M
C
O
•H
4J
cd
4-1
|
1 U
O rH
0 O
O CN
0 O
O 00
O ON
rH
v0 vO
ON ON
X
co O
0 S3
•H -H
"0 'a
0) 0)
PQ PQ
*
cd
•H
C
i-H
I
01
Pi
•d"
O
MH
T3
Cd
PQ
A
«-» *Q 0)
*t ft M
,Q cd cd
cd"
•* CN CN
in CN H
r- ~* CN
*
m co ON
CN m r^
c^ c^ c^
00 CN CM
CN ON ON
1-1
IK
CO CO O
0 O JS
XXX
0) 0) 0)
PQ PQ PQ
*
cd
•H
C
•H
00
tH
4-1
co *
0) cd
ts >
0
oo
V^ **
3 C
,£> O
CO 4-1
•H CO
S oi
01 rl
I-J CJ
I
1 CJ
0 rH
O CN
O rH
O 00
O 00
O ON
rH
''
O O
X
CO O
CO j?^|
X X
0 0
0) 0)
PQ PQ
•K
3
cd
4-1
C
A
4J
•H
O
*4-4
rH
|3
1
1 &
0 rH
O CN
O CN
N
O 00
CJ r*^*
CO ON
i-H
CO CO
CN CN
K
CO O
0 %
X X
1 1
PQ PQ
•K
S
•H
CO
•3
o
^
rl
CU
TJ
•H
«
Q
MH MH
MH MH
CN r>
CM 0
oo co
r-. ON
ON ON
rH rH
CN CN
!
co co
0 0
X X
•O -O
0 0
0) CU
PQ PQ
*
*
cd
•H
rH
m
0 *
0 *
CU O
P"l "rl
M O
•a .
LJ rn
3 3
*S *|
4J 3
4J rH
•H O
PL, cj
MH MH
MH MH
H VO
ON rH
ON St
O 00
CN f»
CN O
CO
0
CO X
O_J
•rl
I 1
*
*
0
•H
co *
0 *
S §
CD O
•H M
•> CO
0) 0)
4J -H
0) O
O 0)
PM Q
4*4 *4H
MH MH
CN O
-* O
CN
CO O
m o
ON 00
rH CO
CN ON
CO
O
X f
Is 0
CU CU
PQ fl
•K
*
cd
0 *
H ed
A!
- CO
co cd
iH JZ
P. CU
A
rl Cd
o> a
CJ 0
MH UH
MH MH
m co
vO CO
rH
m CN
CO vO
ON 00
rH rH
CN CM
rH rH
CO CO
0 O
0 0
Cg Cg
•H -rl
rH H
0 O
S rS
*
*
CO
cd *K
CU CO
E"^ cd
X
T3 H
s .
H 0
01 4-1
•O CO
0) 3
S5 <
MH MH
MH MH
m r»
CN
m co
CN CO
t-« ON
rH O
CN CO
rH rH
CO CO
O O
0 0
•H iH
rH rH
•H -H
JS r£|
*
«
co
S
0)
* H
*
co »
X o
CU 0
H- 1
0 "
0 O
4J
CO 4-1
3 rl
O O
re PM
o
0
CU
60
cd
4J rH
MH Cd
•rl CJ
M O
Q) rQ |^
rl "^
300)
rH Cfl 4J
•H P. Cd
Cd CO 4-1
MH *^ CO
O
iH H >,
H 01 ,£>
01 N
CJ T3
0) 01
M > 4-1
O -H Cd
4J CO 4-1
O CO CO
01 0)
4-1 O 4-1
Q) X O
T) 01 0
1 1 1
T3 Q) MH
0)
rH
<"*
O
P.
^"*
0 H
(U ^-*
rH -H
rO 3
o o
Q) P.-H
CO .CO rl U
0 'C 3 S
o o rH oi a
•H -H -H 4-1 -rl
4-1 4J Cd CO 0
cd cfl MH >-. O
4J 4-1 CO rl
CO CO VI 4-1
CU IS O
<-H 0 ? O 0)
Cd Cd O rH rH
M ,a p- M-I o)
3 rl
ai & i I i
* -K
.\ * cd ,0 o
318
-------
APPENDIX C
AIRBORNE PLATFORM AIR SAMPLING SYSTEM DESIGN
C-l System Description
The aircraft air-sampling system was designed to collect atmospheric
samples with a minimum of sample disturbance and then to provide a known
environment in which these samples could be measured. The system was de-
signed for sampling in'steady, level flight in a light aircraft. In addition
to the air-sampling function, the system was to provide accurate meteoro-
logical data. In particular, measurements of atmospheric pressure and
temperature, indicated and true air speed, and Mach number were of interest.
The principal objectives in designing the sampling apparatus were gas
analyzer measurements and accurate meteorological data. A secondary objec-
tive was to develop a systen. that would also be adequate for particulate
sampling.
A schematic of the overall system is shown in figure C-l. The sampling
probe was straight, except for a slight, large radius bend approximately mid-
way along its length, which was necessary to avoid the aircraft radar unit
located in the forward nose section. This bend should not have an influence
on particle sampling. The outboard section of the probe extended approxi-
mately 0.3 ro beyond the nose. Further extension was limited due to the
possibility of interference with the radar system. At this forward location
there was minimal aerodynamic disturbance and little likelihood of aircraft
contamination in the air.
The probe'was joined to a transition diffuser and stagnation manifold
on board the aircraft. This; unit decelerated the incoming air to stagnation
conditions to provide an almost quiescent environment in which to take sample
measurements. An exhaust damper was provided to adjust the exchange rate of
the air inside the system and to maintain a positive pressure relative to the
aircraft cabin to eliminate the possibility of contaminants entering from the
cabin through the manifold exhaust. Measurements of pressure, temperature, and
dew point, along with gas analyzer samples, particle analyzer samples, and
hot wire anemometer readings, were then taken from the manifold.
320
-------
I )Forward
I 1 Bulkheads
I ' .
Data System
And
Sample Probe
Sample Manifold
Sample Intake Tube
2.5 cm I.D. Teflon
Total
Tenparature Probe
t Tube
0. Analyzer
" Enhaust
Figure C-l. Functional layout of airborne sampling system.
The outboard total temperature and pitot-static probes we^e intended
to provide data on the ambient static pressure (altitude), the stagnation
pressure excess (velocity or Mach number), and the total temperature (ambient)
static temperature). Quantities in parentheses were determined by cali-
bration or calculation from the actual measurements. Air lines and
electrical leads from the outboard instruments were brought into the
cabin, where appropriate transducers were used to provide the dc signals
necessary for the onboard recorder system. The probes were mounted on
the aircraft so as to minimize aircraft aerodynamic error.
An atmospheric particulate sampler, which utilized aerodynamic
3
pressure to cause air flow at approximately 35 cfm (1m /min) through
a standard high volume glass filter, was designed and installed midway
through this program. The system required no power and was mounted ex-
ternally.
The various elements of the system illustrated above are described
in detail in the following sections with appropriate specifications and
design methodology. Sections are also included on calibration and data
reduction.
321
-------
C-2 Sampling Probe and Onboard Stagnation Manifold
The objective of the probe and manifold system described briefly
in the previous section was to provide a straight, ram- type air-sampling
capability, which samples undisturbed air ahead of the plane. The
system had particular advantages in avoiding aircraft contamination and,
at the same time, providing a. straight flow path for participate detect-
tion. A manifold was used to provide a quiescent environment for mea-
surements with an adjustable air exchange rate.
A sketch of the probe and manifold apparatus is shown in figure
C-2. The probe V»s essentially a Teflon tube mounted inside a steel
structural tube. Teflon was used to preclude extraneous chemical reac-
tipa effects and to minimize side wall particulate capture.
The manifold was constructed with an inlet diverging diffuser sec-
tion to allow a deceleration of the flow to a manageable velocity.
Design conditions were chosen to be 25° C and a nominal air speed of
90 m/sec (200 mph) . At this speed, the ambient-to-stagnation tempera-
ture ratio is
- 0.9886
o
and the Mach number is 0.24. Therefore, the velocity of the air inside
the manifold is
/kRgcT/0.9886 • M (1)
where M is the manifold Mach. num&er. The result is
m
V ' "» 1144 M (2)
mm ^ '
The final sizing of the manifold was accomplished by assuming the flow
through the apparatus is isentropic and recognizing the equality of the
flow rate in the inlet and manifold. Thus, WJ - W and
in m
A.f (M) - f (M ) * A (3)
i w w m m
322
-------
-------
where the isentropic flow function, f , is dependent only on the
Mach number, and, when defined in terms of stagnation properties, it is
fw " AP_ / g (4)
Thus,
fw (V - fw OU/CA/V (5)
where f (0.24) » 0.27438. For a given A /A. ratio, f (M ) can be calcu-
w m i w m
lated and the corresponding M read from the isentropic air tables. The
air velocity in the manifold is then calculated from M using equation 2.
m
A table of some of the calculations is shown below in terms of manifold
diameter, D .
m
D
m
102 cm
12.7 cm
15.2 cm
A /A.
m i
16/1
25/1
36/1
f (M )
w m
1.715(10)~2
1.098(10)~2
0.763(10)~2
M
m
0.0145
0.009
0.007
V
m
5.0 m/s-ec
3.1 m/sec
2.4 m/sec
It is seen that a 12.7-cm diameter manifold provides a nominal
3 m/sec manifold velocity for an unobstructed isentropic flow. The
actual system would be expected to behave in a slightly different manner,
as there are inefficiencies in the diffusion process and pressure losses
in the probe. These losses, in effect, serve to obstruct the inlet flow
and reduce the inlet Mach number, which reduces the actual velocity with
the manifold. Back pressure, imposed on the manifold from the cabin,
would also reduce the manifold velocity. A manifold with a 12.7-cm inside
diameter should provide enough theoretical velocity to accommodate a loss
due to friction and also provide a range of adjustability with an exhaust
damper. That is, a 12.7-cm manifold should provide from 1.5 to 3 m/sec,
which can then be adjusted with a damper to provide a velocity down to
zero if desired. Any ve-.locity down to 0.3 to 0.6 m/sec would probably
provide a sufficiently short residence time, although 1.5 m/sec is recommended
to insure no exhaust contamination.
324
-------
Several instruments were located in the manifold for data col-
lection. A gas analyzer probe was located nearest the entrance, but
yet at a distance sufficiently far from the entrance diffuser to allow
flow development. The gas analyzer probe was essentially a 1/4-in.
diameter suction tube which was aligned to receive the incoming air.
A hot-wire anemometer was centrally located to provide a visual indication
of the manifold velocity, an adjustment of which was then obtained by
adjusting the exit damper. The air velocity in the manifold could be main-
tained constant at varying altitudes and air speeds in this manner. A
static pressure tap, temperature thermistor, and dew point sensor were also
centrally located in the manifold to provide continuous recordings of
manifold pressure and temperature data. Dew point measurements were taken
to provide data from which ambient dew point could be determined. The1pri-
mary use of the manifold pressure and temperature data was for correcting
the volumetric gas analyzer's data to equivalent ambient conditions.
I
The location of the sensors and sampling probes on the manifold is shown in
figura C-2.
C-3 Outboard Pitot-Static Probe
The purpose of the outboard pitot-static probe was to provide con-
tinuous measurements of the ambient static pressure in the atmosphere,
altitude, indicated air speed, and Mach number. These standard aircraft
measurements have meteorological as well as air sampling applications,
and therefore it was important that they be taken and recorded as accu-
rately as possible. The pitot-static probe was used to accomplish these
measurements, becoming an integral part of the air sampling system and
thus avoiding the alternative of relying on typically less accurate
light aircraft flight instrumentation.
The location of the pitot-static probe was an extremely important con-
sideration. If improperly located, the indication of the static pressure
would be erroneous because of the aerodynamic distortions by the aircraft. A
typical pressure distribution is illustrated in figure C-3. The two loca-
325
-------
tions which are usually recommended for a pitot-static probe are mount-
ings on a nose boom and a wing tip boom. The latter was inconvenient by
the difficulty in running air lines through the wing sections. A nose
location was complicated by the difficulty in gettin gsufficiently far
out from the nose without interfering with the nose radar. An estimate
of the distance out from the nose necessary to obtain an undisturbed
measurement was made by treating the forward air flow as an imcompres-
sible flow around a half-body nose geometry. Two half-body profiles
Figure C-3. Pressure distribution around aircraft.
are shown in figure C-4, each with a different source location. The
outer profile corresponds to a source location at R from the stagnation
point and the inner profile corresponds to a source location at R . . It
can be seen that the outer profile best fits the actual aircraft profile
downstream whereas the inner profile best fits the aircraft profile near
the nose. Both, however, are blunter near the nose than the actual
skin profile. Thus, estimates based on these profiles near the nose are
probably overestimates. The equation for the pressure field around the
half-body profiles is
P(r.9)-P
(')
(6)
where P is the true static pressure and P(r,6) is the actual pressure in
the vicinity of the aircraft. With the pitot-static tube location as shown
326
-------
O
327
-------
in figure C-4, the outer profile estimate indicates an excess pressure
of 6 percent of the total pressure, and the inner profile estimate
indicates an excess pressure of 0.9 percent. The inner profile is probably
a closer estimate at this location, with a nominal systematic error of
1 to 2 percent probably a good estimate. An error of 1 percent indicated
on the static side of the pitot-static probe will indicate the true static
pressure plus 0.01 (P -- P), where the latter factor is approximately
0.5 psi at 200 mph. Thus, the overprediction is 0.005 psi which corresponds
to about 3 m of altitude at sea level. Thus, the location of the pitot-
static probe as shown should indicate altitude to within 3 to 6 m.
Any standard pitot-static tube designed to operate in the Mach
number range of 0 to 0.25 is satisfactory. An inexpensive one which has
been used successfully Is the model manufactured by Airflow/Davis In-
strument Company, Baltimore, Maryland. This instrument has a 30.5 cm
length, a 4 mm tube diameter and an ellipsoidal nose geometry.
The pitot-static probe itself was located on the air-sampling
tube beyond the nose of the aircraft. Metal air lines were then
brought back into the cabin from the "static" and "total" connec-
tions on the probe and attached along the length of the air sampling
tube. A schematic is shown in figure C-5, which illustrates the recom-
mended manner of instrumenting the pitot-static probe, along with some
of the electrical specifications.
C-4 Outboard Total Temperature Probe
The purpose of the total temperature probe was to achieve an accurate
measurement of the adiabatic stagnation temperature of the air relative
to the moving aircraft. With this measurement and a determination of the
Mach number from the pitot-static probe, the actual static or "ambient"
temperature could be determined. Manufactured total temperature probes are
normally designed for high subsonic civilian applications as well as
supersonic military applications. These commercial probes, such as that
manufactured by the Roseniount Company, are sumewRat overdesigned for the
328
-------
>>
o
cd
o
o
cu
id
en
o o
o
CO
e
cu
en
cui
o
co
N o
cd
'O CJ
£3
r, W
U PM
CU
4J B«
O IT)
CO CM
C5 •
CU O
cn
^-^ *
TJ
H -H
cd CO
CU
>4-l
14-4
co
>H
O
<•
-s
CO
4-1
C
3
CO
cfl
OJ
VI
4J
td
vi
0)
•o
3
CO
co
a)
)-i
P.
4J
co
a
co
SP
CJ
O
t-l
I
CJ
(1)
p
§>
•H
P-i
329
-------
low subsonic application here and excessively priced for the need. There-
fore, a probe was designed to provide the information required and
was specifically intended for low subsonic and steady-state, level-flight
measurements.
A drawing of the total temperature probe is shown in figure C-6. It
was designed as a small diffuser and was constructed from nylon material,
which has a low thermal conductivity. It was taped externally with a
silver reflecting tape to minimize solar heating. With this construction,
it is not thought that the air compression will deviate significantly from
an adiabatic process. Small holes were drilled in the rear of the diffuser
to allow internal air movement and a ventilation sufficient to minimize
extraneous heating effects on the sensed temperature. The probe was not
designed for rapid measurements and was allowed to come to temperature
equilibrium prior to each measurement. Using standard heat transfer
computational methods, it was found that the above probe should require
approximately 1.5 minutes to allow a dissipation of 95 percent of the
heat energy contained in the nylon and to bring it to equilibrium with
the new conditions. This time allowance for the probe adjustment should
preclude any possibility of error due to thermal inertial delay.
The sizing of the probe was accomplished in a manner similar to that
used in the sizing of the air-sampling manifold. The inlet area was somewhat
arbitrary and was chosen small to eliminate excessive angle of attach sensi-
tivity but sufficiently large to allow adequate internal ventilation. A
diffuser area ratio of 25 to 30 to 1 was adequate to stagnate the incoming
air, and an outlet area of 1/3 of the inlet area allowed adequate ventila-
tion around the thermistor head. Three small holes were provided at the
rear for this purpose.
330
-------
•o
rt
JS
I/1
II ill II
ii it n
H ij n
,--
1
i
1
//
/ /
1 1
•
^
~i> o.
o n) - >w
<: Q
'c
o
rH
53
~M
1 .
1 1
1 1
1 1
0
4J
CO
0)
U
4J
}H
-------
The specifications for the probe are shown on the drawing, some of
which were modified for convenience in mounting, etc. The location of
the probe on the aircraft was riot important as long as it was not near the
engines and located outside of the skin boundary layer. The location near
the forward end of the sampling tube was ideal.
The probe was designed to be used with a thermistor sensor, an instru-
ment which has a relatively slow response time and was therefore very com-
patible with the probe itself in terms of responsiveness.
C-5 Airborne High-Volume Sampler
An atmospheric partic.ulate sampler which utilizes the High Volume Air
Sampler Method was designed and used in an airborne application in conjunc-
tion with the air sampling system described previously. The sampler was
used primarily in the collection of airborne sulfates and nitrates and
utilizes the standard glass-fiber filter. The unit was mounted on the for-
ward side of the aircraft nose section, placing it well in front of engine
and prop contamination with a remotely operated flow control apparatus.
