United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-79-039
December 1979
Air
Ozone and Precursor
Transport Into
an Urban Area
Evaluation of Measurement
Approaches
-------
EPA-450/4-79-039
Ozone and Precursor Transport
Into an Urban Area
Evaluation of Measurement Approaches
by
Michael W. Chan, Douglas W. Allard,
and Ivar Tombach
AeroVironment Inc.
145 Vista Avenue
Pasadena, California 91107
Contract No. 68-02-3027
EPA Project Officer: E.L. Martinez
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
December 1979
-------
This document 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 (MD 35), U. S.
Environmental Protection Agency, Research Triangle Park, NC 27711;
or, for a fee, from the National Technical Information Service,
5285 Port Royal Road, Springfield, VA 22161.
This report was furnished to the Environmental Protection Agency
by AeroVironment, Inc., 145 Vista Avenue, Pasadena, CA 91107, in
fulfillment of Contract No. 68-02-3027. The contents of this report
are reproduced herein as received from AeroVironment, Inc. The
opinions, findings and conclusions expressed are those of the author
and not necessarily those of the Environmental Protection Agency.
Publication No. EPA-450/4-79-039
-------
ABSTRACT
This report evaluates five techniques for measuring the transport of ozone and
precursors into an urban area. These techniques were tested in Philadelphia during the
summer of 1978. The data collected in the field program indicate that, in general,
advection of ozone aloft is the main route by which pollution of photochemical
interest is transported into Philadelphia. Transport of ozone along the surface and
transport of oxides of nitrogen and non-methane hydrocarbons, both aloft and along
the surface, are minimal. Thus, the recommended techniques must primarily be able
to quantify the ozone transported aloft. Of the five techniques, three were
determined applicable for quantifying the ozone transported aloft. These three
techniques are: (1) fixed surface monitoring upwind; (2) a dedicated instrumented
aircraft; and (3) a free lift balloon ozonesonde system. The selection of one of these
techniques by an air pollution control agency would depend on the agency's technical
expertise, funding available, and the intended use of the data.
111
-------
EXECUTIVE SUMMARY
The objective of this study was to evaluate and demonstrate approaches for
measuring the transport of ozone and precursors into an urban area. To achieve this
objective, five measurement techniques were tested in Philadelphia during the summer
of 1978. These five approaches were:
1. Surface measurements at fixed sites;
2. Airborne measurements by dedicated instrumented aircraft;
3. Airborne measurements by a portable instrument package in a light
aircraft;
4. Soundings by ozonesonde beneath a free lift balloon; and
5. Soundings by ozonesonde beneath a tethered balloon.
From the data collected by these various techniques we obtained certain useful
information on the transport of ozone and precursors into Philadelphia, including:
1. From the fixed sites located upwind for this study, only minimal transport
of ozone, oxides of nitrogen, or non-methane hydrocarbons was observed
along the surface into Philadelphia during the early morning hours.
2. Transport of oxides of nitrogen and non-methane hydrocarbons aloft is
slight.
3. Transport of ozone aloft, in excess of .08 ppm, was sometimes observed
during transport days.
IV
-------
4. There is no significant difference in ozone concentrations laterally along the
upwind boundary of Philadelphia in the air mass aloft. Differences have been
observed at surface stations, probably due to variabilities in the strengths of
ozone scavengers, and in the strengths of sources of precursors in the vicinity
of the monitoring stations.
5. Trajectories on photochemically active days with strong transport* indicate
that the incoming air mass generally approaches Philadelphia from the west in
a broad, anticyclonically curved path and does not routinely traverse any
particular urban area upwind.
Of the five measurement techniques evaluated, surface measurements at fixed sites,
airborne measurements by dedicated instrumented aircraft, and soundings by ozonesonde
beneath a free lift balloon are all viable means of measuring directly or providing useful
information on ozone concentrations aloft.
An air pollution control agency, in selecting the technique to use, should consider
factors such as its technical capability, funding available, and the intended use of the
data. The table on the following page ranks the three acceptable approaches in terms of
cost, technical expertise required to operate, and the uses of the data generated, and
serves to guide the agencies in making the selection.
^criteria for such days are given on pp. 63-64.
-------
UJ
z
O
N
O
z
o
z
o
H
o
a.
Z
z
z
cQ
O
a;
O
a
z
5
D
(T5
<
al
O
u-
LT)
LU
X
U
<
O
a;
a,
a,
,
D "J
O
uj (J
td
to
Q
TP
0)
> 0,
Sri*
U •;-;
.-i +j
c u
o
O
.t: o
c U
1
+.«
0)
S
XXX
XXX
o
o
o
o
""N
(N
60
s
O
(N
ft
0
O
TJ
3
O
O
oo
o
o
o
ja
o
o
o
o
o
o
o
•o
-a
W5 0)
CO
2
o
N
O
s
E
-------
ACKNOWLEDGEMENTS
Many people at AeroVironment Inc., besides the authors, contributed signifi-
cantly to this study. Other primary participants and their areas of responsibility are:
Mr. Steve Fisher, Field Operations
Mr. Don Gonzales, Data Processing
Dr. Andrew Huang, Quality Assurance
Mr. Martin Ledwitz, Data Analysis
Mr. David Pankratz, Field Operations
Ms. Diane Barker, Report Editing
Mrs. Kay Vessels, Technical Typing
Mrs. Darlene Asamura, Technical Typing
Ms. Gloria Best, Technical Illustrating
Dr. Edgar Stephens of the University of California at Riverside also provided
valuable input into the hydrocarbon data analysis and report.
A number of other organizations were also involved in this study. EPA Environ-
mental Monitoring Systems Laboratory, Las Vegas (EMSL-LV) supplied an instrumented
helicopter. Mr. Charles Fitzsimmons was responsible for EMSL-LV operation, while
Mr. Frank Johnson provided the weather forecasts for making operational decisions.
The helicopter measurements were made by Northrop Services Environmental Science
Center under the supervision of Mr. Calvin Hancock. Hydrocarbon species identifi-
cation was performed by Scott Environmental Technology, Inc. (SET). The project
manager at SET was Mr. Anthony F. Souza. Research Triangle Institute (RTI), under
separate contract to EPA, audited the instruments at the surface monitoring stations
as well as on the helicopter. Mr. David Pasquini, RTI, was primarily responsible for
accomplishment of the audits.
vu
-------
Mr. Tom Weir, Philadelphia Department of Public Health, provided information
on Philadelphia air quality in general. Other agencies that provided data for this study
included the Pennsylvania Department of Environmental Resources and the New
Jersey Department of Environmental Protection.
Five individuals deserve to be mentioned for allowing the use of their property for
equipment placement during the study. These individuals are:
Mr. W. Laurence Bicking, New Castle Department of Parks and Recreation
Mr. William Deck, Hillsborough Golf Resort and Country Club
Mr. Lane R. Jubb, Valley Forge Aviation, Inc.
Mr. Thomas Pankok, Salem County Freeholders
Mr. Robert Shannon, Shannon Aviation Corporation
In addition, the following EPA personnel, listed alphabetically, provided helpful
guidance and assistance during the course of the study:
Mr. William Belanger, EPA, Region III, Philadelphia, Pennsylvania
Mr. Edward Hanks, EPA, Research Triangle Park, North Carolina
Mr. E. L. Martinez, EPA, Research Triangle Park, North Carolina
Mr. Norman Possiel, EPA, Research Triangle Park, North Carolina
via
-------
TABLE OF CONTENTS
ABSTRACT iii
EXECUTIVE SUMMARY iv
ACKNOWLEDGEMENTS vii
1. INTRODUCTION 1
2. REVIEW OF METHODS FOR MEASURING OZONE AND PRECURSOR
TRANSPORT 3
2.1 Discussion of Measurement Techniques 3
2.1.1 Method A: Surface Measurements, Fixed Sites 4
2.1.2 Method B: Surface Measurements, Moving Vehicle 5
2.1.3 Method C: Tower Measurements 7
2.1.4 Method D: Airborne Measurments, Dedicated
Instrumented Aircraft 9
2.1.5 Method E: Airborne Measurements, Portable Instrument
Package 13
2.1.6 Method F: Airborne Measurements, Manned Balloon 16
2.1.7 Method G: Free Lift Balloon Soundings 16
2.1.8 Method H: Soundings by Tethered Balloon 18
2.1.9 Method I: Remote Sensing 20
2.2 Comparative Evaluation 22
3. FIELD PROGRAM TO EVALUATE SELECTED MEASUREMENT TECHNIQUES 25
3.1 Objectives of the Field Program 25
3.2 Program Design 25
3.3 Surface Measurements at Fixed Sites 26
3.3.1 Sites 1, 3, and 3 29
3.3.2 Site 4 29
3.3.3 Site 5 31
3.3.4 Lancaster and Bivalve 31
3.3.5 Ancora 31
3.3.6 Franklin Institute and South Broad Street 31
3.3.7 Somerville 31
3.4 Airborne Measurements by Dedicated Instrumented Aircraft 32
3.5 Airborne Measurements by a Portable Instrument Package 37
3.6 Soundings by Ozonesonde Beneath a Free Lift Balloon 40
3.7 Soundings by Ozonesonde Beneath a Tethered Balloon 44
3.8 Hydrocarbon Sampling for Species Analysis 47
IX
-------
3.9 Quality Assurance 49
3.9.1 Surface Monitoring Program 49
3.9.2 Airborne Measurement Program 53
3.9.3 Audits 54
4. CHARACTERIZATION OF OZONE AND PRECURSOR TRANSPORT INTO
PHILADELPHIA 55
4.1 Average and Maximum Levels 55
4.2 Transported Levels of Ozone and Precursors 63
4.3 Vertical Profiles 64
4.4 Horizontal Gradients 74
4.5 Diurnal Variations 80
4.6 Case Studies of Ozone and Precursor Transport 88
4.6.1 August 22, 23, 24 88
4.6.2 August 19 98
4.7 Hydrocarbon Data 106
4.7.1 Data Validity 106
4.7.2 Analysis of Continuous Hydrocarbon Monitoring 107
4.7.3 Hydrocarbon Species Analysis 109
5. PERFORMANCE EVALUATION OF SELECTED MEASUREMENT
TECHNIQUES 116
5.1 Surface Monitoring Network 117
5.2 Airborne Measurements by Dedicated Instrumented Aircraft 122
5.3 Airborne Measurements by a Portable Instrument Package 122
5.4 Soundings by Ozonesonde Beneath a Free Lift Balloon 126
5.5 Soundings by Ozonesonde Beneath a Tethered Balloon 131
6. COST EVALUATION OF SELECTED MEASUREMENT TECHNIQUES 135
7. RECOMMENDATIONS ON MEASUREMENT TECHNIQUES 138
8. REFERENCES 140
APPENDIX A - Northeast Oxidant Transport Study Audit Data 144
-------
LIST OF FIGURES
Figure Description
1. Typical flight pattern for aircraft surveys. 11
2. Locations of monitoring sites. 27
3. Monitoring station at Site 1. 30
4. Dedicated EPA-Las Vegas instrumented UH-1H helicopter. 33
5. Typical flight pattern followed during 0400 EOT flight. 35
6. Typical late morning flight pattern. 36
7. Cessna 172 aircraft used for flying the portable instrument package. 38
8. Portable instrument package and recorder strapped to passenger's 39
seat of aircraft.
9. Mast 730-9 ozonesonde. 42
10. Atmospheric Instruments Research Company ozonesonde ground 43
station.
11. Tethered balloon ozonesonde system. 46
12. Ozone wind rose for Site 1. 58
13. Ozone wind rose for Site 2. 59
14. Ozone wind rose for Site 3. 60
15. Ozone wind rose for Site 4. 61
16. Ozone wind rose for Site 5. 62
17. Vertical profiles of ozone, oxides of nitrogen, and temperature 67
obtained by the instrumented helicopter over Site 1 at 0517 EDT
on August 24, 1978.
18. Vertical profiles of ozone, oxides of nitrogen, and temperature 68
obtained by the instrumented helicopter over Site 1 at 1006 EDT
on August 24, 1978.
19. Vertical profiles of ozone, oxides of nitrogen, and temperature 69
obtained by the instrumented helicopter at 1617 EDT over Site 1
on August 24, 1978.
XI
-------
Figure Description Page
20. Vertical profiles of ozone, temperature, wind speed, and wind 71
direction over Site 1 on a photochemically active day with little
observed transport.
21. Vertical profiles of ozone, oxides of nitrogen, temperature, and 72
wind obtained by the instrumented helicopter at 0430 EOT over
Site 2 on August 21, 1978.
22. Vertical profiles of ozone, oxides of nitrogen, and temperature 73
obtained by the instrumented helicopter at 0911 EOT over
Site 2 on August 21, 1978.
23. Vertical profiles of ozone and temperature obtained by the 75
instrumented helicopter over Site 1 at 0455 EOT on August
19, 1978.
24. Vertical profiles of ozone and temperature obtained by the 76
instrumented helicopter over Site 2 at 0425 EOT on August
19, 1978.
25. Vertical profiles of ozone and temperature obtained by the 77
instrumented helicopter over Site 3 at 0534 EOT on August
19, 1978.
26. Diurnal variation of ozone at Sites 3 and 5 and the South Broad 82
Street Site on a day of strong observed transport (August 24, 1978).
27. Diurnal variation of ozone at Sites 1 and 4 and the South Broad 84
Street Site on a day of observed photochemical activity but low
transported levels of ozone into the study area (August 2).
28. Diurnal variation of ozone at Sites 3 and 5 and the South 85
Broad Street Site on a day of little observed photochemical
activity (August 21).
29. Synoptic situation, 0700 EDT, August 23, 1978. 89
30. Estimated 36-hour air parcel trajectories beginning 0200 EDT 91
on August 21, 22, and 23 and ending 1400 EDT on August 22,
23, and 24, respectively.
31. Surface trajectories for three air parcels located within the study 92
area on August 22, 1978.
32. Surface trajectories for three air parcels located within the study 93
area on August 23, 1978.
33. Surface trajectories for three air parcels located within the 94
study area on August 24, 1978.
Xll
-------
rigure Description Page
34. Vertical profiles of ozone and temperature obtained by the 95
instrumented helicopter over Site 2 at 0438 EOT on August
22, 1978.
35. Vertical profiles of ozone and temperature obtained by the 96
instrumented helicopter over Site 2 at 0500 EOT on August
23, 1978.
36. Vertical profiles of ozone and temperature obtained by the 97
instrumented helicopter over site 2 at 0450 EOT on August
24, 1978.
37. Air parcel trajectory beginning 0200 EOT August 18 and ending 100
1400 EOT on August 19.
38. Surface trajectories for three air parcels located within the 101
study area on August 19, 1978.
39. Vertical profiles of ozone and temperature obtained by the 103
instrumented helicopter over Site 1 at 0455 EOT on August
19, 1978.
40. Vertical profiles of ozone and temperature obtained by the 104
instrumented helicopter over Site 2 at 0425 EOT on August
19, 1978.
41. Vertical profiles of ozone, oxides of nitrogen, and temperature 105
obtained by the instrumented helicopter over Site 1 at 0919
EDT on August 19, 1978.
42. Acoustic radar record at Site 1 on August 23, 1978. 119
43. Vertical profiles of ozone obtained by the portable instrument 123
ozone package and the instrumented helicopter on the afternoon
of August 24j 1978, over Site 1.
44. Vertical profiles of ozone, temperature, wind speed and wind 127
direction obtained from the free lift ozonesonde sounding
of 0807 EDT on September 24, 1978, over Site 1.
45. Comparison between ozone profiles taken by free lift ozonesonde 130
and portable instrument package on September 30, 1978.
46. Vertical profiles of ozone, temperature, wind speed and wind 132
direction obtained from the tethered ozonesonde sounding
of September 21, 1978, beginning at 0759 EDT.
-------
LIST OF TABLES
Table Description
1. Evaluation of ozone and precursor measurement methods, as applied 24
to the typical urban oxidant transport situation.
2. Equipment used in the surface monitoring network. 28
3. Parameters measured by helicopter and instruments used. 34
4. Dates and times of flights using the portable instrument package, 41
and sites over which vertical profile measurements were taken.
5. Dates and times of successful launches of ozonesonde beneath a 45
free lift balloon.