Several important considerations played a role in the design configura-
tion for the sampler. A summary of these are given below.
First, in air sampling, the air must be slowed relative to the aircraft
in such a way as to minimize the disturbance of the actual particulate
concentration and distribution in the air. That is, the sample that passes
through the high volume fLiter must be as representative of the condition of
the quiescent air as possible.
It was necessary to Bake use of aerodynamic pressures to provide the
flow through the filter rather than using powered devices to create the sam-
ple flow due to the large power requirements of a large air pump. In order
to achieve the maximum flow through the filter possible by aerodynamic
pressure, the diffuser action within the sampler which decelerates the air
must be as efficient as possible in order to avoid pressure losses. Also,
the nozzle action around the exterior and rearward sides of the unit should
be preserved in order to provide the maximum rearward suction.
The volume rate of sampling must be measured continuously and a remote
control apparatus designed to control the flow rate, including a shut-off
332
-------
during take-off, landing, etc. The quantity of flow through the filter
must be sufficiently great to allow efficient filtration and at the same
time sufficiently rapid to allow a large enough volume to be sampled in
the available flight time.
The flow rate versus pressure drop characteristics for the clean
glass-fiber filter material was estimated by assuming the pressure drop
across the standard high volume filter calibration orifice and eighteen-
hole plate assembly was approximately that which would occur across the
clean filter material itself. This is the basis of the flow meter calibra-
tions which are used with the standard units. The CFM per unit of filter
area was then plotted versus pressure drop for the standard orifice. Using
orifice formulae, a similar curve was plotted for the eighteen-hole plate.
For any flow rate, then, the total pressure drop is the sum of that which
occurs across the orifice and that which occurs across the plate. This
sum should be approximately equal to the pressure drop across a clean filter
with the same CFM per unit of its area. This plot is shown in Figure C-7
along with the CFM for the standard filter. It can be seen that-a
pressure drop of approximately 28 in. (71 cm) of water is required to produce
the nominal 60 CFM through the standard filter in a clean state.
The aircraft motion will produce an aerodynamic pressure drop which
can be used to drive the flow through the filter. This pressure drop will
be approximately constant at constant air speed and should not vary
appreciably with changing conditions within the filter. Thus, a prediction
of this pressure drop should yield, using the previous flow rate curve, an
estimate of the flow rate per unit of clean filter area. This predicted
flow rate for the clean condition, then, is used to decide the sufficiency
of the flow generated in this manner. It is assumed at the outset that
heavy accumulation on the filter will not occur above ground level and
that, basically, trace quantities are sought. Thus, a clean filter calcu-
lation should provide a good criteria.
At an air speed of 200 mph and an altitude of 1,000 feet (205 m), the
total ram pressure is approximately 19 in. (48 cm) of water. Approximately
2.5 cm of this will probably be lost at the entrance and about 1.25 cm
left as dynamic pressure at the filter. Thus, the raw pressure which
333
-------
13A31
IV 9NI99013 A9
3i\/H MOIJ
VO
co
co
CM
co
O
co
00
CM
If)
CM
CM
CM
CM
O
CM
00
10
CM
O
00
CM
1 1
V3W a311U zNl/ldD ' J. S
0 0
I 1
^J- Lf>
cn co
o
CO
1
£
o
CM
o
CM
<
CO
CO
p-
-------
might be available at the filter is approximately 17.5 in. (44.5 cm) of
water. Estimating a suction pressure of 40% of the freestream dynamic
pressure [i.e., approximately 7.5 in. (19 cm) of water] produces a total
pressure drop available to the filter of approximately 25 in. (63.5 cm)
of water. From the filter calibration curve, this corresponds to approxi-
2
mately 0.9 CFM/in or 58 CM for the standard filter. Even though this
pressure drop will be approximately constant for constant flight conditions,
the flow rate will decrease some as the filter becomes clogged. These
figures are all shown on the filter calibration curve. It is readily
seen that the operating characteristics of the entire unit will be greatly
dependent upon the aerodynamics of the design. That is, entrance losses
should be minimized, internal diffuser losses must be minimized and, just
as important, the external aerodynamics must not be interfered with in order
to preserve the rearward suction pressure. An estimated operating range
of flow rate is shown on the figure to account for uncertainties in design.
The unit should operate around 30-35 CFM for a 7-in, (17.8 cm) diameter
filter.
The diverging diffuser arrangement was used because it offers least
disturbance to the air in deceleration. That is, the air stream does not
impinge on any surface other than the filter itself and therefore suffers
no turning difficulties with the associated particulate losses. The forward
opening is just sufficient to admit that amount of air which can pass through
the filter at the operating pressure drop. In order to make use of the full
deceleration in the diverging section and avoid unnecessary pressure loss,
assurance had to be made that flow separation did not occur. This was avoided
by specifying a 5 cone angle, a criteria within the separation angle. The
sizing of the inlet area was accomplished using standard gas dynamic calcu-
lations and tables of the properties of air. The specifications which re-
sulted are shown on figure C-8. Note that suction at the rear of the sampler
is accomplished by allowing the external flow to accelerate uninterrupted
around the sampler.
An additional advantage which the above design offered was the rela-
tively easy means by which the volume flow rate could be measured. A
loss coefficient of 0.95 for this diverging apparatus is a relatively well
accepted figure. Thus, the flow rate can be calculated from Bernoulli's
equation if the pressure rise from the inlet to the filter is monitored.
335
-------
•H
T)
in
•
CM
3
!*
iH
p4
i
CO
u
0)
4-1
4J
e
0)
o
CO
0)
u
a
4-1
u
i
o
u
o
H
X
a1
CO
cu
•a
CO
o
(1)
M
336
-------
Thus, pressure taps were located at the inlet and just upstream from
the filter. Three taps were equally spaced around the periphery at each
of these two locations to provide the peripherally averaged static pressure.
2
At the design flow rate of approximately 0.75 CFM/in of filter area, the
pressure rise should be approximately 1-9 cm of water.
C-6 Calibration of System and Components ,
?
The most important developmental aspect of the entire system is the
calibration of both components and the assembled system under flight condi-
i
tions. Each of these is discussed below.
This measurement relies on two devices: the static side of the pitot-
static probe and a strain gage absolute pressure transducer. Each must be
checked for accuracy. A pressure transducer can be calibrated with a dead
weight tester, a water column, or other device. Even with the transducer
working properly, there is no assurance that the pitot-static probe is
aligned properly, etc. Thus, after the transducer itself is calibrated, the
probe was checked in flight to make sure that a correct pressure was being
recorded. This was done by flying down the runway a known height above
the ground and checking the static pressure reading (altitude). A 0 - 15
psia transducer which is accurate to +0.1% of F.S. should be able to resolve
at least +9.1 meters of elevation.
The differential pressure transducer, with a range of +2 psid and an
accuracy of +0.25% F.S., can be easily calibrated by manometer as a check
on the manufacturer's calibration. An inflight calibration, however, is
somewhat more complicated. Several timed, low level flights over a straight
measured distance were conducted. A day was chosen with little or no air
movement. The ground speed was then measured and used as the indicated air
speed. If the velocity is low, it can be used directly in calculating the
total pressure from
p0 =p + V^;
where P is assumed at the test altitude, p is the density of the ambient air
at that altitude, and V is the velocity. Several runs were made at varying
337
-------
speeds from as low as possible up to nominal cruise speed. A plot of
V
(P -P)/pr— versus V was developed and Cp determined as a calibration
tion coefficient of pitot-Sitatie output versus indicated air speed. This
indicated air speed will coincide with the true air speed as long as the
above plot is flat with V. A V dependence indicates compressibility effects
are present and should occur in the neighborhood of 150 mph, raising the
desirability of a determination of "true" air speed in the range of 150 mph
to 200 mph.
The type of pitot-sta1:ic probe used actually minimizes the need for
calibration since it has the modern ellipsoidal nose, with a C normally
accepted to be within 1% oE unity. The error is primarily introduced by
misalignment, a characteristic which the ellipsoidal nose pitot-static probe
is relatively insensitive to. The particular probe used here was checked
and found to be stable with angles greater than +30°; thus this is not
thought to be a problem. Therefore, in the above calibrations, runs in
opposite directions were made to nullify wind effects and the data plotted
and compared to unity. If a spread exists about unity, then this value should
be accepted as the correct value for C . This type of calibration experiment
primarily serves to corroborate and add credibility to a unit C but is not
sufficiently controlled to provide data to alter C , even if the data indicates
some alteration. Thus, in the following sections, C is taken as the commonly
accepted 1.0 + 1%.
The total temperature, measured with the total temperature probe, was
calibrated in the same series of runs as those used for the total pressure
calibration. Prior to the runs, a static .temperature calibration was made
by inserting the temperature probe into a water bath and checking against
a calibrated thermometer.
C-7 Data Reduction Formulas
The following formulations can be used in reducing the transducer
and thermistor outputs from the air sampling system:
Altitude - The 0.1% F.S., 0-15 psia pressure transducer should
provide a + 9.1 meters resjolution of altitude at sea level and should remain
338
-------
SYMBOLS
2
A - Flow area, ft ,
A14. - Altitude, ft,
it 2
A. - Sampling tube flow area, ft ,
2
A - Manifold flow area, ft ,
C - Ambient concentration in mass/vol.,
C - Manifold concentration in mass/vol.,
Cp - Total pressure calibration coefficient, approx. unity,
Cj - Total temperature calibration coefficient,
E - Ground elevation above mean sea level,
f - Isentropic flow function, equ. (4),
g - Gravitational constant, 32.2 (lbm-ft)/(lbf-sec ),
k - Specific heat ratio, =1.4,
K - Tabular conversion of psia to feet of altitude,
Kp - ^^
M - Mainstream Mach number, Vtr//2gckRT,
M - Manifold Mach number,
P - Calibrated ambient static pressure output, psia,
(PQ-P) - Pitot-static probe output, psia,
PQ - Total pressure (= P+(P -P), assuming Cp = 1),
P - Manifold pressure, psia,
P , - Ambient static pressure at mean sea level, psia,
P(r,9) - Ambient static pressure field near aircraft,
R - Gas constant, 53.35 (ft-lbf)/(lbrn-°R),
R - Nose radius, ft, as shown in Fig. 4,
r - Probe location, as shown in Fig. 4,
T - Ambient temperature, °R,
T - Manifold temperature, °R,
T - Stagnation temperature, °R,
Trin - Dew point temperature, °F,
Manifold dew point temperature, F,
dp
Tdp,m
V - Indicated air speed, ft/sec,
Vt - True air speed, ft/sec,
V - Manifold velocity, ft/sec,
•
W - Mass flow rate, Ib^sec
P - Ambient air density, 0.075 Ib^/ft3 at msl and 70°F,
339
-------
approximately this at most altitudes. The pitot-static output at mean
sea level should be
Pmgl = 14.696 (10)
where P - is the output measured at mean sea level. Thus, the altitude
msl '
relative to sea level is
P) (11)
where K is a tabular conversion to feet of altitude and P is the actual
static pressure output. The local ground elevation should be obtainable
by taking a P reading on the ground. That is,
(12)
Dew Point Correction - Because of the compression in the mani-
fold, the indicated dew point temperature must be corrected to ambient
conditions. This is done using the Clapeyron equation for the vapor pressure
curve for
grates to
curve for water vapor. In the neighborhood of 70 F, this expression inte-
7 / ^ /PnA
T. - TJ ") = T. VI. 05x10" / In (-^ (13)
dp,m dp/ dp,m x ' \P /
where T, is the measured manifold dew point temperature and T, is
dp,m dp
the dew point corrected to ambient conditions. The dew point correction
should not be greater than approximately 2°F.
Indicated Air Spee;d - The indicated air speed is developed
from Bernoulli's equation
Po - P = pV2/2gc (14)
or
g (P - P)
V, = K^-^ (15)
where p is the ambient density, P the total pressure, P the ambient
static pressure, and K^ the calibration coefficient which should be
close to unity, within 1%.
340
-------
True Air Speed and Mach Number - The Mach number is computed
from the relation
k-1
(16)
where AP is the P - P above and K_, is approximately unity. The true
O r
air speed then becomes
(17)
where T is the ambient static temperature determined by the total tempera
ture probe.
341
-------
APPENDIX D
OZDNE AND OXIDES OF NITROGEN ANALYZER EVALUATION
AT REDUCED PRESSURE
343
-------
APPENDIX D
OZONE AND OXIDES OF NITROGEN ANALYZER CALIBRATION
AT REDUCED PRESSURE
D-l Introduction and Summary
A series of tests were conducted in an altitude chamber at the
EPA Environmental Monitoring; & Support Laboratory at Las Vegas, Nevada.
The purpose of these tests was to determine the characteristic behavior
of a gas phase chemiluminesc.ent ozone analyzer, Bendix Model 8002, and a gas
phase chemiluminescent oxides of nitrogen analyzer, Bendix Model 8101-B, to
changing altitude, as in an unpressurized aircraft.
The instruments were placed in the chamber and calibrated at ambient
pressure. The chamber was then sealed and partially evacuated simulating
a higher altitude environment. Calibration gases were generated externally
from a calibration system operating at a constant ambient pressure, and
drawn into the chamber to the gas analyzer inputs, to insure constant cali-
bration levels with changing pressure. Tests were run over the range of
pressures corresponding to ground level to an altitude of approximately 7,620
meters (25,000 feet) for the ozone analyzer and 5,790 meters (19,000 feet) for
the oxides of nitrogen analyzer. Test results demonstrated that all instru-
ments responded in a repeatable manner to variations in pressure. From these
data, graphs were constructed whereby a single correction factor could be
determined for each instrument at any given altitude over the test range.
The effects of altitude on the instrument then could be compensated for by
multiplying with the proper correction factor.
D-2 Test Set-up
The GO and N0-N09-N0 analyzers were mounted inside a Weber Pressure
,3 ^ X
Chamber. A schematic diagram of the chamber is shown in figure D-l, illus-
trating the chamber controls, support gases and signal lines to the analyzers
and the sample manifold system. The glass sample manifold was mounted inside
the chamber, with the downstream end exhausted into the chamber. Teflon
tubing was used to connect the sample inlet of each instrument to the manifold.
344
-------
LEVEL MONITOR
0., ANALYZER
SPAN
CONCENTRATION
1
FROM CAL.
SYSTEM
IN ST.
SIGNAL
OUTPUTS
CAPILLARY
RESTRICTION
BY- PASS
•<2
-
THERMISTOR
TEMPERATURE
PROBE
MANIFOLD
ALTIMETER
NO ANALYZER
x
ANALYZER
C2H4
CHAMBER
\
TEFLON FILTER
CHAMBER
CONTROL
I
110 VAC
TO CHAMBER
VACUUM SYSTEM
Figure D-l. Schematic diagram of test setup in pressure chamber,
345
-------
Oxygen and ethylene gases were conveyed to the instruments inside the
chamber, through s.s. tubing passsed through bulkhead Swagelok connectors.
The ethylene exhausted from the ozone analyzer was vented into the chamber
after being passed through a catalytic converter.
A descriptive diagram of the calibration system is shown in figure
D-2. The apparatus consists of a clean air source, compressed cylinder
of nitric oxide in nitrogen, regulating valves, flow meters, an ozone
generator, and mixing flasks.
D-3 Calibration Gases
Ozone concentrations in yg/m^ for various ozone generator sleeve
settings were determined using the Neutral-Buffered Potassium Iodide Method
described in the April 30, 1971 Edition of the Federal Register.
Nitric oxide span concentrations were obtained by diluting pressurized
NO in nitrogen with zero air. Nitrogen dioxide span concentrations were
generated by the gas phase titration technique.
The NO concentration supplied by the manufacturer for the pressurized
cylinder of NO used in this test was referenced to equivalent ozone concen-
trations added during titra.tion.
D-4 Test Procedure
The units under tests were calibrated several times during the study
at ambient conditions of temperature and pressure (25°C and 708 mm Hg).
These calibrations were conducted by introducing the zero air and span
concentrations directly to the sample manifold through the removable by-pass
shown in figure D-l. The pressure differential between the manifold and
ambient conditions during these tests was determined to be less than 0.6 cm
of water. Subsequent calibrations were conducted with the chamber controlled
at 25°C and at various simulated altitudes above ambient. During the reduced
pressure tests, the sample manifold was maintained at chamber pressure by
placing a capillary restriction upstream from the manifold and leaving the
downstream end of the manifold open ended into the chamber (refer to fig. D-3)
346
-------
PEN RAY LAMP
LAMP
TRANSFORMER
P .
* REGULATED 110
L^ 1
VAC
ADJUSTABLE
SLEEVE
CAPILLARY
RESTRICTION
1
1
H
i
WATER
MANOMETER
*
QUA
iRTZ
1 II
TUBE I
*^z~
FLOWMETER
SILICA GEL
MOLECULAR
SIEVE AND
CHARCOAL
VENT
CAPILLARY
RESTRICTION
TO MANIFOLD
Figure D-,2. Ozone generator/gas phase titration system
for calibration of 0, and NO-NO analyzers.
J J\
347
-------
CALIBRATION 5 1/mln» 1
UNIT
» 3.5 1/min
1.5 meters of 3.2 mm o.d.
Teflon
u*vi — • ^ i.j i/ mo.it
Vent and 0^ Level
Monitor Point
11 0.6 cm H20
Differential
from Ambient \
Chamt
v
u u
1 1
To Instruments
X
?er
_ tif»
Differential
from Ambient
Figure IK3. Typical pressure and flow conditions for
chamber setting at simulated altitude of
3048 meters (10k feet).