6. Dates and times of launches of ozonesonde beneath a tethered balloon. 48
7. Frequency, times, and locations of hydrocarbon samples collected 50
for species analyses.
8. Average and maximum ozone and precursor concentrations (in ppm) 56
during August and September.
9. Frequency distribution of observed ozone levels (ppm) for August 57
and September 1978.
10. Ozone and precursor concentrations (in ppm) observed during days 65
of significant transport.
11. Gradients of ozone and oxides of nitrogen observed aloft along 78
the study area upwind boundary.
12. Ozone and gradients of oxides of nitrogen observed aloft along 79
transects from upwind areas to primary source areas.
13. Concentrations of ozone and precursors (in ppm) observed along 81
the study area upwind boundary at the surface on days of strong
transport.
14. Times (EDT) of maximum ozone concentration. 86
15. Ozone concentrations (in ppm) at upwind stations on August 22, 99
23, and 24.
16. Non-methane hydrocarbon (NMHC) flux rose for Site 1. 110
17. Non-methane hydrocarbon (NMHC) flux rose for Site 2. Ill
xiv
-------
Table Description
18. Non-methane hydrocarbon (NMHC) flux rose for Site 3. 112
19. Non-methane hydrocarbon (NMHC) flux rose for Site 4. 113
20. Non-methane hydrocarbon (NMHC) flux rose for Site 5. 114
21. Comparison of surface ozone concentration with early morning 120
concentration aloft.
22. Comparison of ozone data (pprn) provided by the portable 125
instrument package and the instrumented helicopter.
23. Comparison of ozone data (ppm) provided by ozonesonde beneath 129
a free balloon and portable instrument package.
24. Comparison of ozone data (ppm) provided by the ozonesonde beneath 133
a tethered balloon and the portable instrument package.
25. Cost of methods for measuring ozone transport. 136
A-l Summary of the Northeast Oxidant Transport Study nitrogen 145
oxides audit results.
A-2 Summary of the Northeast Oxidant Transport Study ozone audit 146
results.
A-3 Summary of the Northeast Oxidant Transport Study meteorological 147
audit results.
A-4 Regression analysis of Beckrnan 6800 gas chromatograph response 148
on audit concentration.
xv
-------
1. INTRODUCTION
The transport of ozone and its precursors has been reported by many investi-
gators. Evidence of the transport of oxidants from urban to rural locations has been
reported for the Washington, D.C. area (Wanta, et al, 1961), New Jersey (Leone, et al,
1968), Staten Island and the Bronx (Scott Research Labs, 1970), and Santa Ynez Valley
in California (Baboolal, et al, 1975). Long-range transport of ozone and its precursors
by air masses is indicated by data obtained in the Los Angeles Basin (Edinger, et al,
1972; Stephens, 1977), the Washington-Baltimore area (U.S. EPA, 1973), Southern
Ontario (Yap and Chung, 1977), Wisconsin (Lyons and Cole, 1976), and Connecticut
(Wolff, et al, 1977). The ozone laden plume from St. Louis was tracked out to 240 km
and was mapped in detail out to 160 km by White, et al (1976). Cleveland, et al (1976)
demonstrated by statistical analysis that ozone was transported over 300 km from New
York to northeastern Massachusetts.
These and other studies have demonstrated that ozone and precursors can be
transported over long distances. Control strategies designed for attainment of the
ozone standard in individual urban areas must take into consideration the impact of
transported ozone and precursors. The effectiveness of possible control strategies is
usually determined through mathematical modeling. Unfortunately, there has
generally been a lack of monitoring upwind of urban areas to assess the transport of
ozone and precursors. Thus, the impact of such transport on the maximum ozone
concentration downwind of an urban area is difficult to assess and cannot be properly
modeled.
Even though it is obvious that measurements should be made to quantify the
transport of ozone and precusors, there is a general lack of information on when,
where, how, and how much upwind monitoring data are needed to determine the
contribution of the transported ozone and precursors to the downwind ozone maxima.
In order that states can design control strategies to attain and/or maintain the
ozone standard in individual urban areas that are affected by transported ozone and
-------
precursors, guidelines have to be established by EPA on techniques to measure the
transport.
As a step toward developing guidelines for quantifying the impact of transported
ozone and precursors on maximum ozone levels in an individual urban area,
AeroVironment was selected by EPA to evaluate a number of techniques for measuring
such transport. Those techniques were evaluated during a field program conducted in
the summer of 1978 in the vicinity of Philadelphia, Pennsylvania. The specific
question of the effect of transported pollutants on the maximum ozone reading in an
urban area was not the primary concern of this study. Other studies are planned by
EPA to fully address that issue.
This report describes the operation and findings of the field program conducted
by AeroVironment and presents an evaluation of the various measurement techniques
tested in the field.
-------
2. REVIEW OF METHODS FOR MEASURING OZONE AND PRECURSOR TRANSPORT
The two primary mechanisms by which ozone and precursors might be
transported into an urban area are:
Advection of polluted air (ozone and/or precursors) along the surface, and
Advection of polluted air aloft, typically at night and during early morning hours
above the ground-based inversion, with downward mixing when the inversion
dissipates later in the day.
Overnight advection of ozone and precursors aloft appears to be the more
significant mechanism of transport from one urban area to another. Chapter 5
contains a detailed discussion of transport mechanisms.
To determine the degree of ozone and precursor transport, therefore, we must
measure wind speed and direction and the concentrations of ozone, NO/NO-, and
non-methane hydrocarbons at the height of the major portion of the transport flux. In
general, this implies that at least some measurements will have to be made aloft,
some hundreds of meters or more above the surface.
2.1 DISCUSSION OF MEASUREMENT TECHNIQUES
There are a number of specific ways by which we can measure the transport
phenomenon, and we have selected nine such methods for review here. All of them are
reasonably feasible within the range of financial and technical capabilities of state and
local air pollution control agencies. This is a basic requirement of any measurement
technique to be developed since state and local agencies are responsible for identifying
ozone problems in the urban areas within their jurisdiction and for developing control
plans to attain the ozone standard. These nine methods are:
-------
o Method A: Surface measurements at fixed sites.
o Method B: Surface measurements by moving vehicle.
o Method C: Measurements on tall towers.
o Method D: Airborne measurements by dedicated instrumented aircraft.
o Method E: Airborne measurements by a portable instrument package in a light
airplane.
o Method F: Airborne measurements by an instrumented manned balloon.
o Method G: Soundings by instrument package beneath a free lift balloon.
o Method H: Soundings and airborne measurements by tethered balloon.
o Method I: Remote sensing of conditions aloft by sensors on the surface.
This list may not be all-inclusive, but it should represent all the economically
feasible methods within the state-of-the-art at this time. Remote sensing from
aircraft and satellite have definite appeal, but are not yet available for routine use.
Obviously, two or more methods can be combined in one study to form the most
effective approach to data gathering for any given area. Each method will be
discussed separately below.
2.1.1 Method A; Surface Measurements, Fixed Sites
The traditional method for defining pollutant concentrations and transport
utilizes a network of fixed stations, each equipped with sensors for the pollutants of
interest and some with a wind vane and anemometer to define the direction and speed
of surface airflow. Since most urban areas already have such stations, their use for
this purpose has obvious economic appeal.
-------
In reality, however, existing ozone and precursor measurement sites are
generally located within the urban area and downwind of it, in the area of impact, and
seldom does one find a suitably located station upwind. Consequently, an additional
fixed site (or sites) might be needed upwind of the city to measure the O,, NO/NO2,
and hydrocarbon concentrations which are transported into the urban area.
No surface-based network will be able to assess transport aloft when a surface-
based inversion isolates conditions aloft from surface reactions. The surface wind
direction will generally differ by about 45° from the wind direction aloft in relatively
flat areas, with larger deviations possible near mountains. If desired, pilot balloon
tracking measurements, by theodolite, can provide data on winds aloft on days without
low clouds. Also, Doppler acoustic radar is now available for determination of winds
aloft up to one kilometer above the surface. As will be pointed out in Chapter 5 of
this report, a simple non-Doppler acoustic radar can be an indispensable addition to a
surface air quality station, allowing inferences of upper air ozone transport using
surface O., measurements.
Surface stations have the advantages of space and power availability and thus
can measure several parameters of interest. Even on-site hydrocarbon speciation
analyses can be done for individual hydrocarbons by automated gas chromatography.
Since the stations are fixed, however, several might be necessary to adequately define
the spatial variability of the upwind air mass.
If existing monitoring networks can be used for measuring incoming transport,
then the surface-based approach has very low incremental costs. On the other hand,
furnishing and installing a new station for O,, NO/NO-, non-methane hydrocarbon, and
wind measurements costs from $^0,000 to $60,000. Monthly operation and data
processing costs will typically amount to $2,000 to $5,000, depending upon the degree
of station automation.
2.1.2 Method B; Surface Measurements, Moving Vehicle
A versatile form of surface-based monitoring involves an instrumented vehicle
which is totally self-contained and is able to monitor pollution and meteorology while
-------
moving or stopped for brief periods. The moving vehicle suffers from the lack of
ability to measure conditions aloft, much the same as the fixed stations do, but offers
appealing advantages in flexibility. Such a vehicle can follow an air mass as it travels
from upwind areas into the city, and can also measure profiles transverse to the wind
to determine the spatial inhomogeneity of the arriving air mass. By its motion,
however, the moving station loses the ability to observe temporal variability of
conditions at a point, although repeated surveys over a fixed route of travel can
provide both a temporal and spatial picture.
Moving stations of this type are currently being operated by the EPA, several air
pollution control agencies, and many environmental consulting firms. They range in
sophistication from commercially available moving laboratories equipped with compu-
terized navigation systems monitoring the vehicle location and correlating it with the
pollutant data, to simple vans with a few air quality analyzers and strip chart
recorders either temporarily or permanently installed.
Electrical power for operation while moving can be provided by a battery-
inverter arrangement or by a small propane-fueled generator; or battery-operated
equipment can be used. When the vehicle is stopped for a few hours or more,
electricity must be generated by the vehicle generators or an auxiliary generator. In
all cases, the available electrical power is limited, and cooling or heating of the
operating environment to assure proper instrument operation must be provided by the
vehicle's own environmental control systems. In addition, exhaust from the vehicle
engine or a generator can pose vexing sample contamination problems when the
vehicle is stopped.
Vehicle engine exhaust can also affect the instrument measurements while
moving, for two reasons. First, because of a possible recirculating airflow around the
truck or van, exhaust from the engine could contaminate the air samples. Careful
testing of the actual installation (or a wind tunnel model) is necessary to insure that
this does not occur. Second, the exhaust emissions from other vehicles will result in
higher NO/NO2 and hydrocarbon levels and lower O3 levels on the roadway than in the
surrounding area. Consequently, moving laboratory measurements made in heavy
traffic will not be representative of the oxidant and precursor levels for that location.
-------
Reliable wind measurements are difficult to make from a moving vehicle and
require sophisticated sensors of vehicle motion. Thus, it is generally preferable to
make wind measurements when the vehicle is stopped for brief periods. Proper wind
instrument exposure is also a problem since tall masts cannot be mounted on moving
vehicles; telescoping masts can be used to support instruments when the vehicle is
stopped, however.
A typical dedicated moving vehicle will cost from $75,000 to more than
$150,000, with monthly operating and data processing costs varying from about $5,000
to $15,000, depending upon the degree of automation of the data collection equipment
on board. It costs more initially than a fixed station because of the vehicle itself, the
electrical power generation equipment, and the more difficult installation and
monitoring requirements for instrumentation to be used while moving. Data process-
ing costs are higher because when the vehicle is used in a moving mode the position of
the vehicle is a continuously changing parameter.
To provide the flexibility of moving measurements at lower cost, simpler, ad hoc
arrangements of instruments in an available motor-home or van will often suffice for a
few days of measurement. However, a proper operating temperature and provision of
sufficient electricity for environmental control and instrument operation generally
requires a dedicated vehicle.
2.1.3 Method C; Tower Measurements
The depth of the nocturnal surface-based inversion which isolates pollutants
aloft from surface emissions is often fairly shallow — from tens of meters to a few
hundred meters. Also, the major portion of the turning of the wind vector between the
surface and aloft occurs in the first few hundred meters above the ground. Conse-
quently, a sensor situated 100 meters or more (preferably more) above the surface will
often describe conditions within the inversion or even above it, or will at least provide
a basis for estimating these conditions.
Such a height is achievable by installing air quality and meteorological sensors on
existing radio/TV towers or on tall buildings. In effect, a fixed air monitoring station,
-------
such as in Method A, can be installed on an elevated platform. Obviously, such an
approach will work only if an existing tower or building is in the proper location, since
the cost of constructing a tower specifically for this purpose is formidable.
An installation on a tall building, especially on the roof, is relatively straight-
forward, although care has to be taken to locate air quality sample inlets away from
building air exhaust ducts, and the effect of the airflow around the building on the
anemometer has to be considered.
Installations on tall towers can be complex, especially since the sample lines to
air quality analyzers should be no longer than a few meters. Maintenance of
instruments installed on towers can be difficult. Also, the strength of the radio
frequency radiation field around broadcasting towers can interfere with the proper
operation of the electronic portions of analyzers and recorders unless proper, and
sometimes elaborate, shielding techniques are used. (In some instances, adequate
shielding will be impossible and the site will be unsuitable for some of the instru-
ments.) Sometimes, the wind load structural limits of the tower will preclude major
equipment installation on the tower itself, so that only a few sensors or sampling
devices can be mounted on it, and the bulk of the signal processing electronics will
have to be located at the tower base.
In spite of these limitations and difficulties, when a tower exists in the proper
location installation of an air quality and meteorological station on it may be an
effective approach to the acquisition of pollutant transport data. To supplement the
main measurements, carefully matched temperature sensors should also be installed at
several locations on the tower to define the lapse rate profile and locate the inversion
level, thereby identifying the air mass which is being monitored. A standard ground-
based air quality station should be located near the tower to provide a complete
picture of the pollutant transport situation.
An excellent example of an instrumentation installation on a tower is provided
by the Mt. Sutro television tower near San Francisco (DeHay, 1974; Russell and Uthe,
1976). The Mt. Sutro system has been used to study the vertical structure of ozone
concentration, and its relation to the meteorological situation (MacKay, 1977).
-------
Although this tower is comprehensively instrumented for detailed boundary layer
research and, therefore, contains many sensors which would be of no interest for
oxidant transport investigations of concern here, the design of the system is worthy of
review by anyone intending to install a monitoring station on another tower.
A simple installation in a tall building will cost at least as much as the $50,000
to $100,000 range of a fixed-surface site; the cost of an installation on a tower can be
significantly greater depending on the constraints imposed by the tower design and
use. Operating costs start at the same $2,000 to $5,000 range as for a fixed surface
site and extend upward depending on the difficulty of access to the instruments for
maintenance.
2.1.4 Method D; Airborne Measurements, Dedicated Instrumented Aircraft
In the past five years, the use of instrumented aircraft for air pollution surveys
has become rather commonplace. Aircraft offer advantages of both horizontal and
vertical mobility, and can map out the characteristics of large volumes of air in a
short period of time. Example of studies in which instrumented aircraft have been
used to study urban air quality include measurement programs in Los Angeles and San
Francisco (Blumenthal, et al, 1974) and in Denver (Anderson, et al, 1977).
The most effective approach to aircraft measurements involves an aircraft in
which instrumentation has been installed in a relatively fixed manner, appropriate
connections and modifications have been made to the aircraft electrical system to
power the equipment, and properly designed sampling probes have been inserted into
the airstream. Aircraft space, weight-carrying capacity, and electric power are
generally limited, and such installations therefore require careful design to satisfy
these limitations.
The sampling probe location must be chosen to avoid sampling the engine
exhaust. For twin-engine aircraft a probe in the nose or on top of the forward part of
the fuselage is generally satisfactory, while for a single-engine aircraft a location on
the wing (for high-wing models) or on the upper part of the fuselage has been found to
be adequate (Adams and Koppe, 1969). Helicopter sampling probe locations depend
-------
greatly on the helicopter configuration; generally a probe as far forward as possible,
and as low as possible, provides the most representative sample. For the gaseous
pollutants of concern here, isokinetic sampling is not necessary, but, as with fixed
stations, the residence time of the sample in the sampling system should be as brief as
possible.
Because such an installation results in modifications to the aircraft structure and
systems, the Federal Aviation Administration (FAA) requires relicensing of the
aircraft in a "restricted" category --a relatively straightforward process for an
aircraft mechanic who has experience with special-use aircraft modifications. The
fixed installation can also include connection of the aircraft navigational and
operational instruments to a recorder so that the air quality and meteorological data
collected can be correlated with the location of the aircraft.
A typical aircraft flight path for profiling the air mass might look like the
representation in Figure 1. The path consists of spiral soundings to define the
vertical structure and horizontal traverses to define the spatial distribution. Because
of the high speed of the aircraft, any given area can be repeatedly surveyed at
frequent intervals, thereby providing a fairly continuous picture of the situation at any
given location. If the soundings can be made near the ground, as at airports or in open
areas, the aircraft can provide measurements representative of surface conditions. In
general, though, operations below 150 m above the surface are not permitted, and even
flying at that altitude over populated areas requires special authorization from the
FAA.
Aircraft measurements are limited to relatively fair weather conditions, since
the ability of an aircraft to fly patterns which do not conform to normal air traffic
flow is severely limited by air traffic control procedures whenever significant clouds
are present in the flight area or the visibility decreases below 5 kilometers. Also,
operations in the vicinity of major air terminals are difficult because of the
inconvenience they can cause other aircraft.
An instrument system installed in an aircraft can be used to measure O,,
NO/NO-, and total hydrocarbons. Total hydrocarbons can be measured by continuous
10
-------
C\J
U
b
c
0)
g
5.
>>
•
UJ
at
D
CO
11
-------
flame ionization detectors; the gas chromatographic analyses needed for non-methane
hydrocarbon (NMHC) measurement, or for further speciation of hydrocarbons, are
generally not done in aircraft because of the low sampling frequency of gas
chromatographs (once per 5 minutes or less often). Continuous NMHC measurements,
such as by catalytic conversion of the methane in the stream, are not yet reliably
achievable. Hence, the only workable method for airborne NMHC measurements
involves sample collection in evacuated canisters or flow-through glass sampling
containers, with subsequent analysis in a laboratory on the ground.
Aircraft also cannot conveniently measure winds aloft, at least not without
expenditure of the order of $100,000 for inertial or Doppler navigation systems. Thus,
determinations of air pollutant transport will require supporting wind measurements by
surface-based pilot balloon tracking or Doppler acoustic radar.
To resolve the spatial distribution of air pollutant concentrations from a moving
aircraft requires that the instrumentation on board have a faster response to changes
in pollutant concentrations than would be required for measurements at a fixed site.
The performance specifications for automated reference and equivalent methods in
40 CFR53.20 require that, for ozone and NO-, the instrument lag times be no more
than 20 minutes and the rise times be no more than 15 minutes. For spatial resolution
of, say, two kilometers horizontally and 100 m vertically, and with instruments in an
aircraft traveling at 50 m/s (approximately 100 knots) and ascending or descending at
150 m/min (approximately 500 ft/min), the response time of the instruments would
have to be 40 seconds, however. The commercially available O, and NO/NO
j X
analyzers which are of interest for aircraft applications and which have received EPA
designation, do have response times considerably faster than those required by
40CFR53.20, but they do not have a response time of better than 40 seconds. Faster
response times are available on most instruments, usually by changing a switch setting,
but the EPA designation no longer applies when the instrument is operating in this
faster mode. This presents a dilemma, since monitoring instruments used in support of
a State Implementation Plan must be designated by EPA in accordance with 40CFR53,
but use of such currently available instruments in aircraft would not, in general,
provide data with the spatial resolution needed for effective use of the data. Thus, at
the moment, all air pollution research aircraft known to us (including the EPA
12
-------
helicopter used in this study) use instruments which are not in compliance with
40CFR53 because their control settings do not conform to those which are required for
operation as a designated method. Obviously, if aircraft are going to be used routinely
for control strategy development purposes, it will be necessary for instrument
manufacturers to achieve the specifications of 40CFR53 with much faster response
times or for the EPA to recognize the special nature of aircraft measurements and,
therefore, to adjust 40CFR53 to allow otherwise acceptable currently available
instruments to be used on faster response scales with some attendant degradation in
instrument signal quality.
A major drawback to the use of dedicated aircraft is the initial cost, with the
aircraft itself a major component of the cost. A new light, single-engine aircraft
suitable for installation of a minimal ozone profiling measurement system (O.,,
temperature, altitude, recorder) will cost in the vicinity of $50,000; prices for a new
twin-engine aircraft suitable for a more sophisticated installation with NO/NO and
navigational data recording added, and with room for hydrocarbon sampling containers,
begin at $150,000. Consequently, the total cost of a dedicated aircraft will be a
minimum of about $70,000 for a simple installation in a single-engine light plane, and
can easily reach $200,000 or more if twin-engine aircraft and sophisticated recording
systems are desired. The operating cost, which will range from $100 to $500 per flight
hour for light aircraft, including crew but excluding post-flight data processing, can be
very reasonable in terms of the amount of data collected. Data processing costs will
vary widely, depending on the number of parameters recorded, the type of flight
profile, and the type of the data logging systems on the aircraft. Thus, the data
processing cost can range from $50 to $200 or more per flight hour.
2.1.5 Method E; Airborne Measurements, Portable Instrument Package
As was mentioned above, airplanes offer an effective platform for airborne
soundings of air pollution and temperature. The major stumbling block to their use is
the cost of instrumenting and maintaining a dedicated aircraft. On the other hand,
ad hoc instrument installations in aircraft tend to be difficult because of requirements
for electric power, carrying of compressed gas, and electrical and radio frequency
noise and interference. Any electrical connection to an aircraft requires recertifica-
13
-------
tion of the airplane into a "restricted" category. The effect of carrying compressed or
flammable gases (ethylene, for example) is the same. Thus, a simple installation of
standard air quality instruments for short-term studies using a rented airplane is
generally not possible.
The availability of new, flyable, battery-operated O, and NO/NO analyzers*
offers a new approach to airborne measurements. These analyzers, plus battery-
operated recorders, can be installed in small cases which can be strapped down onto
the seats of an aircraft. A thermistor temperature sensor can be included to record
the air temperature, and an event mark button can be clipped to the aircraft control
wheel and used by the pilot for marking cardinal events (for instance, crossing of
selected altitudes or passage of significant landmarks) on the chart records.
Samples can be carried to the analyzers by a 1- to 2-cm diameter Teflon line
threaded through the cabin fresh air inlet of the aircraft and extended out into the
free airstream. The airflow through the sample line is ram air-driven. Only a small
portion of the flow is needed by the analyzers; the rest can be released into the cabin
or used for hydrocarbon grab sampling. The temperature sensor can also be installed
in the same fresh air vent as the sample tube.
Such a system offers many of the advantages of dedicated aircraft and, in
addition, is inexpensive to operate — no installation is required -- so any rented or
chartered small aircraft (such as a Cessna 172 or a Piper Cherokee) can be used; the
only cost is for the aircraft and pilot during the actual time flown. The portable
instrument package may be able to be flown closer to the ground than systems in
larger aircraft simply because of less aircraft noise. It can also be used in a
helicopter. The measurement system is simple enough that, for repeated measure-
ments, the pilot can operate it alone and an observer is not needed. When there is
extensive air traffic, however, the pilot alone may be fully occupied with flying the
aircraft and may not be able to monitor whether the instruments are operating
properly.
Manufactured by Columbia Scientific Instruments, Austin, Texas, (O, and NO/NOx>
and Analytical Instrument Development, Avondale, Pennsylvania (O-, only).
-------
Because the currently-available portable O_ analyzers require compressed
ethylene gas for operation, use of such an analyzer in a normal light plane is not
permitted without special arrangements, however. Two options are possible: (1)
certification of the aircraft in the "restricted" category whenever the analyzer is
being operated on board, or (2) obtaining of an exemption from the Department of
Transportation regulations covering the carriage of hazardous materials in an aircraft,
following 41CFR107.101. The former approach is easy to implement and places no
limitations on use of the aircraft when the analyzer is on board. It may require
additional insurance coverage by the aircraft owner, however. The second approach,
to our knowledge has not yet been tried for this purpose; it will probably present the
same insurance situation. A third approach for using the O, analyzer legally, which
was tried during this study, is to use non-flammable Ethychem gas (a mixture of CO-
and CJr{ ) instead of pure ethylene. As will be discussed in Section 5.3, this approach
was not totally successful because the instrument sensitivity was reduced when the
Ethychem was used.
Other disadvantages of the portable system are much the same as for the
dedicated aircraft system. In particular, hydrocarbon measurements can be done only
by grab sampling, in much the same way as for the dedicated aircraft. Also,
experience with such portable systems in aircraft is still limited. Unless the
manufacturer can provide data on the variation of instrument calibration with
altitude, the user will have to perform such calibration tests himself.
An additional disadvantage of the portable instrument package approach, when
compared to a dedicated aircraft, is the lack of precise position data for the aircraft.
Ail coding of aircraft position is provided by noting altitude and location on a log and
marking the corresponding point on the concentration strip chart with the event
marker. Thus, the aircraft position is recorded at discrete intervals and not
continuously. This becomes a disadvantage when the visibility is poor in areas with
few landmarks to define location or whenever the pilot workload is heavy because of
air traffic control procedures, other aircraft in the area, or a complex flight pattern;
in such cases, the intervals between logged event marks might be greater than
desirable.
15
-------
A portable instrument package which records O~ and temperature will cost about
$8,000; one which also records NO/NO costs about $16,000. Rental of an appropriate
}v
aircraft with pilot costs $40 to $50 per hour. Data processing costs would vary with
the type of profiles flown; our experience with the package tested in this study
indicates a range of $75 to $100 per hour of data collected.
2.1.6 Method F; Airborne Measurements, Manned Balloon
In the last few years, instrumented manned balloons have been used to study
urban air pollution in the area of Los Angeles (Heinsheimer, 1977) and St. Louis (the
DaVinci II experiment; Environmental Science and Technology, 1976, and Forrest, et
al, 1979). Since a balloon is a quasi-Lagrangian tracer of air mass motion, it can
follow the progress of photochemical reactions in an air mass as it approaches an
urban area and passes over it.
Manned balloons are expensive to purchase, difficult to use (require large ground
crews for launch and recovery) and have no on board electrical capability. Never-
theless, their ability to study a specific air mass for long periods (even days) has
appeal for research purposes. Because of their operational disadvantages, however, it
is unlikely that manned balloons would be used routinely by most air pollution control
agencies for assessing pollutant transport at urban areas for purposes of control
strategy development. Consequently, a detailed analysis of this approach will not be
carried out here.
2.1.7 Method G; Free Lift Balloon Soundings
Free balloons are used routinely for ozone soundings of the troposphere and
stratosphere, and for measuring atmospheric temperature, relative humidity, and
winds aloft. The use of free balloons for ozone soundings of the polluted mixed layer
near the surface appears to be a straightforward combination of these well developed
techniques.
16
-------
The measurement system consists of a iodometric electrochemical ozone sensor
which was designed for use with a radiosonde. For use in this study the ozone sensor*
was combined with a minisonde** containing a radio transmitter to telemeter
measurements of temperature, pressure, and ozone to a ground receiver.
The combined ozonesonde/minisonde instrument package is flown under a free
350-gram balloon with a rate of rise of approximately 100 m/min. The temperature
and pressure information transmitted by the package allows calculation of the balloon
altitude, by integration of the barostatic equation; the ozone and temperature
information records then provide profiles of ozone and atmospheric stability with
altitude.
This measurement system provides information on inversion heights and ozone
layers aloft but it does not measure precursors. It would normally be launched near an
existing air monitoring station, which would thus give ground values of ozone and,
possibly, NO/NO and hydrocarbons. Winds aloft can also be obtained by tracking the
balloon with a theodolite, as long as the balloon is clear of clouds.
This package is lost after each flight. Each package could be equipped with a
parachute and tagged with a label offering a reward (say, $25) for its return, but
experience suggests only a relatively small fraction of these packages would be
returned. Those that are returned can be refurbished and reused.
The performance characteristics of the electrochemical ozone sensor were
studied by Torres and Bardy (1978). The ozone sensors measure total oxidants and thus
also respond to SO_ (100% negative interference) and, to a lesser degree, NO- (10-20%
positive interference). These pollutants are of no concern for the stratospheric
measurements for which the ozonesonde was designed, but can severely impair the
effectiveness of the method in the polluted air of the boundary layer. Removal of SCX
and NO_ from air streams, with negligible effect on O, concentrations, has been
Manufactured by Mast Development Corporation, Davenport, Iowa
**manufactured by Atmospheric Instrumentation Research, Boulder, Colorado
17
-------
achieved using an absorber column packed with silica gel soaked in sodium dichromate
and sulfuric acid and then dried. The method is described by Saltzman and Wartburg
(1965) and an application of a variant of the technique is described by Miller, Wilson
and Kling (1971).
The major virtues of this approach are:
o simplicity — only one technician is needed
o flexibility - it can be used anywhere (even near airports, with FAA
concurrence, and in the center of an urban area) and will work in fog and
clouds
o mobility - successive soundings can be performed at different locations as
rapidly as one can drive from one to the next
o rapid response deployment - can be used on short notice since no installa-
tions are required
The primary disadvantage of the approach is its lack of precursor measurement
capability -- it measures only oxidants. It does, however, provide this data continu-
ously from the surface upward and in conjunction with simultaneous lapse rate data
and wind data, if desired.
The purchase price of the ground radio equipment and recorders is about $5,000;
a theodolite adds about $4,000. The cost of each launch is about $250 for the
instrument package and materials. Labor costs for the launch and data reduction will
be about $80 if the balloon is not tracked by theodolite, and about $120 if the wind
information is obtained.
2.1.8 Method H; Soundings by Tethered Balloon
The instrument package of Method G can equally well be suspended beneath a
tethered balloon, which enhances the overall utility of the ozone sounding package.
When supported by a small tethered kite balloon (kytoon) the instrument package is
retrieved after each flight and can be reused.
18
-------
A 4.25 m balloon is large enough to lift the instrument package but can still be
handled by one technician in calm conditions, and by two technicians in moderate
(10 m/s) winds. The balloon is small enough to be stored in a single-car garage
overnight so that daily deflating and refilling with helium is not necessary. A
commercially-available kytoon sounding system* uses a sensor package which also
records winds, and which can be mated to the ozonesonde.
The tethered ozone and temperature sounding package will generally provide
atmospheric lapse rate and ozone data to a maximum height of about one kilo-
meter --in the same manner as the system in Method G. The maximum height would
be less in strong winds (> 10 m/s) and also at high altitudes (such as Denver) where the
lift of the balloon is less.
Most of the advantages and disadvantages of the free balloon system also apply
to the tethered system, with the following differences:
o Soundings can be made repeatedly with the same package.
o The tethered balloon system can be "parked" at a given altitude to monitor
the temporal variability of ozone aloft at a fixed location.
o The tethered system cannot be used near airports because the tether lines
make it a hazard to aircraft.
o The kytoon is clumsy to move and handle in comparison to the small free
balloon.
The efficiency of the NO^/SO- scrubber, which was discussed for use with the
free balloon method, degrades (for NO- removal) as the scrubber is exposed to
humidity for extended periods. Because the tethered system can be flown for
extended periods, a different technique for the removal of the interferents may be
Manufactured by Atmospheric Instrumentation Research, Inc., Boulder, Colorado
19
-------
needed. A technique using CrO. and H_SO. impregnated filter paper to remove SO-
has been discussed by Saltzman and Wartburg (1965), as well as in Methods of Air
Sampling and Analysis (1972, American Public Health Association). This technique has
the disadvantage that it converts NO to NO- and, therefore, increases the concentra-
tion of that interferent; but it would be useful when NO and NO- levels are low and
thus the total NO effect negligible.
x
The cost of the kytoon system with the ground-based receiver and recorder is
about $10,000. Daily operating costs are primarily for the labor and should be less
than $500, while data processing costs depend on the level of detail which is sought.
The same ground receiver is usable with the instrument packages for both the free and
tethered balloon systems.