348
-------
The sample was drawn into the chamber because the chamber was at a
vacuum relative to the calibration system which was at ambient. The
length of the Teflon capillary was adjusted so that the flow drawn into
the chamber was 2.5 to 4 &/min. The remainder of the calibrating gas,
> l&/min, was vented into the room through an unrestricting vent to main-
tain a constant ambient pressure within the calibration system. This,
in effect, provided for a known concentration of the calibration gas (0~,
NO, or N0~ in this case) to be made available to the instrument at the
pressure of interest based on measurements or calibration procedures con-
ducted at ambient pressure. A second 0^ analyzer was used to monitor the
ozone level at the room vent port, thus providing a second check on the
concentration level of the calibration gas. Figure D-3 illustrates the
typical pressure and flow conditions during a reduced pressure test with
the chamber controlled at a simulated altitude of 3,048 meters (10,000 feet).
A standard aircraft altimeter readable to tens of feet was used in the
chamber to establish the altitude/pressure points at which data was taken.
This same altimeter was used on the flights as well.
It was observed that the instruments adjusted much more rapidly to
changes in pressure than did the entire system (instrument and calibration
system) to changes in concentration. Therefore, these tests consisted of
setting the system up for a single calibration gas level input and varying
the pressure stepwise over the range being tested. Runs were repeated using
the same input concentration levels on different days to determine repeat-
ability. Also, different input concentrations were used to check linearity
of the analyzers over the pressure range of interest.
D-5 Results of Tests
The response of 0- and NO analyzers to different concentration of cali-
bration gas and the respective altitude range were determined. These data
were then normalized to an initial reading of unity at ambient pressure
conditions in the laboratory. A best fit was then made for the respective
sets of data, and these altitude correction curves are given in figures
D-4 and D-5 for the 0~ and NO analyzers, respectively.
349
-------
l.Or
w
o
M
o
g -5
.4
.3
1,000 5,000 10,000 15,000
ALTITUDE (FEET ABOVE MSL)
20,000 25,000
Figure D-4. Normalized response versus altitude for Bendix ozone analyzer.
350
-------
i.o r
w
CO
§
PH
CO
O
ss
w
IS]
O
25
1,000 5,000 10,000 15,000
ALTITUDE (FEET ABOVE MSL)
20,000
25,000
Figure D-5. Normalized response versus altitude for Bendix oxides of
nitrogen analyzer.
351
-------
APPENDIX E
The following flight logs, summary sheets, and selected flights
are included:
E-l OTHER
E-la Flight Track Symbols & Designators
E-lb List of Weather Symbols
E-2 LOGS AND SUMMARY SHEETS
E-2a Grab Sample (Hydrocarbon) Log (Gulf Coast Area)
E-2b Grab Sample (Hydrocarbon) Log (Northern Route)
E-2c Selective Filter (Acetylene) Log (Gulf Coast Area)
E-2d Selective Filter (Acetylene) Log (Northern Route)
E-2e Sulfate Filter Log (Gulf Coast)
E-2f Sulfate Filter Log (Northern Route)
E-2g Summary Low Pass Data (Gulf Coast Area)
E-3 SULFATE SAMPLER DATA—TRACKS AND DATA
E-4 SELECTED FLIGHTS-TRACKS AND DATA
E-4a Northern High Pressure System Survey (Flights 021 to 031)
E-4b Down Wind Plume Flight No. 005 (Gulf Coast Area)
E-4c Down Wind Plume Flight No. 006 (Gulf Coast Area)
E-4d Sea-breeze Flight No. 075 (Gulf Coast Area)
E-4e Sea-breeze Flight No. 076 (Gulf Coast Area)
E-4f Box Pattern Flight No. 099-100 (Gulf Coast Area)
E-4g Box Pattern Flight No. 109 (Gulf Coast Area)
E-4h Box Pattern Flight No. 110 (Gulf Coast Area)
354
-------
Table E-la. Flight Track Symbols and Designators
Description
Navigation Reference
©
Landings & Takeoffs
Vertical Profiles
Selective Filters
o
Grab Samples
Table E-lb. List of Weather Symbols
Symbol
0, C
*
0. o
i.-f
CD
H
Description
Clear 0/10
Scattered 1/10 to 5/10
Broken 6/10 to 9/10
Overcast 10/10
Thunderstorm
Haze
EXAMPLE:
Overcas1\ Variable, Broke?
1400 ft 25,000 ft.
10 mi. visibility
Ha.ze
356
-------
LOGS AND SUMMARY SHEETS
E-2a Grab Sample (Hydrocarbon) Log (Gulf Coast Area)
357
-------
Table E-2a. Grab sample (hydrocarbon) log
(gulf coast area)
DATE
(1975)
6/30
II
II
II
8/6
II
II
II
8/7
II
II
II
8/9
II
II
8/10
II
9/19
II
II
11
II
9/19
II
II
II
9/20
II
II
II
II
II
9/21
II
H
II
10/10
II
It
11
n
H
n
10/13
II
II
II
II
II
FLIGHT
NO.
010
II
n
on
035
n
n
n
037
II
II
II
040
II
II
041
II
075
II
II
II
II
076
n
II
n
077
078
II
II
H
It
079
II
II
II
094
II
II
II
095
II
II
096
SAMPLE
NO.
0
1
2
1
1
2
3
4
1
2
3
4
1
2
3
1
2
1
2
3
4
5
1
2
3
4
1
1
2
3
4
5
1
2
3
4
1
2
3
4
1
2
3
1
2
3
4
5
6
COLLECTION TIME
START
1100
1026
1103
1529
1451
1512
1527
1544
1340
1358
1415
1432
1905
2013
2055
1906
2014
1530
1606
1638
1717
1735
2045
2115
2153
2222
1545
1508
1540
1632
1702
1736
1950
2030
2130
2225
1716
1751
1816
1847
2122
2150
2415
1710
1743
1814
1843
1912
1946
(GMT)
STOP
1102
1028
1106
1533
1453
1514
1529
1546
1342
1400
1417
1434
1907
2015
2057
1908
2016
1532
1608
1640
1719
1737
2047
2117
2155
2224
1547
1510
1542
1634
1704
1738
1952
2032
2132
2227
1718
1753
1818
1849
2124
2152
2417
1712
1745
1816
1845
1914
1948
358
-------
Table E-2a. Grab sample (hydrocarbon) log
(gulf coast area) (con.)
DATE
(1975)
10/14
II
II
II
II
II
II
II
II
10/19
II
II
II
II
II
II
II
10/20
H
II
II
II
II
II
It
10/21
II
"
It
10/30
II
II
11
II
H
10/31
II
II
II
tl
II
II
II
II
11/1
II
II
II
H
II
II
II
II
RIGHT
NO.
097
II
It
H
M
098
II
II
If
099
H
11
u
It
100
H
II
101
II
H
II
102
It
11
II
103
II
It
II
109
H
II
II
II
II
110
II
II
II
II
II
H
II
II
111
H
tt
II
II
II
II
It
II
SAMPLE
NO.
1
2
3
4
5
1
2
3
4
1
2
3
4
5
1
2
3
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
5
6
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
B
9
COLLECTION TIME
START
1632
1658
1734
1806
1835
2102
2133
2200
2227
1619
1644
1717
1745
1828
1952
2035
2120
1616
1644
1715
1744
1905
1935
2008
2040
2118
2153
2216
2352
1945
2013
2047
2112
2148
2216
1858
1930
2001
2029
2058
2127
2159
2300
2330
1348
1415
1443
1514
1546
1614
1650
1842
1915
(GMT)
STOP
1634
1660
1736
1808
1837
2104
2135
2202
2229
1621
1646
1719
1747
1830
1954
2037
2122
1618
1646
1717
1746
1907
1937
2010
2042
2120
2155
2218
2354
1947
2015
2049
2114
2150
2218
1860
1932
2003
2031
2100
2129
2201
2302
2332
1350
1417
1445
1516
1548
1616
1652
1844
1917
359
-------
Table E-2b. Grab sample (hydrocarbon) log
(northern route flights)
,L
DATE
(1975)
7/9
ii
7/10
II
7/11
II
7/12
7/13
It
7/24
It
7/25
II
7/26
H
7/27
II
9/5
II
II
II
It
tt
II
9/6
II
II
tl
II
II
II
9/7
II
II
It
II
H
II
II
9/8
H
II
"
II
II
II
1 1
FLIGHT
NO.
014
ii
015
II
016
II
017
018
II
025
"
027
II
028
It
029
030
058
II
t|
059
II
II
It
060
II
II
061
II
062
II
063
II
tl
II
064
II
II
It
065
II
II
066
H
ii
ii
=======
SAMPLE
NO.
1
2
1
2
1
2
1
1
2
1
2
1
2
1
2
1
1
1
2
3
1
2
3
4
1
2
3
1
2
1
2
1
2
3
4
1
2
3
4
1
2
3
1
2
3
4
T '
COLLECTION TIME
START
^ •••^
1142
1932
1303
1458
1233
1415
0934
1116
1402
1719
2149
1708
1924
1643
1910
1550
1918
1713
1745
1811
1942
2053
2116
2148
1733
1810
1844
2031
2102
2255
2323
1647
1720
1743
1823
2113
2144
2216
2242
1506
1530
1616
1745
1817
1848
1920
1
(GMT)
STOP
""•"^ *^
1145
1935
1306
1501
1236
1418
0937
1118
1404
1722
2151
1710
1926
1645
1912
1552
1920
1715
1747
1813
1944
2055
2118
2150
1735
1812
1846
2033
2104
2257
2325
1649
1722
1745
1825
2115
2146
2218
2244
1508
1532
1618
1747
1819
1850
1921
361
-------
Table E-2b. Grab sample (hydrocarbon) log
(northern route flights) (con.)
=«=— ==
DATE
(1975)
9/12
II
11
H
H
n
9/27
II
H
II
n
n
11
u
9/28
U
"
U
n
9/29
II
tl
H
9/30
n
H
"
H
H
aaaBpas
FLIGHT
NO.
069
II
070
H
H
II
084
II
II
H
n
085
II
H
086
tt
tl
H
087
088
089
u
H
090
N
II
091
n
N
»
SAMPLE
NO.
1
2
1
2
3
4
1
2
3
4
5
1
2
3
1
2
3
4
1
1
1
2
3
1
2
3
1
2
3
•=••—;
COLLECTION TIME
START
1605
1634
1739
1848
1925
1945
1558
1632
1705
1740
1817
2030
2159
2245
1435
1502
1535
1628
1748
1546
1718
1748
1859
1512
1715
1835
1906
1940
i •
(GMT)
STOP
1607
1636
1741
1850
1927
1947
1600
1634
1707
1742
1819
2032
2161
2247
1437
1504
1537
1630
1750
1548
1720
1750
1901
1514
1622
>717
1837,
1908
1942
362
-------
Table E-2c. Selective filter (acetylene) log
(gulf coast area)
DATE
(1975)
8/11
"
II
II
II
9/19
II
II
"
II
11
"
II
II
9/20
9/21
II
It
II
II
II
II
II
11
tl
10/9
II
II
It
II
II
"
it
10/10
"
II
"
II
11
II
II
"
II
10/13
It
11
II
tl
II
10/H
II
II
tt
II
II
FLIGHT
NO.
041
II
II
II
II
075
II
II
II
11
076
H
ii
11
077
078
II
II
II
II
079
"
ii
H
II
092
II
II
II
093
II
II
II
094
II
II
II
II
095
II
11
II
II
096
II
(I
"
II
II
097
II
It
II
II
II
FILTER
NO.
1-18
2-31
3-42
4-58
5-61
1-70
2-0
3-134
4-79
5-50
1-53
2-145
3-82
4-118
1-40
1-10
2-36
3-40
4-113
5-24
1-55
2-108
3-132
II
4-142
1-14
2-59
3-34
4-01
1-66
2-150
3-30
4-87
1-13
2-44
3-74
4-91
5-75
1-11
2-18
3-49
II
4-106
1-52
2-94
3-81
4-33
5-78
6-1
1-60
2-103
3-14
4-110
5-113
6-68
SAMPLE COLLECTION
TIME (GMT)
FROM TO
1837
1915
1932
2005
2047
1511
1541
1624
1657
1730
2025
2100
2135
2209
1530
1450
1527
1602
1634
1705
1924
1959
2117
2157
2215
1906
1944
2025
2057
2159
2238
2314
2352
1652
1730
1806
1836
1906
2109
2140
2308
2356
2412
1648
1730
1759
1831
1859
1934
1617
1655
1724
1755
1825
1858
1855
1925
2002
2045
2153
1538
1615
1655
1727
1756
2055
2131
2204
2259
1602
1522
1557
1631
1702
1736
1955
2035
2135
2210
2240
1940
2019
2054
2127
2232
2309
2346
2420
1726
1302
1834
1902
1938
2136
2212
2328
2408
2438
1724
1757
1826
1557
1931
1954
1649
1722
1753
1820
1855
1931
TOTAL
SAMPLE
TIME
(MIN)
18
10
30
40
16
27
34
31
30
26
30
31
29
50
32
32
30
29
32
31
31
36
18
13
25
34
35
29
30
33
31
32
28
34
32
28
26
32
25
32
32
II
26
26
27
27
26
32
20
32
27
29
25
30
33
SAMPLE
FLOW
RATE
(cc/min)
46
75
25
18
20
22.6
21.8
20.7
15.5
23.1
21.0
21.8
23.6
12
24
26
23.5
19
24
23
16
17.6
21.4
11
22.6
22.2
21.8
20
23.1
21
22.6
21.4
20
25
13.9 MAX
22.2
24
20
20
23.1
22.6
II
21.8
21.8
26.7
27.3
23.1
19.4
24.5
26.7
18.8
23.1
19.0
17.6
22.2
364
-------
Table E-2c. Selective filter (acetylene) log
(gulf coast area) (con.)
DATE
(1975)
10/14
Jl
II
10/19
II
"
I)
II
11
"
II
10/20
II
II
"
II
II
II
10/21
11
ii
it
M
10/30
II
II
II
II
II
10/31
II
II
II
"
"
II
II
II
II
II
n/i
it
ii
ii
11
ii
ii
FLIGHT
NO.
098
II
II
099
II
II
II
II
100
II
"
101
II
II
II
102
"
ii
103
II
II
II
tl
109
II
11
II
II
II
no
ii
H
11
ii
ii
ii
ii
11
ii
ii
m
11
ii
ti
"
M
a
ii
FILTER
NO.
1-35
2-67
3-72
1-29
2-69
3-58
4-131
5-61
1-70
2-7
4-172
1-42
2-55
3-134
4-132
1-82
2-118
3-50
1-79
2-149
3-333
4-103
If
1-40
2-53
3-10
4-43
5-136
6-140
1-139
2-99
3-64
4-26
5-327
5A-146
6-56
7-121
tl
8-97
9-2
1-20
2-119
3-112
4-31
5-76
6-14
7-16
8-144
SAMPLE COLLECTION
TIME (GMT)
FROM TO
2043
2117
2152
1601
1633
1702
1733
1808
1938
2008
2040
1558
1630
1656
1730
1851
1926
2004
2107
2136
2210
2237
2347
1928
2000
2030
2101
2131
2202
1846
1920
1948
2018
2047
Control
2118
2145
2231
2247
2315
1336
1401
1432
1501
1533
1603
1635
1829
2114
2150
2281
1631
1651
1730
1805
1850
2005
2037
2110
1627
1653
1727
1756
1923
2002
2027
2134
2208
2235
2249
2405
1958
2038
2059
2129
2200
2232
1918
1946
2016
2045
2114
_
2142
2200
2245
2314
2338
1359
1430
1455
1530
1601
1633
1703
1905
TOTAL
SAMPLE
TIME
(MIN)
31
33
29
30
28
28
32
42
27
29
30
29
23
28
26
32
36
25
27
32
25
30
II
30
28
29
28
29
30
32
26
28
27
27
_
24
29
29
27
23
23
29
23
29
28
30
28
36
-SAMPLE
FLOW
RATE
(cc/min)
20
21.8
19.4
7.0
25
26.5
21.4
15.2
24.4
24.5
22.6
27.3
15.4
30
24
16.4
7.5
20.7
20
21.4
23.1
18.8
II
26.1
27.6
27.2
23.1
21.8
23.1
19.6
18.8
23.1
20.7
17.6
.
18.5
22.6
27.6
19.8
22.2
22.2
23.1
21.4
25.5
21.8
22.6
21.4
24.5
365
-------
Table E-2d. Selective filter (acetylene) log
(northern route flights)
DATE
(1975)
9/5
II
II
M
II
H
II
II
9/6
II
II
.»
II
II
II
9/7
H
II
II
"
II
II
9/8
II
II
II
u
li
II
M
9/12
II
II
II
II
it
ii
9/27
H
II
tl
II
9/28
II
II
9/29
II
II
9/30
II
II
* II
II
FLIGHT
NO.
058
II
II
059
II
11
II
II
060
II
II
061
M
062
U
063
II
II
II
064
11
11
II
065
H
"
H
11
066
II
11
069
II
II
II
070
II
II
084
II
II
085
II
086
II
087
089
11
ii
090
n
11
091
"
FILTER
NO.