2.1.9 Method I: Remote Sensing
All the methods discussed above for the sensing of wind or air pollution aloft are
relatively cumbersome to use. The ideal method would involve continuous ground-
based sensors which could measure pollutant and wind data aloft. Some of these
sensors already exist, and a few are even commercially available or soon will be.
Therefore, reality may not be too far from the ideal, at least for some parameters.
The most developed method for the measurement of pollutant concentrations
aloft is correlation spectroscopy (Millan and Hoff, 1978). Here, the spectrum of light
originating from a source (generally, the sun) and passing through the air mass is
compared against the spectral absorption characteristics of NO- to determine the
total amount of NO2 along the light path. Since most of the NO- is near the earth, in
the mixed layer, this allows calculation of the average concentration of NO- within
that layer; it cannot, however, provide information on the vertical distribution of NO-.
The data obtained, when combined with a mean wind derived from other measure-
ments, allows calculation of the total flux of NO- which is being transported by the
wind. The method has also been used for SO-, but not for O, or hydrocarbons.
Active remote sensing techniques using a laser light source and observing
scattered or re-emitted radiation from the illuminated portion of the air mass are
20
-------
currently being tested. These laser techniques are able to measure pollutant
concentrations in a well-defined remote volume, and thus can measure spatial
distributions. Because of signal strength requirements, their greatest success is likely
to be in the area of measurement of smokestack plumes, but some ambient measure-
ments may also be possible. Remote measurements of ambient O, and other
pollutants have been made over a closed path, but such methods are not useful for
measurement of pollutants aloft. At present, laser measurement methods for gaseous
pollutants are experimental and are unlikely to become fully operational within the
next year or two; therefore, the discussion here will not further explore the various
approaches.
An acoustic radar (or sodar) can be used to measure meteorological conditions
aloft, and has been used to characterize the air pollution meteorology of the San
Francisco Bay Area (Russell and Uthe, 1978), in Los Angeles (Edinger, 1975), and in
other cities. Commercially available instruments will display the locations and extent
of temperature inversions and thus define the extent of the mixing layer. Their
vertical range is typically 1 km, although longer range instruments have been built.
Lidar will also provide the same type of information; Russell, et al (1974), contrast the
capabilities of these two techniques.
Doppler acoustic radar can provide continous data on winds aloft. Recent
developments have allowed the measurement of winds up to 1 km above ground level.
Such Doppler instruments, when combined with airborne measurements of pollution,
offer a workable short-term approach to complete pollutant transport monitoring.
Instruments have recently become commercially available from several firms in the
United States and Europe. A paper by Peters, et al (1978), critically analyzes the
capabilities of some acoustic Doppler approaches. Remote wind measurements can
also be made by Doppler lidar (Post, et al, 1978) and by microwave radar (Chadwick,
et al, 1976), but at significantly higher cost.
Remote measurement sensors vary widely in cost. At the upper end, laser
systems can cost in excess of $200,000. Doppler acoustic radar lies in the $25,000 to
$50,000 range. The correlation spectrometer costs less, and the simple acoustic radar
is the lowest priced at about $10,000. Operating costs also vary widely depending upon
21
-------
the need for an operator and the degree of automation of the data recording method of
any given approach. At this time, only the correlation spectrometer and acoustic
sounders (including Doppler units) are commercially available.
2.2 COMPARATIVE EVALUATION
The methods discussed in the preceding section have offered a spectrum of
capabilities and costs. We have established a number of criteria for evaluating their
usefulness in obtaining oxidant and precursor measurements to be used in developing
oxidant control strategies. In formulating these criteria, we assumed that these
methods would be applied by state and local air pollution agencies during the oxidant
seasons, in the next two to five years.
The evaluation criteria are:
o Initial cost - should be as low as possible.
o Operating cost - should be as low as possible.
o Manpower requirements - should be relatively low.
o Required technical expertise to operate - should be within capability of
most state and local air pollution control agencies.
o Portability - ability to use in more than one area during the same oxidant
season would be desirable.
o Measurements made - O, mandatory, NO/NO- very desirable, HC desirable,
winds mandatory (if at all possible).
o Location of measurements - both surface and aloft is most desirable.
o Site specificity - approach should be usable equally well in several loca-
tions.
22
-------
An evaluation of each method against these criteria is shown in Table 1. The
overall costs and benefits of the various methods have been subjectively ranked, with
the evaluation based on the situation which is expected to prevail in most urban areas
where oxidant and precursor transport is of interest. The ranking could be substan-
tially different for a specific case where a particular capability, for example,
portability, was important. A cost-benefit index, based on this subjective ranking, is
also shown.
Based on this subjective analysis, fixed stations (Method A) appear to be both the
most efficient and the most cost-effective oxidant transport measurement method,
Aircraft measurements, using a portable instrument package (Method E) are less
effective in terms of benefits alone, but may be as cost-effective as the fixed sites.
Free balloon and tethered balloon soundings (Methods G and H) are tied for third place
in the cost-benefit ranking, mainly because of their extremely low cost. Towers and
dedicated aircraft (Methods C and D) are tied for fourth place overall, with the high
cost of the aircraft being the major factor in placing it this far down the list.
Remote sensing could not be ranked as a group because of the wide variety of
capabilities and costs. The absence of an O, measurement capability is a serious
drawback, however. Acoustic sensing of winds and atmosphere stability may be a very
effective tool in conjunction with other techniques for pollution sensing.
To provide further evaluation of these approaches, methods A, D, E, G, and H
were tested in Philadelphia in the summer of 1978. A description of the tests
performed and a more detailed evaluation of these five techniques are provided in the
following chapters.
23
-------
8P
0<
= 2
Hfc
UJ03
0^ ^
a, — '
NUJ
OX
u,1-
OO
zj;
QO
•a
H
"u
c
J2
a!
£
I
a
C
13
0
, "2
8^ "w x
C V
U10£
A-
C .-
8
u
a'^
" u
l/l
0 1
o 3 £
yl
J
u
s
3
«j
fc
o.
6 .2 u
I** UJ Q*
u
flj
% J2
||
5
V
iu £P .
£ +3 £
5 n) ~
£ fc 8
1°
^3
5 8
-u
C
«•,
o
O
0
vo
"£
00 OO OO
- , „ N „ ...5
» *, « w - ~ S
o
Z
„
s .2
.
O
I!
ox
o
Z
rt •
S
=,
-J
4J
n)
*
s
rt
5
i <"
§1
m xi
Surface me
ments, fixe
<
o
Z
^
n °°
tj S
L5
fc
o
Z
o
•^c
O *
8
>-
m
1
a
X
*2
2 ^
5 -^
s 2
•5,
X
L
W C
Surface me
ments, mov
vehicle
CD
gg>
'at;"
5§-gI
+^T1
11
„
O
t-l ^
0 o
< E
|
?^i
O
8
1
•a
X
i
J
1
i
b u
Airborne m
ments, port
instrument
package
ui
o
Z
00
v.£
° 0
< E
0
11
ox
£
•a
£
•a
X
1>
ffl
T?
•a
£
^
S C
Airborne m
ments, man
balloon
u.'
O O c?
z z 3
^ u
X O ^
*•*" ' *^"O *^ S D
° 1 o X Itc
< v! < -Z < J3 O
TJ
1 I 1
m n 0^
o o z
8
8 8 9
_i -i 3
«i~
rt *— '
•a -a p
x £ s S
U V
o ? o 2
" fc *• S3
1 5 !"§
J E ° E 3
J I s
„
* c
MO 1^1
00 0 OO-o
55 5 w c £ oo
if §11 if
£3 £j!.g £ S
ox -:
V
K
.E v
X 5
«S
1!
fs
a .y
8 'S
•- X
V)
& E
^ D
5 JS i S
ill iU
-------
3. FIELD PROGRAM TO EVALUATE SELECTED MEASUREMENT TECHNIQUES
Five selected methods for measuring the transport of photochemical pollutants
into Philadelphia were tested in the summer of 1978. These methods were: (1) a
continuous surface monitoring network; (2) a dedicated instrumented helicopter; (3) a
light aircraft equipped with a portable ozone/temperature instrument package; W a
free lift balloon ozonesonde; and (5) a tethered balloon ozonesonde. In addition, grab
and time-integrated hydrocarbon samples were collected at four of the surface
monitoring stations and by the helicopter and light aircraft on selected days for
species analysis. The surface monitoring stations were operational during July,
August, and September of 1978, while measurements using aircraft and balloons were
made on 15 days during August and September.
3.1 OBJECTIVES OF THE FIELD PROGRAM
The primary objective of the field program was to demonstrate and evaluate the
five selected methods for measuring the transport of photochemical pollutants into
Philadelphia. Other objectives included:
1. Assessment of typical levels of O.,, NO, NO-, and non-methane hydro-
carbons transported into Philadelphia at the surface and in layers aloft.
2. Evaluation of species composition of hydrocarbons transported into Phila-
delphia.
3.2 PROGRAM DESIGN
To achieve the objectives of the field program, AV established five surface
monitoring stations around Philadelphia. Continuous measurements of ozone, nitrogen
oxides, non-methane hydrocarbons, wind speed and direction, temperature, and solar
radiation were made at each of these stations during July, August, and September of
25
-------
1978. An acoustic radar to collect mixing height information was also installed at one
of the stations.
Aircraft measurements of Ov NO, NO , visible light scattering, temperature,
J yC
and dew point were performed by the EPA-Las Vegas Environmental Monitoring and
Support Laboratory (EMSL) on 12 days in August.
The portable ozone/temperature package, the free balloon ozonesonde system,
and the tethered balloon ozonesonde system were tested in August and September. In
addition, grab and time-integrated air samples were taken on the ground and in the air
on seven days during August and September. Those samples were analyzed by Scott
Environmental Technology, Inc. for hydrocarbon species (C? through C]n and
aromatics).
More detailed descriptions of each segment of the field program are presented
below.
3.3 SURFACE MEASUREMENTS AT FIXED SITES
The surface monitoring program consisted of five stations set up specifically for
the study and six supplemental stations whose data were also used in the analysis. The
locations of all stations are shown in Figure 2.
The parameters monitored continuously at Sites 1 through 5 included: O,,
NO/NO , THC/CH., wind speed, wind direction, temperature, and solar radiation.
X f
Mixing height was also measured at Site 1. All of the equipment used, except the
acoustic radar, was supplied by the EPA. The specific instruments used at these
stations are listed in Table 2. The methodologies used for the measurement of oxides
of nitrogen and ozone are EPA reference methods. There are currently no EPA
guidelines on the measurement of non-methane hydrocarbons.
Ozone data collected at all supplemental sites and NO/NO data collected at
Franklin Institute and South Broad Sites were also included in the data analysis.
The purpose of each of the stations is discussed below.
26
-------
a^Wffit//^
MMrli ,M3"..-i*'"""m?r-'"J.
f^-,::':g:^
><&. \ w*,v/™7 ,:,1.; *,»• T,J: .TOUU-,;
s^^sMt^a&s
-» ,j- ,»-„.„ -f, 1/s^;;;,, i j v_i:\iy.,»r-^
^^&^^%./fe^.
*' \,^. ^X^wrj^C
'>r,-.,iL '' -»A, N™ .. V* J ^la'n'l1""'
V- ' ILJ > ,- ""'f:
^/•n^SskV:
'"TX. '.4(*i»"» f' l\ TSll
-M^Xh) -4^ 'i',"^
\ 7'\V- ' V • •' •:^™~ .Viiflim)
J\f^, 7\V
«t' t •. j- /n
/ Vr~p=T /" "v
• BIVALVE
FIGURE 2. Locations of monitoring sites. Numbered sites are transport stations
and dots represent supplemental stations.
27
-------
a;
O
UJ
Z
o
z
5
o
j^
z
o
S
UJ
u
<
LL,
a!
CO
Q
UJ
H
Z
UJ
S
a-
5
a
UJ
m
o
'43
U
"£
Q
4-1
O
o
(U
S
.^
u
nJ
U
U
(U
00
&
5r "o
4-i
I
c
1— 1
c
Chemiluminesce
E
CL
CL
0
o
°'
CL
CL
lf\
*
O
1
O
^
O
lw
N
^•»
"n5
-5
cT
(M
O
O
00
Bendix
c c c
Chemiluminesce
Chemiluminesce
Chemiluminesce
E £ £
CL CL O.
CL CL CL
O O O
O 0 O
+1+1 +1
£ E £
CL CL CL
CL CL CL
• • •
o o o
(N X
OO O
zz z
u
N
>\
•3
c
X
o
~^ •^*
O (S4
X 5.
3 O
I2
i i
>> x
"CL c Q c
rt o n3 o
Gas Chromatogr
Flame lonizati
Gas Chromatogr
Flame lonizati
£ E
a a
a a
o o
+ 1 +1
CL a.
CL CL
o o
«^H ^-t
') |
O 0
U •*
H U
CL
a)
00
*"* (Tl
< E
§1
oo (j
£'^
F
o
.i
"O
L_
Wind Cups
Wind Vane
Thermistor in a i
shield
t/5
£ U
+'|l+f+i
U
oo 0
(VJ if- O
CN ir\ *•*
i i Is**.
c
• -H dj
•a o S
a; ^ £
#a fc
-o TD ?•
.5 .5 ^
b ^ t-
H— 1
JL
f5
S E
1/5 4J
U -M
.-, w
C >,
O oO
s^
1'^
U
Pyranometry
c
'E
">;
, ^
fN
O
*?!
c
• <— t
JE
j
^
•— H
I
O
_0
%
^
cS
to
1_
a;
•M
nome
u.
a?
>s
a>
"H.
a,
UJ
0)
U 3
.— 1 -f-l
3 (L)
Q a
Scattering of Ac
Waves by Tern
Structure
E
o
~H
E
o
o
o
•-H
1
o
JC
op
X
00
c
'x
's
Jrt
• *s
>s fd
-------
3.3.1 Sites 1, 2, and 3
These stations were located upwind of Philadelphia under the prevailing flow,
which was determined to be from the south-southwest to west direction. Site 1 was
located near Woodstown, New Jersey, approximately 40 km south-southwest of the
Philadelphia urban core. Site 2 was located near Hockessin, Delaware, approximately
50 km southwest of the Philadelphia urban core. Site 3 was located near Downing-
town, Pennsylvania, approximately 50 km west of the Philadelphia urban core. Figure
3 shows the monitoring station at Site 1. The monitoring stations at Sites 2, 3, 4, and
5 were identical to the one at Site 1.
The purpose of these sites was to monitor the ozone and precursors transported
into the city, both along the surface and aloft. Surface stations can provide an
indication of transport aloft, since ozone and precursors trapped aloft overnight are
brought to the ground through turbulent vertical mixing after the breakup of the
nocturnal radiative inversion. In areas upwind of the city, where emissions of ozone
precursors are generally insignificant, ozone readings between about 1100 and 1300
local daylight time (LOT), after the vertical mixing has occurred, have been suggested
by U.S. EPA (1977) to be most indicative of ozone levels transported aloft.
3.3.2 Site 4
This site was located 35 krn northwest of Philadelphia near Collegeville, Pennsyl-
vania. The purpose of this site was to determine the effect of ozone and precursors
transported from upwind areas on the ozone and precursors which would be measured
at the urban core, in the absence of the urban emissions. Measurements taken at Sites
1, 2, and 3 quantify the ozone and precursors moving into the urban area. However, on
the way to the urban core, these pollutants diffuse and undergo photochemical
reactions. Measurements taken at Site 4 thus define what the ozone and precursor
levels would be at the urban core if there were no significant emissions in the path
between the upwind sites and the urban core. The difference between pollutant levels
at Site 4 and the urban core can then be considered to be a consequence of the urban
emissions. When flow was from the west or northwest, this site acted as an upwind
site.
29
-------
FIGURE 3. Monitoring station at Site 1.
30
-------
3.3.3 Site 5
This site was located 65 km north-northeast of Philadelphia near Ringoes, New
Jersey. It was downwind of Philadelphia when the prevailing flow was from the
southwest sector and served to monitor the peak ozone concentration resulting from
ozone cind precursors transported into Philadelphia and precursors emitted in the
Philadelphia urban core.
3.3.4 Lancaster and Bivalve
Lancaster was located about 95 km west of Philadelphia, and Bivalve was located
about 90 km south of Philadelphia. The purpose of these sites was similar to that of
Sites 1, 2, and 3. Ozone data from Lancaster and Bivalve were used to quantify ozone
concentrations upwind of Philadelphia.
3.3.5 Ancora
This site was located about *5 km southeast of Philadelphia. Its purpose was
similar to that of Site 4. Under the prevailing flow, ozone data at this site would
indicate what the ozone concentration at the urban core would be in the absence of
urban emissions. It would act as an upwind site when the flow was from the south or
southeast and a downwind site when the flow was from the north or northwest.
3.3.6 Franklin Institute and South Broad Street
These sites were both located in downtown Philadelphia and collected oxides of
nitrogen and ozone data. These data were used in the study to provide information on
the concentrations of ozone and oxides of nitrogen in the urban core.
3.3.7 Somerville
This station served the same purpose as that of Site 5. It is located about 80 km
northeast of Philadelphia.
31
-------
3.4 AIRBORNE MEASUREMENTS BY DEDICATED INSTRUMENTED AIRCRAFT
The primary objective of the aircraft measurements was to obtain vertical
profiles and cross sections of ozone and precursors in order to determine interrelation-
ships of these pollutant concentrations at the surface and aloft. From these profiles
and cross sections, typical levels of ozone and precursors transported into the study
city in layers aloft over various time periods during days of high photochemical
activity could also be determined.
Aircraft measurements were performed by the EPA - Las Vegas Environmental
Monitoring and Support Laboratory between August 8 and 25, 1978. The aircraft used
was a Bell UH/1H helicopter (Figure 4). Parameters measured included Oy NO» NOX>
visible light scattering, temperature, and dew point versus altitude, location, and time.
The methodologies for measuring ozone and oxides of nitrogen are EPA reference
methods. In addition, hydrocarbon grab samples were taken for species analysis. The
hydrocarbon sampling will be discussed in more detail later in this chapter. Table 3
lists the instruments used.
The aircraft (helicopter) measurement program consisted of 12 days of intensive
vertical soundings and horizontal traverses over and upwind of the study city. The
choice of these 12 days was based on forecasts predicting significant photochemical
activity along with consistent transport from upwind areas.
On each measurement day, two flights were made: one in the early morning
hours (generally between 0400 - 0630 EDT) and one in the late morning (generally
between 0900 - 1130 EDT). Each flight consisted of vertical soundings over three
surface monitoring stations from as close to the ground surface as possible up to 2 km
above ground level (AGL), and a horizontal traverse between the stations at the
elevation where maximum ozone levels were observed. Figures 5 and 6 show the
typical flight patterns during the early and late morning flights. Exact flight paths
taken on each sampling day have been documented elsewhere (Hancock, 1978).
The early morning flight was designed to provide information concerning vertical
profiles and horizontal gradients aloft of ozone and precursors upwind. Thus, traverses
32
-------
FIGURE it. Dedicated EPA-Las Vegas instrumented UH-1H helicopter.
33
-------
Table 3. PARAMETERS MEASURED BY HELICOPTER AND INSTRUMENTS USED.
Parameter
Instrument
Hydrocarbons
Ozone
NO/N07/NO
& Jv
Light Scattering
Temperature/Dew Point
Static pressure (altitude)
Position
One liter grab samples using evacuated
stainless steel canisters
REM 612 B chemiluminescent analyzer
Monitor Labs 8440 chemiluminescent analyzer
MRI 1550 nephelometer
EG&G 137 CL-S1ATH
National Semiconductor LX3702A
(2) Collins DME 40 and (1) VOR-31A
34
-------
Shannon 9\
Memorial x Sitp
A > r' I j «.J I ' C
Air Field
- Perkiomen Valley
Air Field
Site 2 x^WilmingtonAirpor
New Castle«#•
VORTAC v
Woodstown
VORTAC
Sounding
NAUTICAL MILES 0| 1
STATUTE MILES 10
1
1
1
I,
1
J, 1
1
0 I0( 201 30
0 lOJ 20| 30|
FIGURE 5. Typical flight pattern followed during 0400 EOT flight.
-------
Shannon ~
Memorial x-site3
Air Field
— PerKiomen Valley
Air Field
Site 2 x ^Wilmington Air
Woodstown
VORTAC
Sounding
NAUTICAL MILES 101 1 1
STATUTE MILES 10)
1 L ! 1 1
1
1
1
0 101 201 301
0 I0( 20! 30|
FIGURE 6. Typical late morning flight pattern.
-------
were made only along the upwind edge of the study area and soundings were taken only
over the three upwind sites. The later morning flight was designed to provide
information concerning vertical profiles and horizontal gradients aloft of ozone and
precursors upwind, as well as the vertical distribution of ozone and precursors over the
downtown area. Thus, soundings were made over two upwind sites and also over
downtown Philadelphia.
A special pattern designed for study of a stagnant air mass was flown on August
15, 1978. On that day, two flights were made, each one consisting of a spiral sounding
near Site 2, a traverse to the downtown area, where a spiral sounding was made, and a
traverse back to Site 2, where a final sounding was made.
3.5 AIRBORNE MEASUREMENTS BY A PORTABLE INSTRUMENT PACKAGE
The second airborne measurement approach tested was the portable instrument
package as described in Chapter 2. The package consisted of a flyable, battery
operated ozone/temperature analyzer which could be placed in a light airplane with no
aircraft recertification requirement. For this project, a Cessna 172 was used (see
Figure 7). The instrument was a Columbia Instruments Model 2000 chemilurninescent
(X, analyzer. The operating gas was a non-flammable ethylene mixture (Ethychem) in
a low pressure container, which was exempt from the hazardous materials transporta-
tion regulations. The analyzer has a range of 0 to 0.5 ppm and an accuracy of
+ 0.005 ppm when operated with pure ethylene; the performance is degraded slightly
when Ethychem is used. This analyzer meets EPA reference method designation
specifications when ethylene is used but does not meet these specifications when
Etiiychern is used.
It was originally planned that a Columbia Scientific Model 2200 chemilurni-
nescent NO/NO analyzer, which was under development at that time, would also be a
component part of the portable instrument package. Unfortunately, the instrument
was not ready by the time evaluations of various measurement techniques were
under way.
37
-------
FIGURE 7. Cessna 172 aircraft used for flying the portable instrument
package.
38
-------
FIGURE 8. Portable instrument package and recorder strapped to
passenger's seat of aircraft.
39
-------
Samples to the ozone analyzer were provided by 6 mm diameter Teflon line
threaded through the cabin fresh air inlet of the aircraft. The temperature sensor was
also installed in the same fresh air vent as the sample tube. The readings from the
analyzer and the temperature sensor were recorded on a battery-operated Linear
Instruments chart recorder. The analyzer and the chart recorder were strapped to the
passenger seat of the aircraft as shown in Figure 8.
Tne portable instrument package was flown on five days in August and three days
in September. Table 4 presents a summary of dates and times of sounding with the
package and sites over which soundings were made. The August flights were made in
conjunction with the EPA helicopter. The purpose was to collect data with the
package and the helicopter instruments simultaneously at the same locations, thus
allowing a performance evaluation of the portable instrument package. The free lift
balloon ozone sounding system (to be described later) was not ready for testing until
September, by which time the EPA helicopter was no longer available. In order to
obtain some ozone profile data against which the free lift balloon ozone sounding data
could be compared, the portable instrument package was flown on three days in
September.
3.6 SOUNDINGS BY OZONESONDE BENEATH A FREE LIFT BALLOON
A third airborne measurement approach tested was a free lift balloon ozone
sounding. This method is described in detail in Chapter 2. As demonstrated in this
study, it consisted of a Mast 730-9 ozonesonde and an Atmospheric Instrumentation
Research Company AS-1C three-channel airsonde (see Figure 9). The radio trans-
mitter on the airsonde transmits temperature, pressure and ozone data to an
Atmospheric Instruments Research Company TS-2A Ground Station (see Figure 10).
The ozonesonde, in its original form, measures total oxidants and thus responds
also to SO2 and NO2 in the ambient air. During the tests an absorber column packed
with glass fiberpaper impregnated with CrO» and H-SO^, (Mast Model 725-30 SO_
filter kit) was added to the intake of the ozonesonde. The column removes SO2;
tests indicate that the removal of SO- is 100% complete if the column is
used for no longer than 540 ppm-hr of sampling. The absorber also completely
-------
Table 4. DATES AND TIMES OF FLIGHTS USING THE PORTABLE
INSTRUMENT PACKAGE, AND SITES OVER WHICH
VERTICAL PROFILE MEASUREMENTS WERE TAKEN.
Date
8/18
8/19
8/22
8/23
8/24
9/21
9/21
9/27
9/27
9/27
9/30
9/30
9/30
9/30
Time in EOT
(beginning of flight)
0902
1325
0948
1000
1618
0700
0940
0740
1500
1800
0700
1100
1500
1800
Sites
1,3
1
1,3
2,4
1
1
1
1,2,3
1
1
1
1
1
1
41
-------
FIGURE 9. Mast 730-9 ozonesonde.
-------
FIGURE 10. Atmospheric Instruments Research Company ozonesonde ground station.
-------
converts NO to NO-, which gives about 10% positive interference to the O,
measurements. The worst case during the ozone sounding days had an NO concen-
tration of 0.075 ppm which could result in approximately 0.008 ppm of O, overesti-
mation. Considering that the ozonesonde has an accuracy of + .01 ppm, its use without
also removing the NO- beforehand seems justified. The range of the ozonesonde is
from 0 to 0.50 ppm. Presently, the ozonesonde does not meet EPA equivalency
requirements.
The package was flown under a helium-filled, free-rise 350 gram balloon.
Tracking with a single theodolite was performed in order to determine wind speed and
direction. (The pressure channel allows calculation of the package altitude.)
Eleven flights were made on three different days in September. Four of the
ozonesondes were found to have linearity problems and data collected were discarded.
Six of these flights were made in conjunction with the portable instrument package for
comparison purposes. All launches were made from Site 1. Table 5 presents dates and
times of all successful launches.
3.7 SOUNDINGS BY OZONESONDE BENEATH A TETHERED BALLOON
A fourth airborne measurement approach tested was that of an ozonesonde
beneath a tethered balloon. The method is described in detail in Chapter 2. As used in
this study, it consisted of a Mast 730-9 ozonesonde and an Atmospheric Instrumenta-
tion Research Company TS-2A tethersonde supported by a 4.25 m kytoon (Model
TS-1BR-2) with a TS-2A Ground Station. Figure 11 shows the tethered balloon
ozonesonde system. An absorber column packed with glass fiber paper impregnated
with CrO., and H?SO& was also added to the intake of the ozonesonde for removing
SO,,. The range and accuracy of this system in measuring ozone are identical to those
of the ozonesonde used with free lift balloons. The combination of the ozonesonde and
the tethersonde made it possible to measure ozone, temperature, relative humidity,
pressure, wind speed, and wind direction, all as functions of elevation.
-------
Table 5. DATES AND TIMES OF SUCCESSFUL LAUNCHES OF
OZONESONDE BENEATH A FREE LIFT BALLOON.
Date
9/24
9/27
9/30
Time (EOT)
0807
1055
1207
1516
1905
0720
1630
-------
FIGURE 11. Tethered balloon ozonesonde system.
-------
The system was flown on four days for a total of eleven flights. Table 6
summarizes the dates and times of tethersonde flights. All flights were made at
Site 1.
3.8 HYDROCARBON SAMPLING FOR SPECIES ANALYSIS
The purpose of this phase of the program was to provide information on the
hydrocarbon species as they are transported into the urban area.
There were two parts to the sampling program: (1) Integrated samples at the
ground level were collected at each of Sites 1, 2, 3, and 4; and (2) grab samples were
collected aloft using the EPA helicopter over Sites 1, 2, 3, and downtown Philadelphia,
and, on one occasion, over Site 4.
Evacuated one-litre stainless steel containers were used to collect the ambient
air samples for analysis. Prior to sampling, the containers were cleaned by repeated
evacuation and purging with hydrocarbon-free nitrogen followed by heating and re-
evacuation.
Ground-level sampling at the monitoring sites was performed by using the
containers in series with a solenoid valve connected to the site's sampling manifold.
At a preset time, the solenoid valve was actuated by a timing device for one hour
during which ambient air "leaked" into the partially opened sampling container. Thus,
some time integration was achieved. The rate at which the container was filled was
not constant. At the beginning of the hour, the flow rate was about 50 cc/min. The
rate dropped to zero at the end of the sampling period.
Aircraft samples were instantaneous grab samples taken through the sampling
manifold. The sampling procedure followed, with minor deviations, was the same from
day to day.
During the early morning helicopter flight (0400 - 0630 EDT), ground samples
were taken at Sites 1, 2, 3, and 4. Three aircraft samples were taken over each of
Sites 1, 2, and 3. One of these samples was taken as close to the ground as possible
-------
Table 6. DATES AND TIMES OF LAUNCHES OF OZONESONDE
BENEATH A TETHERED BALLOON.
Date
8/2*
9/21
9/27
9/30
Time (EDT)
1112
1650
0759
1020
1117
1328
0922
1313
1613
08*5
13*1
-------
and the other two above the surface based inversion (at approximately 1 and 1.5 km
MSL).
During the late morning flight (0900 - 1130 EOT), ground samples were taken at
Sites 1, 3, and 4 and one aircraft sample was taken over each of Sites 1, 3, and
downtown Philadelphia at an elevation of about 1 km. Thus, both horizontal and
vertical hydrocarbon profiles were defined over upwind areas during both flights. In
addition, information about hydrocarbon species aloft over downtown and at the
surface at Site 4 was also obtained.
Samples were analyzed by Scott Environmental Technology for hydrocarbon
species C. - C)0 using a Perkin-Elmer 900 Gas Chromatograph.
Samples were taken on seven photochemically active days*. A total of 126
samples were taken. Table 7 summarizes the sample collection times.
3.9 QUALITY ASSURANCE
This section summarizes the procedures used to assure data quality for the field
program. Specifically, the calibration methods and frequency, station check methods
and frequency, and audit and interlaboratory tests are described for the surface
monitoring program and the airborne measurement program.
3.9.1 Surface Monitoring Program
The equipment listed in Table 2, except for the acoustic radar, was tested and
furnished to AeroVironment by the EPA prior to field use. The equipment and
recording devices were installed in environmentally-controlled shelters to meet the
manufacturers' specifications at each of the five monitoring stations.
In the beginning of the program, however, temperatures at Sites 2, 3, 4, and 5
often exceeded the upper limit of the recommended operating temperature range of
f.
The criteria for a photochemically active day are given on pp. 63-64.
-------
D
u
\-
(J
UJ
_J
J
O
U
t/)
UJ
10
Z
O
CD
U
O
05
Q
U-
O
to
O
H— I
<
8
Q
Z
ZU
al "J
LUQi
QiO
a. a,
03
+-1
OJ
to
(VI
i/5
. — i
OJ
0)
ZJ
-M
— H
+-»
(0
G
_
-"-< OJ M'S l*-i<^
i_O i- O L.O u.O i- O i- O i_O
-'i? D!? ~35! ")5! "'S3 D5 3 i?
VD OO CO «c^ t^.
oo o> — < — H cvi (vt (vi
OO OO 00 OO OO OO CT\
*
^
o
ro
J.
1
Ov
(N
(N
(V|
Z
TOTAL
CN
X)
U
J.)
"o
O
a.
E
id
t/5
73
50
-------
86 F. This was because of a combination of air conditioner malfunctions and the
insufficient capacity of the air conditioners. Those problems were corrected and by
August 1, all stations were operating within the specified temperature range.
3.9.1.1 Calibration of Air Quality Instruments
Multipoint calibration of all air quality instruments at each station was
performed by AV personnel bi-weekly throughout the measurement program. In
addition, calibration of an instrument was carried out whenever any of the following
conditions occurred: (a) the control limit for the span check as specified in the Station
Check List Log was exceeded; (b) after repair of a malfunctioning analyzer; (c) after
replacement of major components of an analyzer; and (d) when the audit results
exceeded the limits established (see Audit Section). Zero plus a minimum of four
calibration points equally spaced over the analyzer range were used to generate a
calibration curve. A "master" calibrator was used for all of the multipoint calibrations
to ensure data comparability. The purpose of the multipoint calibrations was twofold:
(a) to check the instrument linearity; and (b) to assign the values of the on-site
calibration sources so that they would also be NBS traceable.