1-40
2-110
3-15
1-50
2-125
II
3-19
4-53
1-7
2-61
3-35
1-69
2-79
1-134
2-42
1-93
2-58
3-143
4-132
1-92
2-131
3-142
4-124
1-43
2-108
3-140
4-55
II
1-136
2-101
3-10
1-69
2-35
3-7
II
1-61
II
2-23
1-61
2-69
3-7
1-308
2-35
1-140
2-131
1-92
1-110
2-142
3-52
1-125
2-29
3-67
1-72
2-68
SAMPLE COLLECTION
TIME (GMT)
FROM TO
1656
1729
1802
1925
1957
2050
2106
2140
1724
1759
1827
2015
2051
2239
2312
1636
1707
1736
1808
2057
2133
2201
2235
1447
1519
1604
1637
1728
1803
1835
1908
1549
1623
1738
1658
1833
1753
1910
1543
1647
1728
2021
2205
1506
1615
1805
1708
1735
1845
1500
1607
1700
1818
1915
1725
1755
1828
1955
2011
2104
2138
2215
1755
1825
1854
2045
2144
2310
2338
1705
1734
1805
1840
2131
2159
2230
2302
1516
1558
1635
1647
1801
1829
1906
1931
1620
1655
1751
1715
1905
1755
1958
1615
1715
1801
2059
2338
1540
1646
1835
1732
1708
1915
1527
1637
1728
1847
1948
TOTAL
SAMPLE
TIME
(MIN)
29
26
26
30
29
II
32
35
31
26
27
30
53
31
26
29
24
29
32
34
26
29
27
29
39
31
43
II
26
31
23
31
32
30
II
34
II
48
32
28
32
38
33
34
31
30
24
33
30
27
30
28
29
33
SAMPLE
FLOW
RATE
(cc/min)
24
26
18.9
24
18
II
17.9
23
20
18.5
21.8
17.4
14
22.6
23.5
21.5
22.6
24.0
21.5
22.6
23.0
20.0
21.4
21.8
15.2
17.6
12.5
II
24.4
24.5
20
27.9
21.8
19.4
11
19.7
II
12.0
18.5
23.5
24
17.6
19.8
25
26.1
22.2
22.6
18.8
20
20
18.9
23.1
20.7
20.3
367
-------
Table E-2e. Sulfate filter log
(gulf coast area)
DATE
(1975)
9/19
II
9/21
II
10/9
II
10/10
II
10/13
10/14
II
10/19
II
10/20
II
10/30
10/31
II
11/1
II
FLIGHT
MO.
075
076
078
079
092
093
094
095
096
097
098
099
100
101
102
109
110
II
111
FILTER
NO.
9000041
II
9000019
II
9000042
II
9000043
9000018
9000054
II
9300053
II
9300052
11
9300051
9300050
9000049
II
COLLECTION
(GMT)
START
1509
2020
1446
1915
1853
2152
1642
2106
1643
1614
2038
1552
1937
1554
1850
1926
1845
2230
1334
1826
TIME
STOP
1812
2343
1756
2240
2127
2425
1938
2444
1954
1930
2236
1852
2208
1755
2059
2238
2200
2335
1708
1935
TOTAL
COLLECTION TIME
(MRS)
6.43
6.58
5.1
6.6
3.2
5.23
5.52
4.17
3.2
4.33
4.72
368
-------
Table E-2f. Sulfate filter log
(northern route flights)
DATA
(1975)
9/4
It
9/5
II
9/6
II
II
9/7
II
9/8
II
9/11
II
9/12
It
9/15
II
9/27
II
9/28
II
9/29
II
9/30
II
FLIGHT
NO.
055
056
058
059
060
061
062
063
064
065
II
067
068
069
070
071
072
084
085
086
087
088
089
090
091
FILTER
NO.
9000012
tl
9000013
II
9000014
II
11
9000015
9000016
1
9000017
11
9000035
II
9000036
II
9000037
II
9000038
II
9000039
II
9000046
II
COLLECTION
(GMT)
START
2158
0228
1652
1928
1725
2016
2237
1637
2054
1444
1728
1635
2116
1547
1738
2055
0015
1532
2020
1428
1732
1530
1659
1444
1815
TIME
STOP
0115
0454
1825
2215
1905
2158
2339
1927
2303
1931
1648
2032
0108
1715
2010
2302
0232
1833
2249
1652
1836
1610
2009
1727
2144
TOTAL
COLLECTION TIME
(MRS)
5,72
II
4.33
II
4.4
II
4.98
II
4.12
It
7.82
II
4.0
II
5.4
II
5.5
II
3.47
II
3.83
II
6.2
H
370
-------
Table E-2g. Summary low pass data
(gulf coast area)
DATE
(1975)
6/26
6/27
6/28
II
6/29
II
8/6
II
II
8/7
II
II
II
8/7
II
8/8
II
8/9
)l
8/11
II
9/21
II
10/10
II
10/13
II
10/14
II
10/19
1!
10/20
10/21
II
10/22
11
10/24
10/26
M
10/30
10/31
tl
FLIGHT
NO.
005
006
007
008
II
009
035
II
036
II
037
II
II
II
038
II
039
H
040
II
041
II
078
079
094
II
095
096
II
097/098
II
099/100
II
101/102
II
103
II
104
II
105
107
II
109
II
110
TIME
(GMT)
1757
1729
1440
1745
1042
1213
1801
1331
1335
2417
2422
1239
1244
1441
1445
1815
2030
1840
2219
1826
2200
1826
2153
1444
1920
1640
1938
2104
1639
1951
1611
2238
1554
2207
1557
2103
2248
2325
1543
1658
2145
2018
2132
1926
2238
1722
2322
LOCATION
OeP.idder
it
II
11
II
II
II
II
II
II
It
II
II
It
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
11
11
11
»
II
II
II
II
II
II
It
H
OZONE CONCENTRATION
(ugM3)
AIRCRAFT STATION
191
121
72
68
82
84
93
51
51
80
82
29
29
102
102
83
83
95
117
64
81
141
115
72
92
111
140
154
118
131
107
125
118
118
180
180
183
178
no
115
72
77
56
96
108
104
142
63
78
101
72
59
46
44
48
72
78
16
12
118
120
134
122
122
136
93
85
138
144
6Q
106
144
126
128
104
118
112
116
114
119
164
178
178
178
118
120
72
63
72
104
108
103
132
372
-------
Table E-3. Sulfate samples
Date
(1975)
9/4
9/6
9/7
9/11
9/12
9/15
9/19
9/21
9/27
9/28
9/29
9/30
10/9
10/10
10/13
10/14
10/19
10/20
10/30
10/31
Flight f!o.
055,056,057
060,051,062
063,064
067,068
069,070
071,072
075,076
078,079
084,085
086,087
088,089
090,091
092,093
094,095
096
097,098
01:9,100
101,102
109
110
Filter No.
9000012
9000014
9000015
9000017
9000035
9000035
9000041
9000019
9000037
9000038
9000039
9000046
9000042
9000043
9000018
9000054
9000053
9000052
9000051
9000050
-i-
(wg/m3)
B.D.
B.ii.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
B.D.
0.2
B.D.
NO;
o
2.7
2.4
3.9
1.3
1.1
0.6
1.9
3.2
6.6
9.8
1.5
5.6
2.1
2.4
2.1
2.7
4.4
5.1
10.6
5.7
S°A
'}•
(ug/m3)
6,5
1.4
14.5
0.7
B.D.
B.D.
0.7
9.7
21.2
22.2
0.2
18.6
8.6
2.8
2.9
8.6
8.4
11.4
25.5
11.2
11/1 111 9000049 B.D. 1.6 1.4
374
-------
M CO
375
-------
TLXAS
LOUISIANA
9/19/75
(1000V5001)
~ - 1.9
DeRidder
MISSISSIPPI
Z cone. : yg/nf
cone. :
i, wi_>u^, . . (iy/ in ».^
O - indicates flight
origin and termina-
J/JL1/75
(4000V5001)
NOj =3.2
S0= 9.7
GULF OF MEXICO
Figure E-3-2.
AIRCMFT FLIGHTS, 9/19/75-9/21/75
high volume filter measurements for sulfate and
nitrates.
376
-------
W3 co
•H CU
ft W
CO
tO }-i
f- 4J
O T3
-H d
I to
u-l
r^
-------
If *~
I
\ .
tn
r^.
; -
/ -^
/ °
*- oc
• .
CNJ CvJ
ds
o
m i
^ — c
i •-- -
i 'I
CO II <*
^> o
^ v to
CO O
378
-------
o
M •
•H 03
* 2
in n)
r- )-i
4J
.-) -H
. iH
O 3
i-l 03
« t-l
W O
H-l
I
B
O 03
H W
g S
pr. L^
^E £3
K 03
u n)
P4 0)
^
0)
• 4-1
in H
i -H
ro 4-i
w
01
n
3
fe
o n M
CD I CO II ^J
tM O O
— r< 1/1
CO
><
1 _
379
-------
-
SCL cone.
yg/m
O - indicates flight
origin and termina-
tion
Figure E-3-6. AIRCRAFT FLIGHT, 11/1/75
high volume filter measurements for sulfate and
nitrates.
380
-------
SELECTED EXAMPLE OF NORTHERN HIGH PRESSURE SYSTEM
SURVEY FLIGHT - NORTHERN HIGH PRESSURE OXIDANT STUDY
E-4A--Northern High Pressure Survey-Flights No. 021 through
031; July 22 to July 27, 1975
381
-------
•rl
VD
•H H
•a
a) u~>
> r--
n \
3 r-*
CO CN
cu
M
I
co r>
co ~-^
cu CN
}-l CM
O
I CM
-------
Table E-4a. Tabulated data: transition flight,
RDU-LCH (021-022), 7/22/75
TIME
(GMT) POSITION
(Flight 021)
18:22 RDU
24
42
43 LIB
44
46
48
50
52
54
56
58
19:00
02
04
06
08
09 CLT
10
12
14
16
18
20
22
24
26 SPA
28
30
32
34
36
38
40
42
44
46
48
50 TOC
52
54
56
58
20:00
02
04
06
08
10 Nello XN
12
HEADING
TAKEOFF
269°
II
100°
II
»
II
II
"
II
"
II
II
II
11
II
II
259°
II
II
It
II
II
II
II
II
256°
II
11
II
II
II
It
II
II
II
It
"
254°
"
»
II
11
II
II
II
II
II
239°
"
ALTITUDE
observed
(ft)
436
_
10000
II
II
II
"
"
II
II
"
II
"
II
II
II
II
II
II
it
II
II
11
II
11
II
II
II
II
II
If
II
II
II
II
II
"
tl
"
II
II
II
II
11
II
tl
"
11
II
03
(ug/m3)
_
.
87
-
80
80
80
87
87
101
94
80
80
94
87
80
73
80
80
73
73
73
73
73
73
73
66
73
73
73
73
73
73
73
73
73
73
73
73
73
73
80
73
73
80
' 80
87
87
94
TEMP.
34
_
.
-
_
.
.
_
-
.
-
.
-
-
„
„
.
.
-
_
.
.
.
-
—
.
-
.
-
_
-
.
-
-
_
-
-
-
-
_
-
-
-
-
.
-
-
-
-
TRUE
AIR SPEED
observed
(mph)
140
_
200
-
.
.
-
.
-
.
-
-
.
-
^
.
-
.
-
..
_
.
-
-
—
.
.
.
-
_
.
-
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
383
-------
Table E-4a. Tabulated data: transition flight,
RDU-LCH (021-022), 7/22/75 (con.)
TIME
(GMT) POSITION
20:14
16
18
20
22
23 RMG
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
55 BHM
56
58
21:00
02
11 TCL
(Flight 022)
22:03 TCL
04
18
20
22
24
26
28
30
32
34 ME I
36
38
40
42
44
46
48 MZE XN
50
52
HEADING
II
II
II
II
II
25'°
II
II
II
II
II
II
"
II
II
«
II
II
II
II
II
II
230°
II
"
H
II
LANDING
TAKEOFF
223°
II
II
II
II
II
II
II
II
216°
II
II
"
II
II
11
22:9°
1
1.
ALTITUDE
observed
(ft)
II
II
II
II
II
II
II
H
II
II
II
II
II
II
11
II
II
II
11
II
M
II
II
II
-
.
.
169
169
.
8000
II
It
II
II
II
II
II
•M
II
II
"
"
"
11
II
II
II
TRUE
03 TEMP. AIR SPEED
(ug/m3) Ob"™*1 observed
( C) (mph)
87
94
101
94 -
94
* * •
80 -
80
80
115
122
115
94
94 -
87 -
87 -
94
108
122
122
170
156
149
115
135
170
. .
33 140
_
82 - 200
82
88 -
88
88
88 -
88
88
88
88
88
88
95
95
95
'95
95
95
384
-------
Table E-4a. Tabulated data: transition flight,
RDU-LCH (021-022), 7/22/75 (con.)
TIME
(GMT) POSITION HEADING
22:54
56
58
23:00
02
04
06
07 MC8 239°
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
II
56
58
24:07 LCH LANDING
ALTITUDE 03 TEMP.
observed (ua/m3\ observed
(ft) ^9/m ; (oc)
88
82
88
101
88
88
,'! 76
76
70
70
70
70
70
76
7£
76 - -
70
70
70
70
76
76
76
76
70
76
70
40
64
82
82
88
101
16
TRUE
AIR SPEED
observed
(mph)
.
.
_
-
_
_
.
_
-
^
.
_
.
-
-
_
_
-
.
_
.
.
-
^
.
.
_
-
„
,
„
_
-
385
-------
Table E-4a-l. Tabulated data: transition flight,
LCH-SDY (023-024), 7/23/75
TIME
(GMT) POSITION HEADING
(Flight 023)
16:17 LCH TAKEOFF
18 " 32:°
32
34
36
38
40
42
44
46
48
50
52
54
55
53
17:00
02
04
06
08
10
12
14
15
17 6GG 335°
18
20
22 "
24
?6
*3Q "
30
32
34
36
38
40
42
44
46
48
50
52
54
56
ES
18:00
02
ALTITUDE
observed
(ft)
16
_
_
10000
II
II
II
II
II
II
II
II
II
II
II
It
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
tl
II
11
11
II
II
"
.1
II
II
II
II
II
II
II
II
II
II
II
03 TEMP.
(ug/m3) observed
28
.
66
66
80
80
66
60
59
59
66
73
73
80
80
73
66
73
73
73
73
73
73
73
80
. -
80
80
66
66
66
66
66
66
73
73
73
66
66
80
73
59
59
59
66
59
59
66
52
52
TRUE
AIR SPEED
observed
(mph)
140
.
_
200
-
_
-
.
-
-
_
-
.
-
-
.
-
-
-
-
.
-
-
-
-
_
-
-
-
-
.
-
-
-
-
.
-
-
-
-
-
-
-
-
-
.
-
-
-
-
386
-------
Table E-4a-l. Tabulated data: transition flight,
LCH-SDY (023-024), 7/23/75 (con.)
TIME
(GMT) POSITION
18:06
08 MLC
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44 PER
46
48
50
52
54
56
58
19:00
02
04
06
08
09 ICT
(Flight 024)
21:06 ICT
08
22
24
26
28
30
32
34
36
38
40
42
44
•56
HEADING
II
323°
II
II
"
II
II
II
It
11
tl
II
II
II
II
II
II
II
II
338°
II
II
II
II
M
tl
II
"
II
* II
11
II
LANDING
TAKEOEF
334°
n
n
ii
n
ii
it
n
n
n
n
n
n
"
TRUE
ALTITUDE 03 TEMP. AIRSPEED"
observed (,ja/m3\ observed observed
(ft) (vg/m ' (0C) (mph)
52
52
52
52
45
52
52
45
52
11 52 -
52
59
66
73
11 66 -
66
73
73
73
73
73
66
52
38
45
38
52
73
87
66
_ _
...
1332
1332 - - WO
10500 61
11 68 -
76
76
68
11 68 -
68
68
11 68 -
.68
7.6
83
76
387
-------
Table E-4a-l. Tabulated data: transition flight,
LCH-SDY (023-024), 7/23/75 (con.)
TIME
(GMT) POSITION HEADING
o
21:48 334
50
52
54
56
o
58 TKO 325
22:00
02
04
06
08
10
12
14
16
18
20
22 EAR
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
53
23:00
02
04
06
08 " 0
09 AIM 321
10
12
14
16
18
20
22
24
ALTITUDE 03 TEMP.
ob«rved (ug/m3) observed
10500 83
76
76
76
76
76
76
76
68
61
61
61
61
61
54
61
61
61
54
61
47
47
40
47
47
40
47
40
47
40
47
47
47
47
54
54
54
54
61
68
76
11 - -
68
76
76
68
61
54
61
61
TRUE
AIR SPEED
observed
(mph)
-
-
-
-
-
-
-
-
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
_
-
-
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
388
-------
Table E-4a-l. Tabulated data: transition flight,
LCH-SDY (023-024), 7/23/75 (con.)
TIME
(GMT)
23:26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
00:00
02
04
06
08
10
12
14
15
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
01:00
02
POSITION HEADING
II
II
II
II
II
II
II
II
II
M
II
11
II
II
II
II
II
tl
II
II
II
"
II
II
II
DPR 315°
"
it
n
n
ii
n
ii
ii
"
n
n
ii
n
n
it
n
"
n
"
H
"
ii
n
n
ALTITUDE
observed
(ft)
II
ii
II
II
"
II
II
II
ii
M
II
II
II
II
II
II
" "
II
II
II
»
"
H
II
M
II
II
M
II
II
II
II
II
n
ii
n
ir
ii
n
n
M
II
11
II
"
"
l|
II
II
II
03 TEMP.
(pg/m3) obse™5d
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
61
68
68
68
68
76
„ _
76
76
76
76
76
68
68
68
68
68
68
68
68
61
61
61
61
61
61
61
.61
61
54
54
TRUE
AIR SPEED
observed
(mph)
.
_
.
-
—
.
.
.
-
_
_
.
.
-
_
_
_
.
-
_
.
.
.
-
_
.
_
_
-
_
.
_
.
-
.
.
_
_
-
.
.
_
.
-
—
.
.
.
-
389
-------
Table E-Aa-1. Tabulated data: transition flight,
LCH-SDY (023-024), 7/23/75 (con.)
TRUE
TIME ALTITUDE 03 TEMP. AIR SPEED
(GMT) POSITION HEADING observed (Ua/m3) observed observed
(ft) U9/m > (0C) (mph)
01:04 " " 54
06 " - 47
08 " - 47
10 " - 47
12 " - 47
14 " - 47
16 " - 47
18 " - 47
20 " - 54
22 " - 54
24 " - 54
26 " - 54
28 " - 54
30 " - 54
31 SOY LANDING 1983
390
-------
o
L/1
CM
CM
CM
CM
O
I
rH
CM
O
o
JS
60
•H
C
O
•H
cn
C
cfl
CM
-------
Table E-4a-2.
Tabulated data: high pressure survey flight
(025-026), 7/24/75
TIME ALTITUDE
(GMT) POSITION HEADING observed
(ft)
(Flight 025)
17:07 SOY TAKEOFF 1983
08
10 " LOW PASS 2033
12
14 " 117°
16
18 " 4000
20
22
24
26
28
30
32
34
36
38 BELFIELD 118°
40
42
44
46
48
50
52
54
56
58
18:00
02
04
06
08
10
12 -
14
16
18
20
22
24
26
28
30
32 RIDGEVIEW
34
36
38
40
18:42
(I
It
II
II
li
II
11
II
II
II
«
II
11
11
II
II
II
11
II
II
II
II
II
II
44
03
(vg/m3)
-
_
.
42
37
37
37
42
37
37
37
37
37
42
42
42
42
42
42
37
37
37
37
37
37
37
37
37
37
42
37
37
37
37
37
37
37
37
37
37
37
37
37
37
37
42
TEMP.
observed
-
_
-
.
.
-
.
_
.
.
-
—
-
.
.
-
_
.
.
.
-
_
-
.
.
-
_
.
.
.
-
.
-
-
-
-
.
-
-
-
-
.
.
.
-
-
TRUE
AIR SPEED
observed
(mph)
170
.
190
.
.
-
.
.
-
.
-
m
-
.
-
-
.
.
-
-
-
_
.
.
.
-
_
-
-
.
-
_
-
-
-
•
.
-
-
-
-
.
-
-
-
-
392
-------
Table E-4a-2. Tabulated data: high pressure survey flight
(025-026), 7/24/75 (con.)
TIME
(GMT) POSITION
18:46
48
50 PIR
52
54
56
58
19:00
02
04
06
08
10
12
14
16
18 HON
20
23
28
33
39
44
49
56
20:00
04
06
08
09 HON
(Flight 026)
21:32 HON
34
42
44
46
48
50
52
54
56
58
22:00
02
04
06
08
10
12
14
16
HEADING
II
It
076°
II
II
tl
II
II
II
II
11
II
II
II
II
II
VERTICAL
II
II
II
II
II
"
II
tl
tl
It
11
II
LANDING
TAKEOFF
152°
II
II
11
It
II
11
II
II
II
11
II
II
II
«
II
"
II
II
ALTITUDE
observed
(ft)
it
"
4000
II
II
II
II
It
II
II
II
II
II
II
II
II
II
5000
6000
8000
10000
12000
10000
8000
6000
5000
4000
„
*
1287
1287
_
5500
"
"
.,
ii
ii
H
ii
H
-------
Table E-4a-2. Tabulated data: high pressure survey flight
(025-026), 7/24/75 (con.)
TRUE :
TIME ALTITUDE 03 TEMP. AIR SPEED
(GMT) POSITION HEADING observed (ua/m3) observed observed
(ft) IU9/IB (0C)
22:18 " 61
20 " 61
24 RANDOLPH " ".
44 •'-_
57 LINCOLN LANDING 1198
394
-------
MONTAN
D2J08Z
iclrita
A1909X
central
time zone
CGG
17.1.7Z
SCALE: 1 in = 276 mi
TIME: GMT
WEATHER: 0 8+H/120015+H
Lake Charles, LA.
Figure E-4A-3. Transition flight, LCH-SDY (023-024), 7/23/75.
395
-------
N. bAKCH'A
D17UZ
(025)
(LP-i)
MONTANA
SCALE:
TIME:
WKATIIER:
1 in •-= 110 mi
GMT
0 30H1/0 30h
D213'iZ\(02C>)
Huron
mountain
f irae xone
Figure E-4A-4. High pressure survey flight (025-026), 7/24/75.
396
-------
S
i
tfl
4J
cd
Q
3
O
CO
C
O
•H
H-l
2
ex
u
•H
01
V-l
60
•H
o
8
o
O
O
_JL_
g
.-U
o
397
-------
10
in
CM
CM
o
00
•H
rH
co
0)
(-1
co
CO
00
•H
«
vO
sf
W
oo
B
o
. ?
c o
G O
gss
«> H &
398
-------
70000
Tito; (ci-rr)
ASCENDING DESCENDING
18000
16000
14000
« 12000
•3
v 10000
8000
6000
4000
2000
1855
1851
1846
1843
J839
1901
13C5
1911
1915
1918
40 80 120 160
OZONE (wg/m3)
200 240
Figure E-4A-7. Vertical profile, Peoria, Illinois (027), 7/25/75.
399
-------
Table E-4a-3.
Tabulated data: high pressure survey flight
(027), 7/25/75
TIME
(GMT)
16:43
44
50
52
54
56
58
17:00
02
04
06
08
10
12
14
16
18
20
22
24
26
27
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
18:00
02
04
06
08
10
12
14
16
18
20
22
POSITION HEADING
LNK TAKEOFF
078°
11
II
II
II
II
II
II
11
II
II
It
RED OAK
II
II
II
II
II
11
II
CSQ LOW I'ASS
II
073°
II
II
11
II
II
II
II
II
II
II
II
II
II
"
II
11
OTM 100°
II
li
H
11
H
II
II
11
II
ALTITUDE
observed
(ft)
1198
3500
II
II
II
II
II
II
II
II
II
II
11
II
II
t|
II
II
-
.
-
.
3500
II
It
II
II
II
II
II
II
II
II
II
II
II
II
II
M
3500
11
II
11
II
II
ll
II
11
II
TRUE "
03 TEMP. AIR SPEED
dm/m3t observed observed
* 9 (°C) (mph)
140
- - 190
99 -
99 -
94 -
94
89 -
94
94 -
89 -
84 -
84 -
84 -
84
84
84
84 -
84 -
.
_ « -
81 -
_
79
79 -
79
79 -
79
70
75 -
75 -
75 -
75 -
79 -
79 -
84 -
84
84
84
84
89 17
84'
75
84
84 -
79
• 79
79
79
79
400
-------
Table E-4a-3. Tabulated data: high pressure survey flight
(027), 7/25/75 (con.)
TIME
(GMT) POSITION
18:24
26
28 BRL
30
32
34
36
39 PIA
43
46
51
56
19:01
06
11
15
18
20
22
23
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
20:00
02
04
06
08
10
12
14
15
16
18
20
22 TERRE HAUTE
HEADING
ii
11
0880
II
II
..
It
VERTICAL
II
II
II
11
II
II
tl
II
II
II
II
117°
II
II
II
II
II
II
"
II
II
II
»
II
II
II
II
II
II
II
11
II
II
II
II
II
II
II
140°
H
II
II
LANDING
ALTITUDE
observed
(ft)
it
II
3500
ii
it
H
_
5000
6000
8000
10000
12000
10000
8000
6000
5000
3500
II
II
II
II
II
II
II
II
II
II
II
II
11
II
II
II
"
H
II
II
II
II
II
-
11
11
II
II
II
.
«
"
II
585
03 TEMP.
(ug/n.3) observed
79
79
84
_
.
« —
*
96
52
88
45
80 5
52
58
106
117
118
108
108
108
108
104
108
36
108
123
89
89
89
89
89
94
89
79
, 89
89
89
104
104
108
108
118
108
108
108
113
104
99
TRUE
AIR SPEED
observed
(mph)
.
.
_
-
—
_
.
.
-
_
.
.
..
-
,
_
_
_
-
^
_
«.
„
-
.
_
.
*
-
.
.
_
_
-
„
_
_
_
-
„
_
.
.
-
—
_
_
—
_
401
-------
(1324-1900Z)
Chatham
yx^ y~ D1901X,
Terre Haute
D1636Z
eastern
tLme ?:one
A1955Z
Pittsburgh
SCALE: 1 in =
TIME: GMT
WEATHER: 50(j)84-H
110 mi
Figure E-4A-8. High pressure survey flight (028), 7/26/75.
402
-------
B
8
H
vO
CM
8
I
CO
CM
o
4-1
•H
U-<
O
o
•rl
t-l
0)
st
8
•o
o
g
o
§
(isu
aan.ti.uv
403
-------
Table E~4a-4. Tabulated data: high pressure survey flight
(028), 7/26/75
TIME
(GMT) POSITION HEADING
16:35 HUF TAKEOFF
36 " 0*3°
38
40
42
44
46
48
50
52
54
56
58
17:00
01 IND
02
04
06 " '
08
10
12
14
16
18
20
22
24
26
28
29 DEC
30
32
34
36
38
40
42
44
46
48
50
52 ' 0
54 WATERVILLE 0*0°
56
58
18:00
02
04
06
08
10
ALTITUDE
observed
(ft)
585
-
.
3500
II
II
II
II
II
II
II
II
II
II
tl
II
II
II
II
II
II
II
II
II
II .
II
II
II
"
II
II
II
II
II
II
II
II
II
If
II
II
II
II
II
II
II
II
II
II
II
II
03 TEMP.
-
.
118
108
108
108
104
113
118
118
118
128 .
113
-
T08
108
108
104
104
99
99
99
99
99
104
104
104
108
-
113
113
108
104
104
108
108
113
113
118
104
104
104
104
89
104
128
-157
-
84
65
TRUE
AIR SPEED
observed
(mph)
140
-
-
190
-
_
-
-
-
•
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
~
-
-
-
-
*
-
-
-
-
-
404
-------
Table E-4a-4.
Tabulated data: high pressure survey flight
(028), 7/26/75 (con.)
TIME
(GMT) POSITION
18:12
14
16 WINDSOR
18
20
22
24 CHATHAM
26
32
35
37
41-
46
51
54 • • i
57
58
19:00
01
02
04
06
08
10
12
14 LOST NATION
16
18
20
22
24
26
28
30 YMG
32
34
36
38
40
42
44
46
48
50
55 ALLEGHENY
HEADING
II
II
077°
II
II
tl
VERTICAL
II
II
II
II
II
II
"
II
II
II .
II
145°
II
II
II
II
II
II
11
II
II
II
II
M
II
II
11
II
II
II
II
II
II
II
II
II
II
LANDING
ALTITUDE
observed
(ft)
II
tl
II
II
II
II
.
5000
6000
7000
8000
10000
8000
6000
5000
3500
1000
1000
H
_
.
.
3500
II
"
"
"
"
"
»
"
"
H
"
"
"
"
"
"
«
"
"
-
1252
<:3.3,
75
80
89
84
84
89
.
91
79
95
95
115
101 •
57
. 96
70
62
65
75
80
-
.
-
65
60
55
65
80
-
99
84
99
99
113
113
137
128
104
99
104
108
128
-
-
TEMP. .
observed
.
.
-
.
-
.
.
12
11
-
.
7
-
-
-
_
-
—
-
_
-
-
-
-
.
-
-
-
-
_
-
-
-
-
.
-
-
-
-
_
-
-
-
-
TRUE
AIR SPEED
observed
(mph)
.
.
-
-
-
.
>
.
-
-
_
.
-
-
-
.
-
*•
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
-
-
_
-
-
-
-
405
-------
east-em
time zone:
NEW YORK
Bradford
-1619Z.
(LP-l7\
D3542Z
(029)
Pittsburgh
Martha's Vineyard
/- A2112Z
'
Selingsgrove.
165 77.
',(1743-18127.)
Atlantic City
V2015Z
70°42'W, 39°24'N
D1906Z
(030)
SCALE: 1 in = 110 mi
TIME: GMT
WEATHER: 504KH/700
Figure E-4A-10. High pressure survey flight (029-030), 7/27/75.
406
-------
in
CN
r^
ON
CN
O
HI
co
•H
U
C
to
H
O
M
Cu
ffl
o
•H
S
W
(U '
>-l
3
60
S
o
(1SH
407
-------
Table E-4a-5.
Tabulated data: high pressure survey flight
(029-030), 7/27/75
TIME
(GMT) POSITION
(Flight 029)
15:40 AGC
42
44
46
48
50
52
54
56
58
16:00.
02
04
06
08
10
12
14
16
18
19 BFD
20
21
22
24
26
28
30
32
34
36
38
40
42
44 PIPER MEMORIAL
46
48
50
52
54
56
57 SELINGSGROVE
' 58
17:00
02
HEADING
TAKEOFF
040°
II
II
II
II
II
II
II
II
II
II
II
II
"
II
11
II
II
II
LOW PASS
II
040°
It
II
,,
II
II
II
II
II
II
II
II
"
II
II
II
II
II
II
137°
II
II
11
ALTITUDE
observed
(ft)
1252
_
3500
II
II
II
II
II
II
II
II
II
II
II
It
II
tl
II
-
-
_
_
.
. •
3500
it
"
"
"
"
»
"
"
ii
"
••
"
"
"
"
»
it
11
"
it
°3 TEMP.
(yg/m3) observed
.
_
137
_
161
161
171
166
161
161
157
152
147
152
T52
147
147
147
147
147
_ _
122 18
147
147
142
142
137
142
137
137
147
157
132
147
157
142
137
128
147
137
16
.
161
147
TRUE
AIR SPEED
. observed
(mph)
T40
.
190
.
-
.
.
.
.
-
.
-
.
.
-
.
-
.
.
-
_
_
.
.
-
.
-
-
.
-
-
-
-
-
-
-
-
-
-
-
.
-
-
-
-
408
-------
Table E-4a-5. Tabulated data: high pressure survey flight
(029-030), 7/27/75 (con.)
TIME
(GMT) POSITION
17:14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
43 ACY
46
49
52
58
18:03
08
12
19
(Flight 030)
19:05 ACY
06
08
09 COAST
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
HEADING
II
II
II
II
' II
II
II
II
11
11
II
II
II
II
II
VERTICAL
II
II
'll
II
It
II
II
LANDING
TAKEOFF
112°
II
II
II
tl
II
II
It
II
II
It
II
II
II
II
II
II
11
II
II
II
II
II
M
ALTITUDE
observed
(ft)
II
II
"
11
"
II
II
II
II
II
II
II
M
tl
II
_
5000
6000
8000
10000
8000
6000
5000
76
76
-
.
3000
H
II
II
M
rt
ll
ll
ll
ll
il
tt
n
ll
it
il
"
»
tl
it
II
n
°3
(yg/ffl3)
118
123
128
118
118
118
132
132
166
-
214
157
U6
176
166
.
75
79
101
108
101
85
75
•
-
_
-
-
153
139
139
78
115
120
125
no
120
115
115
115
106
87
73
87
- 82
97
101
82
TEMP.
observed
(°C)
_•
.
.
.
-
„
.
.
.
-
_
.
.
.
16
_
.
12
-
10
_
.
.
•
-
.
-
-
.
-
-
-
-
,
.
.
-
-
_
.
-
-
-
.
.
-
-
-
TRUE
AIR SPEED
observed
(mph)
—
-
-
.
-
»
.
.
-
-
.
-
-
-
-
.
-
-
-
-
.
-
-
~
140
-
-
190
-
_
-
-
-
-
_
-
-
-
-
.
-
-
-
-
-
-
-
-
-
409
-------
Table E-4a-5.
Tabulated data: high pressure survey flight
(029-030), 7/27/75 (con.)
TIME
(GMT)
19:52
54
56
58
20:00
02
04
06
08
10
12
14
15
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
. 58
21:00
02
04
05
12
POSITION HEADING
H
II
U
N
n
II
"
"
H
H
II
II
70042'W,39024'M 355°
II
II
n
n
n
H
n
n
n
11
n
it
H
H
H
II
H
II
II
n
n
NANTUCKET
II
n
H
II
MARTHA'S VINEYARD LANDING
ALTITUDE
observed
(ft)
n
n
H
H
-
3500
11
n
»
n
n
n
H
H
n
n
"
it
H
II
II
It
II
II
II
II
II
It
II
II
It
II
II
II
H
II
II
II
-
68
TRUE
°3 TEMP. AIR SPEED
(yg/m3) observed observed
(°C) (mph)
82 -
78 -
78
55
16
46 18
46 -
70
55
70 -
80
65 -
70 -
67
75 -
80
80 -
80
70
55
70
65
46 -
51 -
46
46 -
46
51 -
51
51 -
51
51
46 -
51
46 -
75 -
84 -
.
-
410
-------
eastern
time zone
D2208Z ^Martha's Vineyard
t—' I
SCALE: 1 in = 110 ml
TIME: GMT
Figure E-4A-12. Transition flight, ACK-RDU (031), 7/27/75.
411
-------
Table E-4a-6. Tabulated data: transition flight,
ACK.-RDU (031), 7/27/75
TIME
(GMT)
22:08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
23:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
37
38
40
42
44
POSITION HEADING
MARTHA'S VINEYARD TAKEOFF
260°
II
II
II
II
II
II
II
II
II
II
II
II
"
II
II
11
II
II
II
II
23(5°
II
II
II
II
II
II
II
II
11
11
II
"
II
II
II
t|
II
II
11
"
11
II
213°
it
It
H
II
ALTITUDE
observed
(ft)
68
.
4500
U
II
II
II
II
II
II
II
II
II
II
II
II
u
tl
II
"
tl
II
II
11
II
II
a
n
u
"
ti
u
u
u
u
it
it
u
ti
u
u
u
u
11
ii
ii
u
ii
u
u
TRUE :
°3 TEMP. AIR SPEED;
(uq/m3) observed observed
9 ' (°C) (mph)
140
_ -
190
. -
94
88 -
104
88 -
104
114
99 -
_
94 -
99 -
109
104
104
109
88
88
B3
83 -
78 -
78
78 -
78 -
78
73 -
73 -
73
78
73 -
73
73
78 -
78
78
78 -
78 -
78
83
83 -
78
94
124
.