Dynamic calibration of the air quality instruments was carried out for NO-NO--
NO and O., analyzers by use of a Bendix calibrator. For NO and NO channels, an
X j X
NBS SRM cylinder gas containing approximately 50 ppm NO in N2 was diluted to
ambient levels for the calibration. For the NO_ channel, the technique of gas phase
titration (GPT) of NO and O3 prescribed by EPA (Federal Regulation, Title 40, Part
50, Appendix F) was employed.
For the O, analyzers, the calibrator's O_ output was determined on-site by both
the GPT technique and UV photometry method (Federal Register, Vol. 44, No. 28, PR
8221-8233, February 8, 1978). For the former, the data obtained during the NO?
calibration was conveniently utilized, and for the latter, a modified Dasibi Model
1003 AH ozone monitor served as a transfer standard. Although the GPT results were
used as the O., reference method for this project, the UV photometry calibration
results were also documented.
51
-------
For the THC/CH. instruments, two compressed gas cylinders containing approx-
imately 0 and 8 ppm of CI-L in an ultrapure air were used for the calibration.
Multipoint calibration was accomplished by using an AV Model BB-100 dilution system.
The ChL cylinder gas used was NBS traceable.
The NO flowrate and dilution system of the calibrator, and the flowrate for the
AV Model BB-100 dilution system were all calibrated by use of a bubble flowmeter and
a dry test meter at the beginning, midway through, and at the end of the measurement
program.
3.9.1.2 Calibration of Meteorological Equipment
Calibration checks of the meteorological equipment were done on-site. The wind
cups were turned to ascertain that the bearings were in normal condition. The wind
vanes were aligned to the true north, true south, and true north again to calibrate the
recorder output. The local magnetic north declination of +10 was accounted for. The
temperature output from each station was checked daily by using an NBS traceable
mercury thermometer.
3.9.1.3 Station Checks
Station checks were performed daily by AV personnel using the Daily Check List
form prepared for the project. The main concerns here were to ensure that the
analyzers and the supporting equipment were in proper working order.
Zero and span were checked for the analyzers during the station check. For
NO/NO and O, instruments, this was accomplished by using the on-site Bendix Model
A J
8861 or 8851, or Monitor Labs Model 8500 calibrator. For THC and CH , a
hydrocarbon-free air cylinder and a span cylinder containing approximately 8 ppm of
CH. in ultrapure air were used.
The zero and span values for each instrument were used to determine control
limits. The control limits were a useful tool to detect early instrumental problems.
52
-------
Any instrument problems were reported by the station check personnel to the
Field Manager on the same day so that corrective action could be initiated as soon as
possible. In addition, an instrument status report was provided weekly by the Field
Manager to AV's Program Manager in Pasadena to update him with the field program
status, and to enable him to provide direction when needed.
3.9.2 Airborne Measurement Program
The EPA helicopter air quality instruments, portable instrument package, and
ozonesonde were involved in the measurement program. Calibration of each of the
three measurement methods is described in this section.
3.9.2.1 EPA Helicopter Air Quality Instruments
The NO/NO and O, instruments were calibrated before the first flight and
•rt. J
shortly after the last flight of a given flight day. The on-site transfer standards used
were cross-referenced in EPA-Las Vegas against NBS SRM prior to the program.
Multipoint calibration was performed.
3.9.2.2 Portable Instrument Package
Calibration of the portable instrument package was performed after completing
each day's flights. The reference method used was O- GPT.
3.9.2.3 Ozonesonde
The ozonesonde used in the free balloon soundings was calibrated prior to each
launch using the GPT method. In addition, an operational check was performed to
ascertain that the pressure sensor was responding properly and that the zero/span was
stable. Calibration of the ozonesonde used in the tethered balloon soundings was
performed prior to and shortly after each sounding, also using the GPT technique.
Since this ozonesonde was the same as that used for the free balloon sounding, the
same operational checks were performed.
-------
3.9.3 Audits
Quality assurance verification of the O, and NO-NO--NO systems was
j £, X
performed at the EPA-EMSL facility at Research Triangle Park, North Carolina, at
the start of the measurement program. The agreements for both systems were within
5%. No such verifications for THC/CH. were performed.
During the monitoring phase, the project was further audited by Research
Triangle Institute (RTI), under contract to the EPA. Three audits were conducted for
the (X and NO-NO^-NO analyzers and one audit for the THC/CH^ analyzers at each
of the five surface monitoring stations. The EPA helicopter air quality instruments,
and the O., portable instrument package, were audited by RTI once during the
measurement program. The supplemental stations were also audited by RTI. Results
of these audits are documented elsewhere (Pasquini and Hackworth, 1978; Grimes and
Pasquini, 1978; Pasquini and Grimes, 1978). A summary of the results is presented in
Appendix A.
The acceptable criteria for all audits were set at -tl5% agreement for the slope
of regression analysis of the audit data, and jfO.015 ppm for the intercept. Any audit
results indicating that these limits were exceeded resulted in recalibration or
repair/recalibration of the instrument(s) in question as promptly as possible.
-------
4. CHARACTERIZATION OF OZONE AND PRECURSOR TRANSPORT
INTO PHILADELPHIA
This chapter presents results of the analysis of data collected in Philadelphia
during the summer of 1978. Although monitoring at Sites 1 through 5 began in 3uly,
there were many operational and instrumentation problems in the first month of the
program. Thus, only data collected during August and September were used for the
analysis.
4.1 AVERAGE AND MAXIMUM LEVELS
Table 8 presents the average and maximum hourly averaged ozone and precursor
concentrations observed at all monitoring locations during August and September. Site
locations were shown in Figure 2 . Ozone levels were quite high, even at sites located
predominantly upwind. Precursor concentrations were generally quite low except at
downtown locations. High precursor concentrations at upwind locations were infre-
quent and usually not sustained, indicating a lack of significant large scale precursor
transport into the study area.
Table 9 presents a frequency distribution of ozone levels observed at all sites.
The greatest frequency of concentrations in excess of 0.120 ppm occurred at Ancora,
located southeast of Philadelphia, while the greatest frequency of concentrations in
excess of 0.080 ppm occurred at Site 3, located west of Philadelphia, predominantly
upwind. The lowest frequency of concentrations in excess of 0.080 ppm occurred at
the South Broad Street site in downtown Philadelphia. Sites located at Franklin
Institute, Somerville, and Bivalve also recorded a relatively low frequency of ozone
concentrations in excess of 0.080 ppm.
Figures 12 through 16 present ozone wind roses for all sites. These depict,
simultaneously, the frequency distribution of ozone concentrations and wind direction.
The wind directions associated with high ozone concentrations are easily identified
55
-------
CO
D
o
O
N-4
OL
D
Q
I
I/)
z
o
a;
z
II]
u
z
o
u
D
U
tU
a;
a
Q
UJ
Z
O
N
O
z «
< s
UJ H
> Z
oo
rt)
g
•D
>,
X
at
1
I
c
4J
O
k*
iz
"o
V)
I*
•a
'2
§
O
&
*8
£
<
&
18
>
<
1
£
1
a.
*
1
<
J*
to
£
1
V
a.
V
(/)
1
DO
*
O
Cu
s
u
s.
4J
t/1
3
2f
J^
(4
4;
a.
D
»/i
_ r^ «^ — • (N ' ' ' ' I 1
O O O O O
-H (M >^ —• — ' ' ' ' ' '
O O O C3 O
A so O o i i i i
— o O ON r^ O 1^
O 0 0 O 0 O 0
r4»*\— i — 'fMO(Nf*>
ooooooooooo
ooooooooooo
f*>fM^-t^rM— 'fNm — jj-r^
ooooooooooo
ooooooooooo
"*iON
V D
v -
| 1 , £
« c i! -
O s ui > m u
*1 21 1!
O k- fl O -"
— *rsim.»<^i7)U.i—l(/l«tcQ
g
-------
Table 9. FREQUENCY DISTRIBUTION OF OBSERVED OZONE LEVELS (ppm)
FOR AUGUST AND SEPTEMBER 1978.
Site
1
2
3
4
5
South Broad Street
Franklin Institute
Lancaster
Somerville
Ancora
Bivalve
Total Hours Observed With Specified Ozone Concentration
>.120
7
>4
10
15
13
1
*
1
1
22
0
.100-.120
29
22
43
49
32
1
13
10
1
24
2
.080-.099
51
36
112
93
43
10
19
34
12
53
31
.040-.079
317
287
447
374
216
133
184
340
107
338
374
.000-.040
888
961
829
883
548
1,256
1,178
837
1,267
817
998
Total
1,292
1,310
1,441
1,414
852
1,401
1,398
1,222
1,388
1 , 254
1,405
57
-------
0-.040 .081-.120
t
N
.041-.080 >.120
Ozone Classes (ppm)
0
I
2.5%
I
5.0%
7.5%
Relative Frequency of Occurrence
10.0%
FIGURE 12. Ozone wind rose for Site 1.
58
-------
0-.040 .081-.120
t
1
N
0
1
2.5%
1
5.0%
I
7.5%
I
10.0%
I
.041-.OSO XI20
Ozone Classes (ppm)
Relative Frequency of Occurrence
FIGURE 13. Ozone wind rose for Site 2.
59
-------
O-.OfO ,081^.120
t
N
.041-.080 XI20
Ozone Classes (ppm)
FIGURE 1*.
0
I
2.5%
1
5.0%
I
7.5%
I
Relative Frequency of Occurence
10.0%
I
Ozone wind rose for Site 3.
60
-------
0-.040 .081-.120
t
1
N
0
1
2.5%
I
5.0%
I
7.5%
I
10.0%
I
.041-.080 >.120
Ozone Classes (ppm)
Relative Frequency of Occurrence
FIGURE 15. Ozone wind rose for Site 4.
61
-------
0-.040 .081-.120
t
N
2.5%
I
5.0%
I
7.5%
1
.041 -.080 >.120
Ozone Classes (ppm)
Relative Frequency of Occurrence
10.0%
I
FIGURE 16. Ozone wind rose for Site 5.
62
-------
from these figures. Some data are not included in Figures 12 through 16 since
coincident wind direction and ozone data were not available. These data do not
significantly alter the wind direction frequency distribution. An analysis of the figures
shows that the most preferred wind directions for high ozone concentrations at all
sites are from south to west. These directions put Sites 1, 2, and 3 upwind of
Philadelphia and indicate significant transport of ozone, sometimes in excess of
0.10 ppm, from the southwest. Wind directions from north to northeast are almost
never associated with high ozone concentrations, probably because these wind
directions are often associated with extensive cloudiness and a resultant inhibition of
photochemical activity.
4-2 TRANSPORTED LEVELS OF OZONE AND PRECURSORS
The 61 days of observation during August and September were classified into
four groups of transport days based on observed ozone concentrations (especially at
upwind stations) and persistence of flow (surface and aloft). The four groups and the
criteria for classification are:
1. Photochemically active days with strong transport.
a) Ozone concentrations in excess of 0.08 ppm between 1100 and 1300
EOT or 0.10 ppm later in the day at an upwind station.
b) Prevailing flow from 180 to 315 observed at upwind stations both
surface and aloft during the 24 hours prior to 1400 EOT on the day of
interest.
2. Photochemically active days with weak transport.
a) Ozone concentrations in excess of 0.08 ppm but less than 0.10 ppm at
upwind stations.
b) Prevailing flow from 180° to 315° observed at upwind stations both
surface and aloft.
63
-------
3. Photochemically active days with no well-defined transport.
Ozone concentrations in excess of 0.08 ppm in the study area but not
at upwind stations.
4. Days with little photochemical activity.
No ozone concentrations in excess of 0.08 ppm observed.
Eight out of the sixty-one days were identified as photochemically active days
with strong transport. The magnitude of transported ozone and precursors is suggested
in Table 10 which presents ozone and precursor levels observed at the surface and aloft
on the eight days during which strong transport was observed. The wind directions
presented in Table 10 are prevailing wind directions at the surface between 1100 and
1700 EOT. Peak ozone levels in excess of 0.10 ppm were consistently observed at
upwind sites while concentrations between 1100 and 1300 EOT and average concentra-
tions aloft (measured by airborne instruments) were generally in excess of 0.06 ppm
and sometimes in excess of 0.08 ppm. (The average concentration observed at a
surface station between 1100 and 1300 EOT was suggested by U.S. EPA (1977) as an
indication of the transport of ozone aloft.)
Days with strong transport were further examined and compared with other days
in order to identify the peculiar characteristics of the transport situation, especially
with regard to vertical profiles, horizontal gradients, and diurnal variation of ozone
and precursors.
4.3 VERTICAL PROFILES
Figures 17 through 19 present vertical profiles of ozone, oxides of nitrogen, and
temperature data obtained by the instrumented helicopter at Site 1 on a day with
strong transport. The flight times shown indicate the beginning time of the spiral,
which encompasses approximately fifteen minutes. The wind data presented in these
figures were obtained by pibal tracking near the time of the helicopter spiral. Wind
direction arrows point downwind. These figures typify the profiles obtained on most
64
-------
z
I
z
aJ
U
Z
O
u
u
UJ
Qi
a.
1
k.
•a
X
ethane
Npn-M
c
Z
"o
» ' ' ' ' iAts.^1 INI I 1 1 I It7\( I K\ IA (N ' I I 1 00 VO ^ ' 'fs.fvl ' ' <
— O O O C4 fM OOO *•> O O»* O — O r-\«""\
oooooo ooo o o ooo ooo oo
oooooo ooo o o ooo ooo oo
OOOO--— OOO OO OO O — fN OO OO —
oooooo ooo oo oo ooo oo ooo
rsi — i — • i i i i i i t i iiiriiiiiii r*> oo — . i i i i t i i t o r«i *N ( ' f i • ' ' '
oooooo 4-4-r^ r^^to
ooo ooo ooo
ooo ooo ooo
— ooo oooooo oooooo oooo o — oooooooo oooooooooo
oooo oooooo oooooo oooo oooooooooo oooooooooo
— » O — * ON >£ OO f-«. rC (N OO BOv£oOeOhk^OvOr>^^)9>9O O^fC^OO-^lNoO^ff* lfc».Off^*-'»,DOOaOxfi*OO
-«-«-«0 0000 — 0 00000000000 -,_.— — — _.-C)_o 0— O — OOOO — 0
oooo oooooo ooooooooooo oooooooooo oooooooooo
-Q (/» t/i ZZ ^(/iZ Z t" ^ &
i ^ i i* & i i i i i t i & t ^ i i t i i i i i w^ t ^ :* i i i t i i iZ^ZZiiriit
is '** t ** h " *• **
"O^^JjS "O",!}— ^""Sj^S "O*~*tJ —
N. ,J\ fN|
^ "^ —4 (\J
«-• ** M «J
3 3 S 3
»0 00 00 So
< J D 3
oi
UJ (/)
ZZ
o<
N as
OH
o
(0
-------
0)
1
c
2
s
•0
I
J-
.C
5
i
0
Z
c
00
0
L*
Z
S
o
V
o
^
D
S ST
'S S
tn <
S Si
fl ™
1/1 <
fl
"o ^
"I
(U g>
* 2
IT It
J? <
Mr;
£ C
> £j
a "c
•*
o
o
**^ I—
^ UJ
-^
o
o
p jr
SO
o
o
*>-
o ^
-~
o
o
?s
o
o
o t-
SuJ
o
o
o'LU
-H
u
I
•s
ft)
s
0
o
§2
**^
4J
4->
V
rt
Q
O —• O — " O I 1 I • ' t OOOOO'1 ill — a«. o **> O I ' ' • I ' — • r^ — < — IA 1 I I t i I
ooooo ooooo ooooo ooooo
O — » — • O O ' 1 1 I 1 » O—^^OI ' 1 1 I ' (N V> 1 1 1 I < I
OOOOO OOOOO OOOOO OOOOO
8£ o ' S3 ' ' ' ' ' -^''-^ri S^'g11'111 S2S2f5S ' ' ' ' '
OOOOO OO OOO OOO OOOOOO
OOOOO OO OOO OOO OOOOOO
O O O O — < O OO OOO OOOOO— OOOOO — —
OOO OOO OO OOO OOOOOO OOOOOOO
O oQ i i i i i i ' ' ' fsjoioo'i'1111 iiiriiliitr iii &
,1111 , ?»£»*,,,,,, ,*% ,» %»%**,,,,,,
~S «2 ~H ~ £
£"3 Jit: «! § $ '
5 « s fc s «5
8|al2S 3|il£S S|||2*| SjSlgt;
6jisj§> ^yi- ^^P- ^y^-
\O (N
OJ (M & S
*-• *-i C C
3 5 4J 41
ff S1 & |
Q
01
t
8.