124
129
149
144
412
-------
Table E-4a-6. Tabulated data: transition flight,
ACK-RDU (031), 7/27/75 (con.)
TIME
(GMT)
23:46
48
50
51
52
54
56
58
24:00
02
03
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
01:00
02
04
06
08
10
12
14
16
18
20
22
23
POSITION HEADING
II
II
II
If
. If
If
II
II
11
II
240°
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
II
»
II
FRANKLIN 246°
11
II
II
II
II
II
II
II
II
II
"
'*
II
II
II
II
t|
"
RDU LANDING
ALTITUDE
observed
(ft)
ii
it
ii
ii
n
n
n
ii
6500
II
II
II
II
11
II
II
II
II
II
M
..
II
II
II
II
II
11
"
II
II
II
II
II
II
n
ii
it
n
11
n
M
11
n
n
ii
.
.
.
.
.
*
436
03 TEMP.
(ug/m3) observed
134
119
119
.
-
m —
.
.
98
93
— m
98
98
93
104
104
104
104
109
115
121
121
121
126
126
115
115
109
109
109
109
115
121
121
121
126
115
115
115
115
115
115
121
137
137
154
165
154
149
137
137
-
TRUE
AIR SPEED
observed
(mph)
—
.
_
180
-
m
.
.
190
-
—
.
_
.
-
—
_
_
_
-
_
_
.
_
-
_
_
.
.
-
_
,
.
.
-
-
.
.
-
_
.
.
-
-
_
_
.
.
.
-
-
413
-------
SELECTED EXAMPLES OF THE DOWN WIND PLUME FLIGHT
GULF COAST OXIDANT STUDY
E-4B--Flight No. 005 on June 26, 1975.
E-4C--Flight No. 006 on June 27, 1975.
414
-------
Henderson
- 20447.
1'airfield
LOUISIANA
•D180QZ
(LP-U
DeRidder
San Ravburn. Rca.
central time, zone
A2142Z
LAKE CHARLES
1916Z
Freeport
GULF 0? MEXICO
SCALE: 1 in ^ 35
TIME: GMT
WEATHER: 35
-------
Table E-4b-l.
Tabulated data:
(005), 6/26/75
downwind plume
TIME
(GMT) POSITION
17:52 ORI
17:57
18:00
02
•04
06
•08
10
12
14 , JASPER
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44 SABINE PASS
46
48
50
52
54
56
58
19:00
• 02
04
06
08
10
12
14
16 FREEPORT
18
20
22
24
26
28
30
32
34
HEADING
TAKEOFF
LOW PASS
285°
H
185°
II
II
H
II
II
H
II
II
II
II
II
H
II
II
240°
II
H
II
H
II
„
n
ii
n
"
n
u
u
n
n
345°
II
II
II
II
II
II
II
II
II
ALTITUDE
observed
(ft)
- 203
253
1500
II
n
H
H
u
H
II
u
II
II
II
II
II
H
11
II
II
II
H
II
II
II
H
n
n
n
n
H
n
n
it
n
it
n
ii
n
ii
n
u
n
n
ii
it
n
u
it
u
«*,
191
204
182
182
200
196
187
196
174
200
187.
182
204
191
209
178
174
138
200
196
182
178
174
182
244
191
152
147
156
213
261
182
174
191
204
165
160
130
116
112
103
116
116
121
121
169
310
191
116
MANIFOLD TFMp
TEMP. TEWt
_ _
. .
.
•
„ m
.
. .
• •
28.6
28.0
25.3
11
25.9
26.4
n
11 *
n _
u ^
H _
II _
H -
H _
II ^
« ^
II _
25.9
26.4
II- ^
II _
27.0
28.6
It _
II _
29.6
II _
II _
tl m
29.1
28.6
II _
28.0
n _
n ^
11
27.5
"
"
"
TRUE
AIR SPEED
observed
(mph)
160
m
180
.
-
—
_
_
-
—
.
.
.
- •
— .
-
.
.
-
_
.
-
•
-
.
.
.
-
-
„
-
-
-
-
_
>
-
-
-
.
-
.
-
-
.
-
-
-
-
416
-------
Table E-4b-l. Tabulated data: downwind plume
(005), 6/26/75 (con.)
TIM
(GM • POSITION
19: i
!
!
;
I
i
i
)
'
I
i
3
20: i
>
1
CS
C 3
10
12
14
16
17 FAIRFIELD
18
20
22
24
26
28
30
32
34
36
38
40
42
44 HENDERSON
46
48
50
52
54
E6
53
21:00
02
04
06
08
10
12
ALTITUDE 03
HEADING observed (ug/m3)
(ft)
TV
99
99
90
112
112
108
" " 1 47
121
116
86
121
143
147
182
178
204
226
209
204
178
080°
196
200
226
209
222
218
204
204
196
182
143
t ^
M
n
it
it ^
"
"
25.9
26.4
it ^
"
27,5
"
"
,.
24.3
23,7
2^, '.
1
11
H
'*
23,7
24.3
II
'.4.8
(i
•'
25,3
TRUE
AIR SPEED
observed
(mph)
.
_
_
-
.
.
_
-
.
_
«.
»
„
.
™
_
„
-
.
.
„
_
-
.
.
_
„
.
*
„
_
-
_
_
,
-
.
*
-
_
-
w
«i
«
„
^
417
-------
Table E-4b-l. Tabulated data: downwind plume
(003), 6/26/75 (con.)
TIME
(GMT) POSITION
21:14
16
18
20
22
24
25 JASPER
26
28
30
32
34
36
38
42 LCH
HEADING
H
145°
II
II
H
II
It
H
II
LANDING
ALTITUDE
observed
(ft)
ii
H
H
ii
H
H
II
II
II
II
II
H
II
_
16
03
(ug/m3)
160
165
165
156
165
165
196
156
165
147
239
244
143
MANIFOLD TFMP
Trup TEMr.
^ "TO*1
II _
II _
tt _
H _
25.9
26.4
H _
H _
U _
II _
II _
11
II _
. .
•
TRUE
AIR SPEED
observed
(mph)
.
_
—
-
—
—
.
-
—
.
.
.
.
418
-------
HENDERSON
FairfieId
LOUISIANA.
Sam Rayburn. Res..
central
time zone
20382 DeRiclder
A20552
D1730Z
(LP-1)
\_1841Z
Freeport
GULF Cr MEXICO
SCALE: 1 in « . 35 mi
TIME: GMT
WEATHER: 25$vf, Haze
Figure E-4C-I. Downwind plume flight (006), 6/27/75.
419
-------
Table E-4c-l.
Tabulated data:
(006), 6/27/75
downwind plume
TIME
(GOT) POSITION HEADING
17:25 DRI TAKEOFF
28
29 " LOW PASS
30 " 286°
32
34
36
38
40
42
44
45 JASPER IS 5°
46
48
50
52
54
56
58
18:00
02
04
06
08 "
10
12 SABINE PASS 242°
14
16
18
20
22
24
26
28
30
32
34
36
38
40
41 FREEPORT 349°
42
44
46
48
50
52
54
56
58
ALTITUDE
observed
(ft)
203
-
253
_
1500
11
it
It
SI
it
il
il
ti
tl
it
»
il
>i
11
u
il
il
ll
u
11
n
u
"
u
II
II
II
11
11
II
II
II
II
II
II
II
11
II
II
tl
It
II
II
II
II
°3
(ug/m3)
.
121
_
130
116
121
99
81
95
143
116
134
112
121
103
95
99
99
125
86
147
95
55
59
20
.
.
-
„
-
73
95
59
59
68
59
55
51
_
55
64
55
55
55
68
77
86
73
MANIFOLD
TEMP.
observed
(°C)
_
.
_
-
„
.
.
.
-
29.6
II
II
II
II
tt
11
II
II
II
n
ii
u
ii
11
il
u
il
u
u
'•
29.1
El
28.6
II
II
II
28.0
11
tl
28.6
II
II
U
28.0
27.5
II
il
II
**
420
-------
Table E-4c-l. Tabulated data: downwind plume
(006), 6/27/75 (con.)
TIME
(GMT)
19:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
20:00
02
04
06
08
09
10
12
14
16
18
20
22
24
26
28
30
32
34
36
POSITION HEADING
II
II
II
II
, II
II
II
II
11
II
11
"
II
II
II
II
tl
II
'II
11
II
II
FAIRFIELD 80°
tl
II
II
II
II
II
ll
II
II
II
II
II ,
HENDERSON 152°
II
II
»
11
"
II
"
II
II
II
II
II
II
II
ALTITUDE
observed
(ft)
II
II
II
It
II
II
II
II
II
II
II
II
"
II
II
II
II
II
II
II
II
II
"
II
II
II
II
II
II
II
II
II
II
II
11
II
II
11
II
II
II
It
11
II
II
II
II
II
II
II
°3
(vg/m3)
73
86
77
68
73
86
86
68
68
77
64
64
64
95
95
59
59
64
68
95
86
81
99
77
90
95
99
112
112
125
125
112
112
125
165
_
174
169
147
169
121
156
147
130
116
121
95
95
112
134
MANIFOLD
TEMP.
observed
(°C)
26.4
25.9
26.4
II
II
II
27.5
II
II
II
II
11
II
27.5
tl
II
II
28.0
It
28.6
II
II
29.1
29.6
II
»
II
II
II
II
29.6
II
30.2
II
II
II
11
II
II
II
30.7
30.2
30.7
II
II
II
II
II
II
II
421
-------
Table E-4c-l. Tabulated data: downwind plume
(006), 6/27/75 (con.)
TIME ALTITUDE 0 MANIFOLD
(GMT) POSITION HEADING observed 3 TEMP.
(ft) (ug/m3) observed
20:38 JASPER 104° " 130
40 " " 156
42 " " 147
44 " " 138
46 " " 125
48 " " 121
50 " " - 31.8
52 " - - 32.9
55 DRI LANDING 203
422
-------
SELECTED EXAMPLES OF THE SEA-BREEZE FLIGHTS-
GULF COAST OXIDANT STUDY
E-4D--Flight No. 075 (Morning Flight) on October 19, 1975.
E-4E--Flight No. 076 (Afternoon Flight) on October 19, 1975.
423
-------
Sam Rayburn Res.
18002
LOUISIANA
TEXAS
Houston
0
X_1645Z
94 00'W
DeRidclor
D 15087.
A 1817Z
central tine zone
SCALE: 1 in = 35 mi
TIME: GMT
WEATHER: 15
-------
Table E-4d-l. Tabulated data: sea-breeze flight
(075), 9/19/75
TIME
(GMT) POSITION
15:03 DRI
08
10
12
14
16
18
20 SAM RAYBURN RES.
22
24
38
40
42
44
46
48
50
52
54
56
58
16:00
02
04
06
08
10
12
14
16
18
20
22
' 24
26
28
30
32
34
36
38
40
42
44
45 94°W,28°N
46
48
HEADING
TAKEOFF
280°
M
U
II
II
"
170°
tl
"
II
II
II
tl
II
II
II
U
II
II
U
U
II
11
II
M
n
n
n
n
n
»
n
it
n
H
II
U
n
"
n
n
"
it
350°'
II
11
ALTITUDE
(ft)
203
1183
.
3394
.
4108
-
4132
.
-
-
4124
-
4096.
-
4131 '
_
4140
.
4323
-
4128
.
4072
'.
4062
_
4088
*_
4073
-
4062
_
4078
-
4098
_
4080
_
4090
-
4088
-
4077
.
3028
03
(yg/m3)
66
66
30
42
66
66
66
60
30
- 42
48
48
48
54
54
54
54
54
60
48
60
54
60
60
60
54
48
54
48
48
42
48
48
42
42
36
36
30
30
36
30
30
30
30
30
TEMP.
CO
24.7
.
20.9
.
18.8
-
19.5
.
-
-
18.6
_
18.4
..
18.6
_
18.7
_
18.9
-
18.4
,.
18.1
.
18.4
.
18.4
_
18.2
-
18.4
_
18.5
_
18.6
_
18.7
_
18.9
-
18.8
-
18.5
.
18.5
DEWPT.
CO
20.2
_
12.6
.
9. "3
-
5.2
.
-
-
12.3
-
11.3
.
12.0
_
11.1
12.1
-
13.5
-
13.0
.
13.0
_
13.5
-
14.4
-
12.9
-
12.3
-
11.5
*
12.3
-
11.5
-
11.6
-
13.1
_
13.0
MANIFOLD
TEMP.
CO
28.2
.
25.1
_
23.6
-
24.1
.
-
-
22.6
-
22.8
22.9
_
22.7
-
22.8
-
22.5
-
22.7
-
23.1
_
22.8
-
22.9
-
23.0
-
23.2
-
23.3
_
23.0
-
23.6
-
23.4
-
23.3
_
24.3
TRUE
AIR SPEED
(mph)
158
.
140
-
178
-
183
.
-
-
183
_
186
.
185
_
184
_
173
-
183
..
184
_
183
_
183
-
182
-
183
-
183
-
184
_
183
-
180
-
182
-
183
_
209
425
-------
Table E-Ad-1. Tabulated data: sea-breeze flight
(075), 9/19/75 (con.)
TIME
(GMT)
50
52
54
56
58
17:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
18:00
02
04
06
08
10
12
14
17
POSITION HEADING
350°
M
II
It
It
(I
II
II
II
II
II
II
11
II
11
II
II
"
II
II
II
tl
II
II
"
"
"
M
11
II
..
"
11
"
H
SAM RAYBURN RES. 100°
ft
11
II
II
tt
II
11
DRI LANDING
ALTITUDE
(ft)
923
_
460
-
455
.
519
.
473
_
484
—
472
-
513
.
525
_
501
—
933
_
1038
-
503
_
646
.
723
.
734
.
1042
-
870
.
946
-
928
.
928
.
203
03
(wg/m3)
30
42
36
36
41
41
61
61
56
56
51
51
46
46
46
46
46
41
41
47
47
47
68
68
72
83
78
78
83
83
67
57
62
78
78
88
78
62
78
98
109
114
114
-
TEMP.
CO
23.3
_
26.3
-
26.6
.
26.4
.
26.6
_
26.4
.
26.4
-
26.4
.
. 26.5
_
26.4
_
26.7
.
25.9
-
28.0
_
28.3
.
28.1
—
27.6
_
27.1
-
27.4
.
27.0
.
27.0
_
27.0
_
-
DEWPT.
CO
22.5
„
21.6
-
21.4
21.7
,
21.6
„
21.6
_
21.6
-
21.5
.
21.6
.
21.4
—
20.6
.
20.3
-
19.9
*
18.9
.
19.4
.
19.0
.
18.5
-
18.7
_
12.9
.
18.4
_
18.0
_
-
MANIFOLD
TEMP.
CO
27.6
29.6
30.2
30.2
30.3
—
30.3
30.4
-
30.4
30.2
—
30.5
—
30.2
„
30.2
31.3
31.8
31.7
31.7
31.2
31.4
31.1
.
31.0
—
31.1
_
-
TRUE
AIR SPEED
(mph)
199
180
176
179
.
177
—
177
178
-
176
176
•
176
—
166
..
184
183
.
174
173
.
178
_
177
-
176
.
178
.
181
—
180
_
-
426
-------
x
LU
in
n
n
c «
•"£;*•
m
VO
r>.
o
a)
N
a)
-------
g
r§
vD
r^
o
S3
o
00
60
•H
i—i
m
-------
Table E-4e-l.
Tabulated data: sea-breeze flight
(076), 9/19/75
TIME
(GMT) POSITION HEADING
20:17 DRI TAKEOFF
18 " 1980
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
21:00
02
04
05 94°W; 29«N 090°
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
38 92°W; 28°N VERTICAL
41
44
46
49
53
' 57
22:01
04
06
08
n
15
ALTITUDE
(ft)
203
642
-
2123
-
2078
-
2056
-
2031
.
2030
. "
2030
.
2025
-
2026
.
1989
.
2014
.
1979
.
.
2136
-
2137
_
2115
.
2135
-
2135
-
2140
-
2139
_
2115
_
3115
4187
5201
6208
8377
10619
8202
6177
5190
4091
3106
2086
03
10~2
104
109
98
93
98
98
98
87
82
87
87
65
60
49
55
49
55
60
55
60
55
49
55
.
55
55
55
65
76
60
65
71
71
82
87
87
87
87
87
_
86
84
82
93
96
101
104
86
82
90
86
98
TEMP.
CC)
_
29.1
»
25.7
«,
24.7
-
24.2
»
24.0
»
23.1
—
23.0
.
23.2
-
22.4
•
22.5
.
22.6
.
22.7
.
_
22.4
-
22.5
_
22.0
_
22.2
.
22.2
_
22.1
-
21.9
_
22.0
.
20.7
19.2
16.9
15.2
11.9
10.4
10.2
13.1
14.8
17.2
20.5
22.2
DEWPT.
CC)
17.2
_
16.8
_
16.4
-
18.5
.
19.1
.
20.0
.
19.0
.
16.6
-
1V.7
_
19.2
_
20.0
.
19.2
_
_
19.4
-
19.6
_
20.4
_
18.9
.
19.5
.
20.0
-
19.4
_
19.8
10.6
9.5
6.1
5.1
1.2
-3.6
0.6
3.7
5.2
10.0
12.2
19.6
MANIFOLD
TEMP.
CO
33.6
_
30.8
_
29.8
-
29.5
.
28.6
_
29.1
.
28.0
27.6
-
27.2
27.1
.
27.7
.
34.2
*
_
27.4
-
27.5
_
27.2
_
27.4
27.2
.
26.9
-
26.9
27.1
26.0
24.9
23.0
21.6
19.2
17.2
17.6
19.6
20.9
22.5
24.9
26.3
TRUE
AIR SPEED
(mph)
161
.
172
_
179
-
183
_
174
.
181
.
180
179
-
176
175
—
177
T76
_
—
175
-
175
—
180
_
177
176
_
175
-
177
176
180
182
186
182
188
163
209
200
197
195
183
179
429
-------
Table E-4e-l. Tabulated data: sea-breeze flight
(076), 9/19/75 (con.)
TIME
(GMT) POSITION HEADING
22:16 92°W;28°N 360"
10
20
22
24
26
28
30
32
36 LFT LANDING
23:03 LFT TAKEOFF
04 " 360°
06
08
10
12
14
16
18
20 92°W;31°N 262°
22
24
26
28
30
32
34
36
38
40
41 DRI LANDING
ALTITUDE
(ft)
2080
2108
..
2125
..
1378
42
42
2141
2144
2144
..
2168
•-
2150
2151
..
2158
„
2134
2145
1533
203
03
(yg/m3)
104
87
98
98
98
98
104
115
115
-
_
153
142
131
136
136
136
120
120
125
125
109
115
98
104
109
109
109
125
158
-
TEMP.
(*c)
22.2
-
22.7
-
23.3
-
24.0
_
28.9
-
_
23.9
_
24.3
.
24.3
.
24.2
-
24.3
-
24.4
-
24.6
_
24.6
.
24.4
-
24.9
-
DEWPT.
(°C)
18.4
-
17.3
-
13.6
-
18.9
.
19.1
-
.
16.8
-
16.3
_
16.7
.
16.8
-
. 15.7
17.2
-
15.5
_
17.3
.
18.3
-
18.7
-
MANIFOLD
TEMP.
CO-
26.6
-
27.1
-
27.9
-
28.7
_
30.1
-
_
29.2
-
29.2
_
29.1
-
28.7
-
29.1
-
29.0
-
29.1
_
29.2
-
29.2
-
29.7
-
TRUE
AIR SPEED
(mph)
180
-
181
-
183
-
185
.
-
182
-
183
.
179
-
180
-
180
-
178
-
180
_
181
-
181
-
193
-
430
-------
SELECTED EXAMPLES OF THE BOX PATTERN FLIGHTS-
GULF COAST OXIDANT STUDY
E-4F--Flights No. 099, 100 on October 19, 1975.
E-4G—Flight No. 109 on October 30, 1975.
E-4H--Flight No. 110 on October 31, 1975.
431
-------
SCAt.r.: I in - 60 i;ii
nilf: Gi'if
UuAriU K: 200 M5H
LITTLE ROCK
A 1857Z
4
<$>
2000' MSL
central
time zone
2058Z
93°18'U
33°50'N
LOUISIANA
TEXAS
DeRidder (099)
AHZ11Z
(l-P-2)
1732Z
96°19'W
2 \ 1638Z
^ O 'i " 10 I I
29°00'N
GULF OF MEXICO
Figure E-4F-1. Box pattern flights (099-100), 10/19/75.
432
-------
Table E-4f-l. Tabulated data: box pattern flights
(099-100), 10/19/75
TIME ALTITUDE
(GOT) POSITION HEADING (ft)
(Flight 099)
15:51 DRI TAKEOFF
54 " LOW PASS
56 " 175"
58
16:00
02
04 "
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38 93°22'W,29°N 263°
40
42
44
46
48
50
52
54
56
58
17:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
203
253
_
1744
-
1751
-
1721
_
1750
_
1755
.
1761
-
1768
.
1709
_
1757
_
1769
_
1745
-
1771
_
1772
_
1771
^
1766
_
1827
-
1813
.
1825
_
1791
_
1801
.
1846
-
1834
.
1822
_
1835
°33
.
118
125
125
131
131
131
142
147
153
158
174
142
142
147
142
142
142
136
136
131
131
125
131
125
131
136
142
147
142
142
153
164
158
153
153
153
147
174
164
169
1?4
153
169
169
174
164
180 '
174
164
NO
(ug/m3)
.
.
_
.
-
17
_
17
_
17
—
17
_
10
-
6
4
.
-
.
.
..
8
-
0
_
0
_
14
.
17
.
22
-
8
_
9
_
9
.
3
_
0
-
0
.
0
_
0
TEMP.
_
17.0
_
14.4
-
14.5
_
14.5
_
14.3
.
14.3
.
14.2
-
14.4
.
14.4
_
14.6
.
14.5
..
14.5
-
14.8
_
15.1
_
14.9
15.0
.
15.3
-
15.0
„
15.0
.
15.2
.
15.8
_
16.4
-
16.4
_
16.4
_
16.5
DEWPT. MANIFOLD
(°C) TEMP.
«.
4.5
_
2.3
-
0.9
.
1.0
2.2
.
2.0
„
2.5
5.2
4.7
_
5.2
.
7.1
_
7.9
-
8.1
_
7.5
6.8
5.7
.
6.4
5.3
_
5.4
—
5.0
2.8
2.2
-
2.7
1.9
2.4
1 _
32.9
.
32.4
-
.
_
32.2
31.7
.
31.8
—
32.4
31.8
31.9
31.9
.
31.9
_
32.0
-
32.0
„
32.7
_
32.0
31.8
„
31.7
31.8
_
31.5
..
31.5
31.3
-
31.3
-
31.3
31.3
_
-
TRUE
AIR SPEED-
(mph)
—
154
_
188
188
—
187
187
.
187
187
188
185
186
188
„
186
187
_
183
185
185
185
185
187
184
186
184
188
183
183
433
-------
Table E-4f-l. Tabulated data: box pattern flights
(099-100), 10/19/75 (con.)
TIME ALTITUDE
(GMT) POSITION HEADING (ft)
17:32 96°19'W,29*N 350°
34
36
38
40
42
44
46
48
50
52
54
56
58
18:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52 018°
54
57 TERRELL LANDING
1850
_
1821
-
1835
.
1825
.
1867
—
1856
.
1935
-
1837
.
1855-
_
1921
_
1833
_
1838
-
1861
.
1857
.
1865
_
1914
.
1870
-
1821
.
1860
-
1716
_
.
479
°33
(ug/nT)
164
158
164
169
164
164
164
164
153
158
153
164
158
158
153
147
147
153
153
158
158
147
147
147
147
147
142
147
142
147
153
153
153
153
158
158
158
1169
158
164
164
164
-
NO
(ug/m3)
0
.
0
-
0
.
0
.
5
.
6
.
9
10
.
10
.
8
.
9
.
15
-
20
_
20
.
20
.
20
.
4
-
0
.
4
-
6
_
.
.
TEMP.
(°c>
16.1
_
16.2
-
16.5
_
16.8
.
16.8
.
16.9
_
17.1
16.9
.
16.8
_
16.8
.
17.2
_
17.0
-
16.9
.
17.1
.
17.5
.
17.5
_
17.3
-
17.2
_
16.7
.
16.6
m
20.0
.
DEWPT. MANIFOLD TRUE
(°C) TEMP. AIR SPEED
(°C) (mph)
2.7
.
2.1
-
2.3
_
1.1
_
1.7
.
1.7
.
1.7
1.6
.
0.1
_
0.7
.
0.9
1.6
-
1.7
.
1.2
.
1.6
.
0.6
_
1.1
-
0.9
.
1.6
.
1.5
—
1.5
.
32.2
_
31.8
-
_
.
31.9
_
31.2
.
31.0
.
32.1
31.8
.
32,8
_
32.2
.
32.3
.
31.3
*_
31.4
.
31.5
.
31.6
.
32.2
.
33.0
-
32.4
.
31.3
.
31.3
_
31.6
.
188
_
189
-
190
.
181
_
184
.
181
.
182
182
_
182
.
184
.
181
_
181
-
186
^
183
.
181
.
176
_
178
-
182
.
187
_
185
—
104
-
434
-------
Table E-4f-l. Tabulated data: box pattern flights
(099-100), 10/19/75 (con.)
TIME
(GMT) POSITION
(Flight 100)
19:38 TERRELL
40
42
44
46
48
50
52
54
56
58
20:00
02
04
06 96°22'W,
33°50'N
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54"
56
93°18'W,
58 33°50'N
21:00
02
04
06
08
10
12
14
16
HEADING
TAKEOFF
300°
350°
It
II
II
II
II
II
II
II
(1
II
II
080°
II
II
M
II
II
II
II
"
It
ti
II
H
II
M
"
"
II
li
It
"
It
II
II
tl
11
174°
^
it
11
n
M
II
11
II
II
ALTITUDE
(ft)
479
2043
.
1912
-
1847
.
1883
_
1997
_
1832
.
1807
.
1913
.
1833
.
1976
_
1957
.
1901
-
..
.
1976
.
1872
_
1976
_
1972
-
1968
.
1949
.
1800
.
1915
_
1907
-
1911
.
1878
_
1862
°33
(ug/mj)
_
164
169
169
180
180
180
169
169
158
153
153
153
142
136
142
142
136
131
131
131
131
136
125
125
_
125
125
115
115
104
109
104
109
109
98
109
109
98
98
98
98
104
104
109
104
104
131
115
109
NO
(u9/m3)
_
0
.
0
-
0
-
0
_
0
—
0
.
0
.
0
_
0
_
0
.
0
_
0
-
0
.
0
_
0
.
0
.
0
-
0
_
0
_
0
_
0
_
0
-
0
0
,
0
TEMP.
(°C)
_
18.8
.
17.4
-
17.4
.
17.4
.
17.4
.
17.9
.
17.7
.
17.5
_
17.1
_
16.9
.
16.9
.
16.6
-
—
_
16.1
.
15.8
.
16.1
_
15.5
-
14.9
-
15.6
«.
15.6
_
14.9
..
15.7
-
15.3
15.8
_
16.0
DEWPT. MANIFOLD
(°C) TEMP.
rc)
.
3.5
_
0.8
0.9
_
1.0
—
0.8
.
0.8
1.2
.
1.0
_
1.0
0.6
.
0.4
-1.6
.
-
-1.9
_
-3.1
.
-2.8
—
-1.8
-
-4.3
_
-4.5
_
-6.4
_
-4.7
_
-5.3
-
-5.0
-3.7
-6.2
.
31.8
.
31.9
32.2
.
32.1
_
32.9
.
32.9
_
32.3
_
32.6
_
34.2
34.4
.
33.4
.
33.6
.
.
34.8
34.1
,
.
^
32.3
-
.
^
32.8
^
32.0
^
35.8
_
35.6
-
.
^
^
^
.
TRUE
AIR SPEED
(mph)
.
161
.
186
-
187
_
184
—
187
.
183
183
.
183
_
188
_
187
183
187
.
_
186
_
187
186
_
187
-
195
188
_
182
_.
185
_
175
-
185
183
187
435
-------
Table E-4f-l. Tabulated data: box pattern flights
(099-100), LO/19/75 (con.)
TIME
(GMT)
21:18
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
22:00
02
04
07
08
n
ALTITUDE
POSITION HEADING (ft)
It ^
1967
tl _
1860
tl
1841
"
2092
II
1799
II
1824
"
1802
II
1828
II _
1822
"
1829
11
1847
tl
549
DRI LOW PASS 253
II
LANDING 203
°33
(ug/nr)
115
115
109
115
109
115
125
136
158
169
153
131
125
125
125
125
125
125
125
131
131
125
125
125
118
.
-
NO
(ug/m3)
0
_
0
-
0
.
0
_
0
—
0
_
0
-
0
_
0
.
0
.
0
_
0
.
0
-
TEMP.
(°C)
16.5
.
16.7
-
16.5
_
16.3
_
16.5
.
16.7
_
17.0
-
17.3
.
17.3
_
17.5
.
17.5
_
20.3
.
23.6
-
DEUPT. MANIFOLD
(°C) TEMP.
(°C)
-6.6
_
-5.6
-
-3.8
_ _
-3.1
_
-2.6 33.8
.
-0.4
-
-1.2 33.1
-
-0.3 34.2
, -
-0.4 35.0
_
-0.3 35.1
_
1.0
_
2.2
-.
2.0 33.5
-
TRUE
AIR SPEED
(mph)
184
.
180
-
186
_
182
_
187
.
187
_
186
-
187
_
187
_
183
.
183
_
182
.
_
..
436
-------
o
en
o
4J
4-1
cd
o
PQ
t
W
0)
M
§)
437
-------
o
o
(S
o
cfl
0)
M
13
•
V3
o
1-1
p.
CO
o
•H
-------
Table E-4g-l. Box pattern flight (109), 10/30/75
TIME
(GMT) POSITION HEADING
19:08 LCH TAKEOFF
10 " 340°
12
14
16
18
20
22
24
26 DRI LOW PASS
28 " 255°
30
32
34
36
38
40
42
44
46 SILSBEE 178°
48
50 132°
52
54
56
58
20:00 228°
02
04
06
08 312°
10
12
14
16 045°
18
20
22
24 178°
26
28
30
32 " .
34 S.W. BEAUMONT VERTICAL/1780
36
38
40 " 0
42 178°
44
46 038°
ALTITUDE
observed
(ft)
16
_
.
1000
11
II
11
II
II
253
ii
11
11
11
ii
H
n
H
11
n
n
11
"
11
"
n
11
11
n
"
ii
n
11
11
11
n
11
11
"
"
•"
"
n
it
2000
3000
5000
_
1000
°33
(ug/m )
100
98
98
98
98
98
100
100
96
98
108
108
104
100
100
98
98
100
100
100
100
92
98
116
108
140
144
190
206
256
156
156
120
112
116
116
112
104
100
128
152
174
236
124
118
116
_
-
TEMP.
observed
(°C)
20.0
-
-
20.0
_
.
_
.
24.0
^
-
.
.
-
_
.
.
_
20.5
.
20.5
-
.
-
.
19.5
.
.
-
.
20.5
-
_
20.5
_
-
_
20.5
-
_
-
_
.
20.0
18.0
13.0
.
20.0
DEWPT.
(°0
15.7
-
13.0
-
12.1
.
8.0
.
7.1
_
8.2
.
12.6
-
13.7
-
14.8
.
15.0
.
15.0
-
15.8
-
15.8
15.7
.
16.1
—
16.3
-
16.9
15.6
_
15.8
15.6
_
16.2
_
16.5
11.9
10.9
6.2
.
-
TRUE
AIR SPEED
observed
(mph)
179
-
-
173
_
-
.
.
180
_
.
-
-
-
_
.
.
_
179
_
180
.
-
-
_
178
-
-
-
_
180
.
.
178
_
.
.
176
-
.
-
_
-
170
165
166
.
.
190
439
-------
Table E-4g-l. Box pattern flight (109), 10/30/75 (con.)
TIME
(GMT)
20:48
50
52
54
56
58
21:00
02
04
06
08
10
12
14
16
18
20
22
24
26
28
30
32
34
36
37
38
40
42
44
46
48
50
52
54
56
53
22:00
02
04
06
08
10
12
14
16
18
20
22
24
26
POSITION HEADING
it
"
ii
M
II
II
II
II
312°
II
II
11
11
II
II
II
«
SUsbee 225°
H
"
..
11
n
n
132°
It
II
11
II
II
II
II
II
358°
II
II
II
II
II
II
II
II
II
II
II
SILS8EE 075°
II
"
II
M
ALTITUDE
observed
(ft)
II
II
II
II
II
II
II
II
II
II
«
II
II
II
II
II
«
II
II
II
II
II
II
II
II
II
M
II
11
II
II
II
II
II
II
11
II
II
II
11
II
"
II
II
II
II
II
II
II
II
II
(uS-3)
186
156
164
186
128
_
124
156
198
116
116
100
120
104
100
104
100
104'
108
112
112
112
104
100
104
_
100
108
116
144
170
178
280
198
280
248
226
194
186
186
148
116
100
100
100
104
104
104
108
116
116
TEMP.
observed
.
.
.
-
_
19.5
_
20.5
-
.
.
.
.
21.0
.
_
_
_
21.5
.
22.0
.
_
.
.
21.0
.
-
.
-
_
.
_
20.0
-
—
22.0
.
_
-
_
.
.
22.0
-
22.0
.
.
.
-
DEWPT.
.
.
_
-
15.5
-
16.1
.
16.2
_
15.5
_
15.5
-
15.3
_
14.7
.
15.5
„
15.1
.
16.8
.
_
16.9
.
16.8
.
16.8
_
15.8
»
16.1
-
16.3
'
16.8
.
16.3
_
15.6
.
15.2
-
14.1
,
13.3
.
6.0
TRUE
AIR SPEED
observed
(mph)
.
-
.
-
_
180
_
183
-
_
.
.
.
179
_
_
.
.
181
„
180
.
.
.
—
178
.
-
.
-
.
.
.
180
-
_
180
.
.
-
•
-
-
181
-
180
-*
.
.
-
440
-------
Table E-4g-l. Box pattern flight (109), 10/30/75 (con.)
TIME
(GMT)
22:28
30
32
34
36
38
41
ALTITUDE
POSITION HEADING observed
(ft)
II II
II II
II II
II II
II II
DRI LOW PASS 253
LANDING 16
°33
(ug/nr)
112
112
112
112
112
108
.
TEMP.
observed
(°C)
22.0
_
.
22.0
24.0
-
TRUE
DEWPT. AI« SPEE°
observed
( C) (mph)
7.0 183
-
5.9
185
6.7 183
.
Oxides of nitrogen below the minimum detectable, concentration of the analyzer.