5X
x> Q-
£«
il
"§"0
= fc t
tj)5s 5
Cm a.
EJ 2
fi r
vt £
2=fc S
s s- *
oS >.
•O u J3
-O -C 13
ti " 41
^ 1
II I
»i 1
S2 2
^ o
u m c
Js~ •=
66
-------
NOUD3HIQ QNI/ft
L I I 1
ci
u.) -<
001
Sfc
O (/)
r-S ^
- o
X
0!>
Z
Q
01
OJ
CL
tn
Q
Z
(0 CM
00
tv
ON
a
§
o
o
s
o
o
o
o
o
(JJ
a.
o5
— tu
«»*
1o
O
U)
U
S Z
°8
o
o
01
OH
D
O
E
(nsw 3Aoav
saniuiv
67
-------
i i i i
Q
tU <*>
O UJ
Sfc
~g
o ~i
"rsi a! £-
_o
•O
D
UJ
UJ
a
in
O
Z
oo
O 3
3A09V
00
(M
f-
\r>
D
U
"*£
UJ
a:
a;
01
a,
O
•3-
O
(M
O
I
Z
O
2
UJ
(J
8
cot
— < "•
a-
O 0)
eo
4-> T)
o ct- c
« O O
U 0)
•jg .5 o
o a; («
00
U
a:
D
a
E
o
o
o
o
o
o
ir\
68
-------
oo
(N
(-
I/)
§
oo
. u
o_,
tu
CM
Qi
01
0,
_o
w—(
o
.
e v
C -H
4) ro
•xS
O O.
« 8
p
O 4-.
«M -3
o 2
k *••
O.g
—
UJ
a:
D
O
3AOQV
69
-------
transport days. During the early morning flights on this and other transport days a
substantial reservoir of ozone was observed aloft with depressed ozone levels near the
surface. The depressed ozone levels were often accompanied by slightly elevated
concentrations of oxides of nitrogen. The late morning flights, usually made after
1000 EDT, indicate some buildup in surface ozone, although higher concentrations
continue to exist aloft. In the afternoon, a nearly uniform ozone profile to the top of
the mixed layer was observed.
Figure 20 presents vertical profiles of ozone, temperature, wind speed and wind
direction at a site upwind of Philadelphia for a day during which photochemical
activity was observed but little ozone transport into the study area was apparent. The
soundings were obtained by an ozonesonde beneath a free lift balloon. The ozone
reservoir aloft in the mixing layer is somewhat less than that observed on days of
significant ozone transport, although the difference is not substantial. The tempera-
ture profile for both the early morning and mid-afternoon sounding indicates a limit to
vertical mixing in the form of an elevated inversion at about 1,850 meters above
ground level. The maximum ozone concentration observed in the early morning
sounding below 1,850 meters is 0.055 ppm, which is generally less than the levels
observed on days of significant transport. The wind direction is southerly at this level,
a direction often associated with transport of higher levels of ozone (see Figure 12).
Another important aspect of this sounding is the appearance of high ozone
concentrations above the elevated inversion. Wind flow at this level is from the
southwest, a direction more often associated with ozone transport.
Figures 21 and 22 present vertical profiles of ozone and oxides of nitrogen
obtained by the instrumented helicopter during a day in which photochemical activity
was relatively low, due to the fact that a relatively clean air mass behind a fast
moving cold front had moved into the area. The profiles indicate somewhat lower
ozone concentrations than those obtained on days of significant photochemical
activity. Surface depletion of ozone is still noticeable, however. It is interesting to
note that concentrations near the surface in the morning hours are of the same
magnitude as on transport days. Thus, until the time of mixing, surface ozone
monitors contribute little or no information about ozone levels aloft.
70
-------
NOLLD3HICI QNI/tt
-g
w
o &
Q
WN Z
ss
•s c «*
u
•o o
(Tj ,^1
0> —' j)
O» *-" A)
""•Si
4) CJ
3 -rj
15*
-Ex
N 85R
O 4^ _rt «*v
o
8
o
o
8
8 8
O ^
UJ
*/^ ^
t- ^H
Q* ... .—4 h_
_ «; HH a>
flj .— P •"
> O
•
O
(S
in
cm
D
E
c
0
(is\n sja^w) aaniuiv
71
-------
QNIA
a
UJ co
O UJ
Oj-
o Co
O
(N
O
-O
_o
Q
2° 01 (N
-H O
CN
D
O
Q
tU fN
O UJ
,
o
o
0
A
8
•^
O
z
Ann it.Ai
o
o
o
» l\ 1
K/
A/I/ V
ft Ar^v
§
•A
i
UJ
UJ
10
Q
2
5
UJ
ft!
05
UJ
a
S §
a:
t-
•z
UJ
U
O
u
O -
«**4
"O tN
00
<
C
UJ
C
*o
I
o
•
00
o
ex
•£•
o
X £
o a.
- o
4> O
c a
sji
o
f/j C
V V
ig e
"o 2
t •!-•
U. M
13 -
>x?
•
-------
i-
Q
UJ
ON
O
o
.Z
D
LU
UJ
a
>
Q
Z
oo
^
O
TO
1.O
H-
Q
LU CM
— U.I
—1 L_
ON CH
O IO
o
oo
o
o
CM
o
O
O
o
CM
o
o
o
o
o
o
o
u
o
^-^
UJ
&
D
a:
UJ
a,
£
UJ
H
I
Z
O
Qi
H
Z
UJ
U
Z
o
u
u O
J D.
2 _
£ nj
'c ^ u
•«o^
O ° C
4> ££
'•g I .3
Sfc C
4) >
f-
•£ O
CM
TM
UJ
Oi
D
a
3A09V
73
-------
4.4 HORIZONTAL GRADIENTS
Most of the data collected, with some notable exceptions, suggests that, while
horizontal gradients of ozone and precursors exist at the surface along the upwind
boundary of the study area, the features of the transported air mass are generally
homogeneous.
Vertical profiles of ozone and precursors aloft obtained by the instrumented
helicopter over upwind locations during the early morning on days of observed
transport (August 9, 19, 22, 23, 24) and six other days were examined to determine
similarities and differences. In general, while some differences were evident, the
gross features of the profiles and the average concentrations aloft in the transport
layer (above the layer of surface ozone depletion) were similar from one location to
the next, despite some differences in sounding times. Figures 23 through 25 illustrate
a group of upwind profiles obtained during the early morning hours of a transport day.
The profiles are very similar despite the time span of over an hour required to perform
all three soundings.
Helicopter transects were examined from site to site along the upwind boundary
on days of observed transport. A typical transect between sites takes approximately
20 minutes, allowing little time for increase in photochemical activity. Table 11
presents the maximum gradients of ozone and oxides of nitrogen observed during these
transects. The times shown are the beginning times of the transect. Winds presented
were obtained by pibal observation at the height of the transect within an hour of the
transect time. The gradients shown represent the difference between the minimum
and maximum concentrations observed during the transect flight. The average
absolute difference between concentrations at the beginning and end of a transect
along the upwind boundary is 0.012 ppm ozone and 0.009 ppm NO . Ocassionally, large
differences were observed along the transect. These may be attributed to point
sources upwind.
Helicopter transects from upwind areas to primary source areas near downtown
Philadelphia were examined on these same five days of observed transport. Table 12
presents the gradients observed along these transects. Most of these were flown
74
-------
NOUD3HIQ QNIA
o in
3:
I
Q
(L)
01
a,
V)
Q
Z
oo
Q
UJ -H
""•* LU
s
iu
IU
..o
Q
tu —
j.
o
o
o
O
o
o
o
o
o
o
00
o
o
vo
o
o
^~
o
I
tWw'
UJ
8
T>
0)
•fj
h
li
VJ <"
»&
l<
><«;
•°R
TD ON
|1
'«32
*-> ^^
•fi ^
O v>
2 3> .
h 3 4)
5
-------
•s-
NOIJLDHtfld QNIA
.0!
i
D
UJ
UJ
a
00
Q
*«M
s*
•o
.
^ 3 4>
2
O 1-
tu -
"3
S
U
1
UJ
Z
O
N
O
N (M >
O -
^ o> w>
o •- .5
111
£S|
—•* 4-* O
rt 0.0
>££
CN
UJ
OL
3
a
(isw HAoav
aannnv
76
-------
QNI/ft
Q
UJ
o
• o
»-^
X
~i
-r> Z
G
UJ
UJ
IX
in
O
2
oo
(^
ON
§
Q
UJ f>!
•» UJ
<*> t_
10, C
o to
UJ
C*
a;
UJ
a
s s
Q
UJ
O
o
o
00
o
o
NC
o
o
•fl-
ea
J
UJ
O
N
O
o
8
fSI
1
ii
il
5S
*-
l<
*!{
X) 0\
4) -H
Is
I-
O v>
"3
^g S
g°^
8«*
•s*?
iP^ e
111
kfct
•«3 S.8
i§^
I2S
(N
UJ
a
a
C
(1SVM 3A09V SSH13VM) 3QniinV
77
-------
UJ
Q
fe
UJ
I
H
U
z
o
6
Q
UJ
UJ
10
cQ
O
UJ
a
o
on,
*•»<
z
a.
O
to
UJ
Q
1—4
X
o
Q
UJ
Z
N
O
U.
UJ Q
a! a,
OD
I
o.
£
0
<
c
o
icentra
o
U
O
E
a.
o
<
g
JJU3DUC
u
c
o
N
o
5
s
s •» 8 r,
il£l*
1*1^
, i
&£ fe 8
jQ 3 t-i C
< a "
IP
£
Is
£
£ ftj u *•}
3 ° aJ .° c
«J J3 £J •< L.
1 tl ..,
Q "
G
gl
H
s ^
c 3
H
1U
Q {;
S t>0
u
**> 3.
i/i •-*
*«§
> w
s ~
sl
S E Q
i^PS
-85-1
?^E
-c"^
SJS S
J
U
fl
a
OOO-- OO OO OOO OO OO OOO OO O O
oo— — oo o^o ooo oo oo oo<*> oo o o
OOOO OO OO OOO OO OO OOO OO O O
OO OO (N
OOOO OO OO OOO OO OO OOO OO O
.« ..» »*- »- 0000 .0. ..00
OOOO OO OO OOO OO OO OOO OO O O
OO— O (M(N --(N Of^O O — ' OO OOCM O — ' O —
OOOO OO OO OOO OO OO OOO OO O O
OO OO QO OQ •il'N. 0^^ \D •&• \D f^sfi h^oO oor^OO vflr^. 0>
OOOO OO OO OOO OO OO OOO OO O
OOoOr^-, r^^ SOOO ^ *> ^ f^^ ^ (N 00 f*> (M ^ *> ^ '
OOOO OO — >O OOO OO OO OOO OO — •
OOOO OO OO OOO OO OO OOO OO 1
rgo->r^ N.ON a-o> fM^(N VODO m ^ oooa- — • (N CN fN W)
oQ 00 OO OO i-
JU
2
(0
78
-------
o
&
a.
CO
H
on
Z
<
a;
H
a
z
o
H
a.
O
Q
U-I
UJ
<^)
02
O
Z
01
a
o
a;
a)
al
N Ou
OP
1
a.
*-*
c
o
n)
*j
C
4J
U
o
i
CL
"o
c
(fl
c
8
c
U
c
s
0
„
o
C
jllll
3 |
1~ g
la"
ll
Q"
c
D
6 4) U £J
3 3 £ g1 v
2 a s ° s
5 * Q H
1^5
-0 Q
ii
Q *-
•o
c
D
B* W
4J
<*> T3
5 ,-.
*J
Q «;
4J S
" Si
i IQ
t/l f— i LU
||I
c s^s
SJ 4) O
S *- ffl
t« 8
^-J
N.-- *-* O^OO r-vCTv O
^•r^ — o-a- oo — <— ' o — « —•
oo o oo oo oo oo o
^jO sD 4 r>*o oo o— o— — •
oo o oo oo oo oo o
OQWi — OSOO V\f-. ONOO Tvtf^ — r
OOC^ sO OO9\ sOfx (v. OO SO 00 00
oo o oo oo oo oo o
oo o oo oo o — < oo o
oo o oo oo oo oo
SNSO ON *NfN 0«M -so OOQN
K, >A fo rsi _. »o rv r-.
-------
during the late morning hours. Table 12 indicates that average ozone concentrations
at the transect level over upwind areas were nearly identical to those over downtown
locations while NO levels were only slightly lower at upwind locations. The average
A
absolute difference between concentrations observed aloft along the upwind boundary
and those observed downtown was 0.011 ppm for ozone
somewhat higher differences observed along the transect.
and those observed downtown was 0.011 ppm for ozone and 0.020 ppm for NO with
While horizontal gradients aloft may be generally minor, substantial gradients
exist at ground level along the upwind study area boundary. Table 13 presents
concentrations of ozone, oxides of nitrogen, and non-methane hydrocarbon (NMHC) at
0600, 0900, and 1200 EOT along the upwind boundary on days of observed transport.
Ozone gradients of 0.030 ppm are common while precursors, although at low concen-
trations, also exhibit substantial differences. The differences in ozone and precursor
concentrations observed at surface stations are probably due to differences in source
strengths of ozone scavengers and precursors upwind and in the vicinity of the surface
monitoring stations.
4.5 DIURNAL VARIATIONS
Diurnal variations at all sites are similar on days of strong observed transport of
ozone into the study area and differ signficantly from diurnal variations on other days,
especially between sites upwind and downwind of Philadelphia.
Figure 26 shows the diurnal variation of ozone at three sites on a day with strong
observed transport. The diurnal variation and levels observed at other upwind sites (2
and 4) were nearly identical to Site 3. This figure shows essentially no difference
between Site 3 (upwind boundary) and Site 5 (downwind boundary). It should be noted
that the flow on August 24 (west to west-southwest) may not have placed Site 5 in the
path of the Philadelphia plume. The downtown (South Broad Street) site shows roughly
the same variation with slightly depressed values. The consistency of the diurnal
variation over the study area indicates that the mechanisms of ozone formation and
transport are large enough in scale to mask the effects of individual source areas.
However, the effect of a Philadelphia ozone plume probably was not recorded by any
of the sites, since most of the stations were concentrated in the upwind areas.
80
-------
Q
D
t_.
£
XH
o°
Z to
O Z
Q^
||
CO *•"
0£
'E o
o!cri
c ^
°Q
§0
S n
p <
oj bu
Q/ Si}
Q uJ
X
z<
N^*
Oz
z o
O ffl
H Q
g|
§ D
Otf
u<
•
1
1
§
£
0)
1
D
1
&
o
fa
z
0
*
O
?