441
-------
CO
O
4J
•a
rv| 01 rg
442
-------
Table E-4h-l. Tabulated data: box pattern flight
(110), 10/31/75
TIME
(GMT) POSITION
17:18 DRI
20
22
24
26
28
30
32
34
36
38
40
42
44
46
47 LCH
18:37 LCH
38
40
42
44
46
48
50
52
54
56
58
19:00
02
04
06
08
10
12
14
16
18
20
22
24 94°261W,31042'N
26
28
30
32
34
36
38
40
42
HEADING
TAKEOFF
.
LOW PASS
.
-
165°
tl
II
II
II
II
II
II
II
LANDING
TAKEOFF
.
322°
II
II
II
II
H
It
II
II
II
II
II
II
II
H
II
II
II
11
II
tl
II
230°
II
II
II
It
»
II
II
II
II
ALTITUDE
observed
(ft)
203
600
253
1000
II
It
II
H
II
II
tl
II
II
.
-
16
16
.
1000
II
It
II
II
II
II
II
11
II
11
II
II
1200
II
II
II
II
1500
II
II
II
II
II
II
II
II
"
II
11
II
II
°33
104
104
104
100
104
104
104
104
112
120
128
120
132
128
130
•
—
138
178
174
174
182
154
224
202
152
.
132
132
136
132
126
126
130
126
122
120
120
120
116
116
120
124
124
128
140
128
132
136
140
TEMP.
observed
(S0
19.0
-
20.5
-
-
_
18.0
-
-
-
20.0
-
-
.
-
~
_
.
.
19.5
_
-
-
-
-
.
-
-
21.0
-
_
21.0
-
-
-
_
20.5
-
-
-
—
21.0
-
21.0
-
.
-
-
-
-
°-
-
5.4
-
4.0
_
5.5
-
5.3
-
7.5
-
8.7
-
10.2
—
_
-
-
2.1
.
10.0
-
8.5
-
8.0
-
6.6
-
6.0
.
3.7
-
4.6
-
5.9
-
1.4
-
4.8
_
3.9
-
3.8
-
4.8
-
6.6
-
7.6
TRUE
AIR SPEED
observed
(mph)
176
180
-
-
_
182
-
-
-
184
-
-
-
-
"
_
-
-
171
-
-
-
-
-
-
-
-
176
-
.
178
-
-
-
.
177
-
-
-
_
180
182
-
.
-
-
-
-
443
-------
Table E-4h-l. Tabulated data: box pattern flight
(110), 10/31/75 (con.)
TIME
(GMT) POSITION
19:44
46
48
50
52
54
56
58
20:00
02
04
06
08
10
12
14
16
18
20
22
24
25
28
30
32
34
36
38
40 „
42 97°41'W,29051'N
44
46
48
50
52
54
56
58
21:00
02
04
06
08
10
12
14
16
18
20
22
HEADING
..
II
II
II
'*
11
II
H
II
II
II
II
II
II
"
II
11
II
II
11
H
II
II
II
II
II
II
11
II
140°
II
II
II
II
"
II
II
"
11
II
II
II
II
il
II
II
"
II
II
II
ALTITUDE
observed
(ft)
11
II
II
II
II
II
II
II
II
"
II
II
II
"
II
II
II
II
II
II
II
It
II
11
II
II
II
II
"
11
II
It
II
II
II
II
II
II
II
11
11
II
II
"
tl
11
1000
II
II
II
(U9/.3)
136
136
136
148
148
148
164
152
168
188
222
266
200
172
156
152
148
152
148
140
124
124
128
132
132
132
136
132
128
116
120
120
124
128
128
128
128
128
132
132
128
128
124
128
128
124
124
124
128
128
TEMP.
observed
.
21.5
.
.
-
_
.
.
22.5
-
—
-
.
.
-
_
'23.0
.
.
-
.
.
_
23.5
-
_
-
.
_
24.0
_
23.5
_
.
-
.
-
_
22.0
-
_
-
-
-
-
_ .
23.0
-
-
-
DEWPT.
(bc)
_
-
.
9.5
-
10.2
.
8.9
-
10.2
_
12.2
.
12.4
-
12.5
.
13.4
.
13.9
_
14.'2
_
13.6
-
13.5
.
13.7
.
14.1
—
14.4
.
14.7
-
14.2
.
14.1
-
14.7
.
15.0
-
15.4
-
14.9
-
14.7
-
15.5
TRUE
AIR SPEED
observed
(mph)
.
181
-
-
-
_
-
.
180
-
_
-
.
.
-
.
178
-
-
-
.
-
-
182
-
.
-
-
.
182
_
181
-
-
-
.
-
-
180
-
.
-
-
-
-
.
183
-
-
-
444
-------
Table E-4h-l. Tabulated data: box pattern flight
(110), 10/31/75 (con.)
TIME
(GMT) POSITION
21:24
25
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
22:00
02
04
06
07 GALVESTON
22:29 GALVESTON
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
23:00 93042'W,29052'N
02
04
06
08
10
12
14
16
HEADING
II
050°
II
II
H
H
II
H
H
H
H
II
H
H
H
II
II
II
H
II
"
II
II
LANDING
DPR.
050°
II
II
U
II
II
U
H
H
H
II
H
II
II
II
280°
II
II
II
H
II
H
II
II
ALTITUDE
observed
(ft)
II
II
II
H
H
II
II
H
II
II
H
II
H
II
II
II
II
II
II
II
II
II
II
-
1000
II
II
II
II
II
II
H
II
It
H
II
II
II
II
II
II
II
II
II
II
II
II
II
°33
—
_
128
128
136
144
144
148
144
140
144
144
148
148
144
144
148
148
148
148
152
152
156
-
128
148
148
144
148
160
160
156
156
152
152
152
152
152
144
144
148
152
148
152
162
174
174
166
TEMP
observed
—
_
23.0
.
22.5
-
_
.
22.5
.
-
_
22.0
.
.
-
_
.
.
21.5
-
—
.
-
.
20.5
.
-
m
.
.
.
20.5
_
.
.
.
-
.
20.0
.
.
-
_
.
.
.
20.5
OEWPT.
_
.
15.4
_
14.6
-
14.0
_
14.6
_
13.3
_
12.0
.
10.6
-
15.2
.
14.8
.
13.8
.
14.0
-
33.1
.
975
-
12.0
.
11.7
.
11.9
—
11.6
.
11.6
-
11.7
11.4
.
10.7
.
10.7
.
8.1
-
TRUE
AIR SPEED
observed
(mph)
—
_
180
.
178
-
_
.
181
.
-
.
181
.
.
-
^
.
.
_
-
.
.
*
.
178
.
-
w
.
,
.
180
_
.
.
.
-
_
180
.
.
-
_
.
.
.
181
445
-------
Table E-4h-l. Tabulated data: box pattern flight
(110), 10/31/75 (con.)
TIME
(GMT)
23:18
20
22
24
26
28
30
32
34
36
38
40
41
ALTITUDE
POSITION HEADING observed
II H
II H
DRI LOW PASS 253
160° 1000
ii ii
H it
H
ii
ii
H
H
-
LCH LANDING 16
°33
182
166
142
166
178
182
170
174
174
170
166
166
-
TEMP
observed
.
.
22.0
20.5
-
_
21.0
-
-
-
20.0
-
-
DEWPT.
7.1
-
6.0
7.6
-
8.0
-
5.1
.
7.1
_
11.6
-
TRUE
AIR SPEED
observed
(mph)
_
-
190
178
-
.
181
-
-
-
180
-
-
Oxides of nitrogen below the minimum detectable concentration of the analyzer.
446
-------
APPENDIX F
BACKGROUND DATA AND EMISSION STUDY
FOR TEXAS GULF COASTAL AREA
F-l. Background Data and Emission Study for Texas Gulf Coastal Area
A partial analysis of historical data was initiated using ozone
data for Nederland, Texas from July 1 to September 30, 1972 and for
Houston, Texas from August 1 to September 30, 1972 as reported by
Johnson, et al.-'. Meteorological data from the Local Climatological
Data for Jefferson County Airport and Houston Intercontinental Airport
and from the Daily Weather Map, Weekly Series were examined. The
Nederland ozone station is located at the Jefferson County Airport.
The Houston ozone station is located in the Houston Ship Channel area,
about 29 km south of the airport. Concentrations of ozone less than
3
100 yg/m were not reported. During the measurement periods, the ozone
exceeded NAAQS on 43 percent of the days at both locations.
The resultant wind direction (i.e., net transport wind direction)
for those days when the ozone exceeded the NAAQS (160 yg/m ) for at
least one hour were compared with the occurrences of the resultant wind
direction in all circumstances. The data presented in figure F-l seem to
indicate that the relatively infrequent occurring northwest quadrant
daily resultant wind is disproportionately associated with high ozone
concentrations at both Houston and Nederland. In the summer of 1972,
these winds—west (250°-290C), northwest (300°-330°), and north (340°-
020°)—occurred only 10 percent of the days at Houston and 15 percent
of the days at Nederland. Table F-l ranks all maximum daily one hour
3
average ozone concentrations; above 300 yg/m at the two sites. The
underlined values indicate days when there was a northwest quadrant
resultant wind. There was no consistent association of wind speed with
ozone concentration.
During the respective study periods, the following conditions were
found to exist at the respective locations:
448
-------
NEDERLAND
(July, August, September 1972)
100% 100%
43% of Days studied,
On exceeded 1 hour STD
64%
50%
N
NE
41%
I
36%
_~
SE
Wi NO 6 11 17 22
n =«• 1 6 Q o 3 7 7 a
3 max >tg/ni3
501,
21%
S
24
5
SW
!
W
NW
i 8 2 -(92)
2 6 2 -(40)
HOUSTON
(August, September 1972)
43% of days studied,
0, exceeded 1 hour SID
TOO/ iOO!
r~~? r'"i
50%
60%
27"!,
-• *-" " B
n i
:i
ti
I I
A!
!
-»
H
"3na>
W!»IO
HO
4
1
10
S
Figure F--1. Ratio of daily resultant wiita ctirHcLJon ocriiv>-f-ii'.>os w'.
ozone concentre f." CD ex«. eed"^ NA.AO^ !.•' all nf--ii.rv,- .(_ -••-
the daily re£u]tant win,! ri ~ i '.-•<-,; jo'-,
449
-------
Table F-l. Ranking of maximum daily-ozone concentration exceeding
300 yg/ni
RANK
1)
2)
3)
4)
5)
6)
7)
8)
9)
10)
11)
12)
13)
14)
15)
16)
17)
18)
19)
20)
21)
NEDERLAND
715*
615
600
500
425
410
405
390
380.
375
365
360
355
355
335
325
320
315
305
300
300
(9/9)**
(8/5)
(7/25)
(9/21)
(8/20)
(8/21)
(8/6)
(8/22)
HOUSTON
715
515
450
445
445
425
405
375
365
345
315
(8/5)
(9/21)
(9/3)
Underlinedx.conceatrations associated with NW resultant winds.
**
Date of Occurrence.
450
-------
A. NEDERLAND:
1) Seventy-nine percent of the time a NW quadrant resultant
wind occurred and the maximum one hour ozone concentration
o
exceeded 160 ug/m .
2) Fifty-seven percent of the time a NW quadrant resultant
wind occurred and the maximum one hour ozone concentration
o
exceeded 300 yg/m .
3) The maximum daily one hour average ozone concentration exceeded
3
300 yg/m twenty-one times, eight days of which (38 percent)
had a NW quadrant resultant wind.
4) The maximum daily one hour average ozone concentration exceeded
400 yg/m seven times, three days of which (43 percent) had a
NW quafdrant resultant wind.
5) The following six days had NW quadrant resultant winds and maximum
ozone less than 300 yg/m .
8/15 270 yg/m3
7/29 260 yg/m3
9/5 205 yg/m3
*3
7/5 and 9/30 <100 yg/m (following the only cold
fronts which passed
during the study).
8/27 Missing Data
B. HOUSTON:
1) Eighty-three percent of the time a NW quadrant resultant
wind occurred and the maximum one hour ozone concentration exceeded
160 yg/m3.
2) Fifty percent of the time a NW quadrant resultant wind
occurred and the maximum one hour ozone concentration exceeded
300 yg/m3.
3) In August and September, the maximum daily one hour average ozone
concentration exceeded 300 yg/m eleven times, three days of which
(27 percent) had a NW quadrant resultant wind.
451
-------
4) In August and September, the maximum daily one hour average
3
ozone concentration exceeded 400 yg/m seven times, three days
of whicn (43 percent) had a NW quadrant resultant wind.
5) The to-lowing three days had NW quadrant resultant winds and
maximum ozone less than 300 yg/m~:
8/27 290 yg/m3
8/6 190 yg/m3
9/30 < 100 yg/m3 (frontal passage)
The maximum hourly ozone concentrations were compared on those days
3
when both locations reported more than 100 yg/m concentrations. The
data are shown In figure F-2. The least squares regression gives the
Nederland (NED) concentration as a function of the Houston (HOU) concen-
tration as:
NED - 0.6 (HOU) -I- 155 yg/m3
The linear correlation coefficient is 0.61. These two results verify
the impression of visual inspection of the data that the maximum Houston
and Nader Land oy.oiie coneer trat Lous are not well correlated.
An Inventory of available hourly ozone, concentrations in the 3AROAD
data file v/as obtained,. The inventory contained only measurements made
b/ tut s. he-ir; ! j-,unescpnt method in the states of Alabama, Mississippi,
Arkansas, Loi'JsL?na and Texas.
The da'.a \«*e examined for the date and time of ozone concentrations
grca: er *• !•)£•> '-.'i? NAAQS, for apparent relationships among several nearby
Rfjt'ors nr.' r--.r- su.r-h occurrences over widespread areas. Historic
surfsv (r ••-s .
While the data were interesting, it was extremely difficult to relate
the oione measurements at a station among stations, or to weather conditions
isi a systematic manner. Research efforts were redirected to activities
ot (.'tie fie]-! research programs.
452
-------
Dates: August & September, 1972
Conditions: Maximum 24 hour ozone concentrations exceeding 100
micrograms per cubic meter at both Houston and
Nederland sites.
800|-
700-
600-
500-
_s
oo
o
4J
CO
I
400-
300-
200 _
NED = 0.6 (HOU) + 155
r = 0.61
100
100
200
Figure F-2.
300 400
Nederland
500
600
700
800
Scatter diagram of maximum daily ozone concen-
trations (>100 yg/m3) at Nederland and Houston,
August, September, 1972.
453
-------
F.2. Hydrocarbon Emission Data
Annual average hydrocarbon emission data (tons/year) for each county
of each state bordering or east of the Rocky Mountains were requested
from the NEDS data file. A priority was assigned at RTI to the needed
data. The lowest was assigned to those states where analyses of hydro-
carbon data had been done previously. A higher priority was given to
those states, outside of the primary study area (e.g., South Carolina,
Georgia) that had not been analyzed. The highest priority was given to
the remaining states. The annual emissions by county and annual emission
density (tons per square mile) were "computed for the two higher priority
states. The logarithm of those values were computed and plotted. A
manual analysis of the resulting distribution was made at unit increments
of the logarithm, i.e., at order of magnitude intervals of emission or
emission density. The emission density map is available from RTI through
the EPA Project Officer.
The emissions by county in Ohio, as tabulated and plotted previously,
were compared with the latest data from the NEDS files. The regression
equation is
E?0 = 1.155 E?5
where E7» and E7,. are the 1970 and 1975 emissions, respectively. The
linear correlation coefficient is 0.998. The reduced emissions in the
1975 data occur primarily in the six largest population centers. In the
remainder of the counties, the emission changes are minimal. The high
correlation coefficient and the small changes of emissions suggested that
reanalysis of the data in the previously analyzed states was unnecessary.
F.3. Reference
1. Johnson, C. E., D. J. Johnson and R. R. Wallis, "Ozone Concentrations
on the Upper Texas GulE Coast, July, August, September 1972", Air
Quality Evaluation Program, Texas State Department of Health, Austin,
Texas, March 1973.
454
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/3-76-033
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Formation and Transport of Oxidants Along Gulf
Coast and in Northern U.S.
5. REPORT DATE
August, 1976
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
Research Triangle Institute
9. PERFORMING ORG "\NIZATION NAME AND ADDRESS
Research Triangle Institute
Research Triangle Park, North Carolina
10. PROGRAM ELEMENT NO.
2AH137, 2AE137, 2AC129
11. CONTRACT/GRANT NO.
27709
68-02-2048
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Monitoring and Data Analysis Division
Research Triangle Park, North Carolina 27711
13. TVPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
This publication reports on two concurrent sets of field measurements of
ozone and precursors which were being conducted in separate regions of the United
States from July 1 - October 31, 1975.
The first set of measurements spanned the northern portion of the United States
from Montana to Pennsylvania. Three continuously operated ground stations (Wolf
Point, Mt.; Creston, la.; Bradford, Pa.) were used to monitor ambient levels of ozone
NO/NOX and 24-hour suspended particulate levels plus the analysis of bag samples for
organic pollutants. The main objective of the northern study was to determine the
extent to which ozone buildup under conducive meteorological conditions differs in
large areas having low and high precursor emission densities.
The second set of measurements were taken along the Gulf Coast, primarily in
Texas and Louisiana. Continuous monitoring of ozone, NO/NOX and 24-hour total sus-
pended particulates plus collection of bag samples for analysis of organic pollutants
was performed at a rural site near DeRidder, Louisiana. The objectives of the
southern set of measurements were to document the extent to which ozone levels exceed
the Federal ambient standard in this region and to assess the relative importance
of long-range transport and local synthesis in determining the high levels of ambient
ozone observed near several Texas cities.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Photochemical Air Pollutants
Precursors
Ozone
Nitrogen Oxides
Organic Pollutants
Atmospheric ozone levels.
Ozone formation and
transport related to
precursor levels and
weather conditions.
Atmospheric
Photochemistry/
Air Pollution
13. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
481
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
455
«US GOVERNMENT PRINTING OFFICE 1976 641-317/5534
------- |