N
O
»
i/I
ftj
J/l
4)
l/l
•3-
D
i/i
o>
t/i
(N
4)
a
l/l
.- _
CD "J3
si
•a-
«
-
«
IB
rt
O
«- »no
o o o
Jt^O
o o o
S S S
^™
o o' S
o o o
o o o
o o o
o o o
— O 0
o o o
--(MO
o o o
— — Csl
o o o
O O M)
o o o
O (M CM
O «n fx
ooo
o o o
(N iA o
ooo
^ VO C'J
§§°
1
ooo
ooo
ooo
ooo
SS5
ooo
ooo
555
00 M
O — •
OOO
OOO
OJ (N T\
OOO
(M
•a- oo oo
—• ff\ \D
ooo
(M J-
* 00 -«
§(N *A
O O
ooo
ooo
ooo
\e ^ (N
0 0 —
1
0 -• 0
ooo
O — ' O
000
SS5
o *«o — «
ooo
sss
000
o r^
o o
•=»• ^J-
— i f^ o
0 0 -H
O 00 o
O -H K.
ooo
— 00
o ^~
o o o
ooo
I
—
o o IS
fW 0
o o ^
—. — . o
ooo
ooo
ooo
5S5
f-v fSI O
ooo
ooo
00 •£> f^
O (*> O
ooo
(NJ — • vD
ooo
O O iA
ooo
o oo o
8—- ^
o o
N, OO M
o 3- r-
ooo
O CM 00
§ra >A
o o
5 S ?
ooo
ooo
fN
b£
ON fN O
ooo
o — • o
-H O O
fM--«
ooo
ooo
ooo
55S
(M N K
(N —i O
ooo
(\J ff\ *• H
ooo
1"-. d- d-
000
ooo
li
O 00 d-
O rv oo
ooo
— fX CN
•* vfi 00
ooo
ooo
ooo
ooo
1
«• (M O
ooo
J- -H O
ooo
— 0 0
ooo
ooo
ooo
*»
rg -< —
ooo
ooo
555
(N O
§00
- ,°.
4- oo
o o —
23g
ooo
? 5 5
ooo
ooo
\f> ON (N
O O "^
1
-H O M
ooo
-H — rt
ooo
J- IX, 00
ooo
-00 a\ oj
September 6
-. — o
ooo
(S| (M —
ooo
fN*<*>
ooo
(M — —
ooo
O fM —
ooo
O it Jt
ooo
ooo
-------
s?
o
o
X
3
(fl
o
OJ
*2
«?5
*••
0)
<1J
•fc
on
•o
2
is
*-•
P I
Q
»- -o
^^
r<^ 00
«k
«£
«*•
+j M
rtJ ^
g 3
§1
0<
° k
I I
ss
'S*5
5-o
-S
£S
.2 J
O o
VO
r>«
OJ
a:
D
o
(uudd) 3NOZO
82
-------
The diurnal variation of ozone at the South Broad Street Site on August 24
corresponds to the variation in NO levels. A decrease in NO levels from 0900 EOT to
1100 EOT is accompanied by the observed increase in ozone concentrations. No such
direct relationship existed at Sites 3 and 5 on this day since both sites recorded low
NO levels (less than 0.01 ppm) all day long. Acoustic radar measurement of mixing
height at Site 1 shows a continuous increase in the depth of the surface mixed layer
from 30 rn at 0800 EOT to greater than 1,000 meters by 1100 EOT. This period of
inversion disintegration corresponds closely to the period of rapid ozone increase at all
three sites shown. Average ozone concentration aloft, as measured by the instru-
mented helicopter was 0.07 ppm, a level reached at upwind stations during the period
of inversion breakup.
Figure 27 presents the diurnal variation of ozone at the surface on a photo-
chemically active day with weak transport of ozone or precursors into the study area.
Prevailing wind direction was southeasterly making Site 1 representative of concentra-
tions on the upwind side of the study area. A supplemental site at Ancora, New Jersey
showed a similar profile. Also shown are concentrations for the South Broad Street
Site and Site 4, which was downwind of Philadelphia on this day. Comparing this with
Figure 26, it is apparent that levels upwind and downtown are much depressed without
the benefit of transport, while the concentration at Site 4 is of the magnitude of
concentrations observed at upwind stations on transport days.
The diurnal variation of ozone during a day of little photochemical activity is
illustrated in Figure 28. It differs from the transport day shown in Figure 27 primarily
in the magnitude of the peak concentrations. The time of the peak is still rather late
in the afternoon. Other days of this type show great variability in the time of the
peak.
Table 14 presents examples of ozone peak times on three types of days: days
with strong transport, days with (relatively) no transport, and days with little
photochemical activity. In general, ozone maxima seem to occur later in the day in
cases of strong transport, although the limited sample size and existence of several
exceptions prevent generalization of this observation.
83
-------
,
•§<
C 0)
V 0)
o
g.s
is
-C O
+3 N
D O
4-> 0>
- s
irt rd
•s
E
a o
c o
u •«->
UJ
od
D
O
(Wdd) NOI1VH1N3DNOD 3NOZO
84
-------
:—O
(0
•o
a
o
4;
<-»
i/5
T)
(d
O
oo
^
£
-r
(N
*-•
"»
&
3
£
>>
oo
.2 ^
n) c>
S
"3 o
EJH
00
3NOZO
85
-------
z
o
p
0!
Z
UJ
U
Z
o
U
UJ
Z
o
N
0
5
D
^
X
<£
U,
O
P
Q
EU
UJ
s
P
j-'
»— 4
_Q
ctj
J 1
.? §
^ E .I-1
+-• .c .i
I-l
*£
d) 4-t
> L.
+3 2
U en
< 0 C
_^z 2
ftl ^~ '
° '$ "S
'i «.§
•^ ^r!
X i
+-> z*
J2 j»
a
_E Q
j— . UJ
E
SO)—-
•| g Q.
nj O ^
S
D
fl
Q
^
.E Q
HUJ
^3--
^
r^«NvOOsOS 'r^CTvONiA i^ONON^M)
O — i O — < O OOOO OOOOO
^ssss ^ss§§ ^ii§§
O' 1 'O OOOOO OOOOO
O O OOOOO OOOOO
oo -ct- oO t-x. vD vX) ^ (N •—< fx. ir\ o
1 1 1
0 00
O 0 O
(N lf> .*
OOI 1 'VO in^O-HO^ CTNfMOOfO— H
— i OO OOOOCMOJ — i •* O <^\ r<^
— ( O — < -H O — 1 — 1 — 1 -H — ( — 1 — 1
ONCN^C^ OVOX^J-VO ONOV^^SD
"^^* •"-< (VJ {\J ^^^ "*^*- ~~* C^J CN "^^** "^*" "™* Cxi C^J "^^
oooooo oooooo oooooo
— i CS co,
OOOOO
OOOOO
>— 1 -H O O •— 1
OOOOO
^ _ ,TN ^. 1^
^jd-Ci CT^-Ci
OO OO C3N
OOOOO
OOOOO
"^ ""\ ^~ vO tx.
— t vD
1
o
o
S tN 0 f5 2
O — 1 — 1 — 1 -H
^ ^
*^^» »-H (V) (VJ "*^^.
OO "^**. ^*** *^*^ O^
OO 00 OO
*
86
-------
C
O
U
(d
Q^ „__,
4~> cd
+-• 0
13 'e >^
t ±!
"£ Ji: .-
'> u M
Jr'
"J CL
> i-
'•*•' a.
t/ (^
< o c
z3 jr t—
U '5 ^
1 £'J
O ^ i
4_j •— H
j? *
a ^
OJ
._
t~) &0
"^ 0
>% is -1-1
•3 ^ °
u x: ^
!!(?
U >£
^ -* £}
O
a.
^,
£ Q
H iB
D *u ^^
i§i
nj O-2-
s
a>
•M
fd
Q
EQ
HfTl
O^
£
o <*>
o o o o o o o
m — i ITN J- t/-s r<> — . i^ j- U-N
— CMCM^~-CM — -CMCM-^CM
00 OO ON 00 00 ON
1 IOOO OOOOO
OOO OOOOO
ir\|v.oo "NJ-r^r^-*
1 ^ ^
^D ^^
^3 ^5 ^3
O CM CM
I if\r^O OOOOO
c^N "~^ C^ rO ^D ^3" OO ^^
__H_ OOO— i O
CMCMUNOh^ CMCMUNOI^
•~^-H— iCNCM ~^—i'-iCMfM
oO "^"^^ *^*^ "**^*« *^^«. oO **^^ ^*^^ *^*N,. *^-^«
OOOOONON OOOOONON
IOOO 1 OOOOO
OOO OOOOO
I^OO^X) OOOtv.vOvO
0 0
0 O
-d- •*
i i^\ — i r^ i ooooo
o — < CM vjjoooor^
' — i —H • — t O *— < O —^ O
ONONr«^^}-^O ONONr<^^-\^>
•~ iCMCM — • -^- — i CM (N — •
00— ~,~~~ .ON OO --^ -— , — ^ ON
OO CO CO CO OO OO
^ jc-v%
•'-' fti (1)
3 0 £
° >; -1-"
t/) cC on
i
87
-------
4.6 CASE STUDIES OF OZONE AND PRECURSOR TRANSPORT
This section presents analyses of two cases of strong ozone transport into the
Philadelphia area which illustrate transport situations of particular importance.
Synoptic scale trajectories referred to in the following discussion were drawn
using the assumption of geostrophic flow (motion parallel to isobars). Pressure data
was obtained from three hourly National Weather Service surface maps. The
geostrophic assumption necessarily means that trajectories are calculated for parcels
above the friction layer (above approximately 500 meters AGL). A more complete
discussion of trajectories can be found in Saucier (1955) and Pettersen (1956).
Mesoscale trajectories were derived using surface wind information from five
locations within the study area. The interpolation scheme uses weighted averages of
the wind components. The weighting factors are inversely proportional to the square
of the distance between the air parcel and the wind site. More weight is given to wind
observations when they are made directly upwind or downwind of the air parcel than
when they are removed to one side. This feature is included to reflect the tendency of
winds to change more rapidly in cross-streamline directions than along the
streamlines.
14.6.1 August 22, 23, 24
This series of days represents the only extended "transport episode" observed
during the monitoring period for which data are available. It illustrates the
cumulative effect of a somewhat prolonged transport situation.
High pressure centered over the study area on August 22 moved to the south and
was centered over North Carolina by August 24. A frontal passage on August 25
terminated the episode. Figure 29 shows the surface synoptic weather situation on
August 23. The location of the high over the Washington, D.C. area was also observed
on several other strong transport days.
88
-------
FIGURE 29. Synoptic situation, 0700 EOT, August 23, 1979.
89
-------
Figure 30 presents trajectories of air parcels arriving in Philadelphia at 1400
EOT on August 22, 23, and 24, respectively. As the high pressure cell moved over the
study area and then to the south, flow changed from variable, mostly northerly, to
southwesterly. The last two days show parcels approaching along a broad, anticyclo-
nically curved arc. This trajectory is typical of most of the cases examined.
Using five surface wind stations, surface trajectories for selected air parcels
within the study area were developed. Figures 31 through 33 present study area
trajectories for three parcels: (1) an air parcel at the upwind boundary at 0700 EDT;
(2) an air parcel located near downtown Philadelphia at 0700 EDT; and (3) an air parcel
at Site 5 at the time when the maximum ozone level was observed at that location.
The mesoscale trajectories for August 22 indicate that surface flow was somewhat
weaker than on August 23 and 24. Air parcels located at the upwind boundary (Site 3)
at 0700 were advected toward the east, across the primary source areas (downtown
Philadelphia) and to the east or southeast. Air parcels located over downtown
Philadelphia at 0700 EDT were advected from the west and southwest on August 23
and 24 but from the north on August 22. In fact, a southwesterly flow regime did not
establish itself until August 23. These same air parcels were then advected eastward
over downwind areas generally arriving at locations 40 km or more downwind by early
afternoon. Unfortunately, no monitoring was performed in that portion of the study
area. High ozone concentrations were recorded at Site 5 on these three days (0.11,
0.11, and 0.13 ppm on August 22, 23, and 24, respectively). Backward trajectories
from Site 5 were developed for the hours during which these concentrations were
observed. The trajectories indicate that these air parcels did not traverse metro-
politan Philadelphia. Instead, the parcels approached Site 5 from the west to
northwest over primarily rural areas, although the urban centers of Bethlehem and
Reading were upwind. Since high ozone concentrations were observed at Site 5
apparently without the benefit of a trajectory over the Philadelphia source areas, the
air parcels shown which have a trajectory over these source areas should have even
higher ozone concentrations downwind.
Vertical profiles obtained upwind over Site 2 by the instrumented helicopter
during the early morning hours are presented in Figures 34 through 36. Average ozone
concentrations aloft increase slightly from day to day. Late morning flights actually
90
-------
FIGURE 30. Estimated 36-hour air parcel trajectories beginning 0200 EDT on
August 21, 22, and 23 and ending 1400 EDT on August 22, 23, and
respectively. (•) indicates position every 3 hours.
91
-------
MBCMHOMMEM7 MC
IO Mile
-\
10 Km
I7 SOMERVILLE
PENNSYLVANIA
|LANCASTER
MARYLAND
NEW JERSEYS
BIVALVE
FIGURE 31. Surface trajectories for three air parcels located within the study
area on August 22, 1978. Numbers shown indicate clock hour in
EDT. Times and locations of the air parcels tracked are as follows:
: 0700 at Site 3 (Upwind Boundary)
: 0700 at South Broad Street (Downtown)
: 1700 at Site 5 (High Ozone Level)
92
-------
10 Km
PENNSYLVANIA
LANCASTER
07
8/22
18
•—
MARYLAND
FIGURE 32. Surface trajectories for three air parcels located within the study
area on August 23, 1978. Numbers shown indicate clock hour in
EOT. Times and locations of the air parcels tracked are as follows:
——: 0700 at Site 3 (Upwind Boundary)
: 0700 at South Broad Street (Downtown)
: 1700 at Site 5 (High Ozone Level)
93
-------
FIGURE 33. Surface trajectories for three air parcels located within the study
area on August 2k, 1978. Numbers shown indicate clock hour in
EOT. Times and locations of the air parcels tracked are as follows:
: 0700 at Site 3 (Upwind Boundary)
: 0700 at South Broad Street (Downtown)
: 1700 at Site 5 (High Ozone Level)
-------
Q
U4
O
a
NOLLD3HIQ QNIA
^ OS
S i
O
UJ
o SH
•—i flu
O
| "S
* £
00
H
a
UJ (N
OO UJ
.-i w* C
O t/5
D
a
I/) >
'•^ U *=
^
UJ ^
_O
Q
UJ CM
00 UJ
O
O
O
00
O
O
vC
O
O
4t
O
O
i_ C O
5 ° k
^•H a
ill
•o oo *
o "J o3
l||
i/) ^
1> L> C
ra 4> 2
|!g
—« ^-> o
Cv ^^ tJ
UJ
QL
nsw
aaniinv
95
-------
NOLLDHHId
a
UJ
.28
t/1
a
z
X
I-
o£
O
z
T>
Is
* ll
in v>
~8
oo
tx.
t/5
D
O
a
tU (N
UJ
Q
UU CM
O UJ
$fc
o in
L°o"
U.1
IX'
u.1
a,
I
o
o
o
00
o
o P
>* o.
°-£
|S«
•s.s.^
«/>">•£
£.2
iu K
§ UJ
<=! g g
g
o
CN
O
-------
1 1 1 1 1 J.
oo
fv.
(N
H
D
a
Q
a) CM
o tu
•^ H
Si—
t/i
o
o
"to
£
Q
UJ
UJ
cu
V>
C
*-M
!*
O
z
XJ
0)
•M
V
e
D
i^
*-»
(/)
C
.«H
C
O
x:
««
8
^<
"J .5 -
a:
UJ
a
•2 «-
O (/»
i- C O
&.. Q.
4-* MM ^
•DO ?
C *f\ ""*
rt ct- fl
«at
O ™ a>
N
CJ
DC
(/> "'
4> i_
oo
o
-
o
NO
o
o
o
•
1
UJ
o
fyj
0
SO
^
UJ
QL
D
a
o
0
£i
t->
••^
CM
O
fM
o
O
•^
•a-
O
o
8
CM
8
o
o
§
gannnv
97
-------
show a decrease in both peak and average concentrations aloft from August 22 to
August 24. At the same time, surface concentrations were increasing from day to day
as shown in Table 15. This table presents 1100 EOT and peak surface ozone
concentrations as well as average concentrations aloft at upwind sites during the early
morning for the period under study. Concentrations aloft were determined by taking
the average in the transport layer (above the layer of surface depletion). It can be
seen that while transport of ozone aloft into the study area in the early morning
increases slightly, ozone at the surface in the late morning increases substantially.
4.6.2 August 19
On several transport days, oxides of nitrogen concentrations in excess of .05 ppm
in the lower 300 meters of the atmosphere were observed. August 19 illustrates such a
case. The NO peaks are often associated with depressed ozone concentrations at the
same altitude. August 19 also shows dramatic increases in ozone levels aloft through
the morning hours. Flow over the study area on this day was generally westerly around
high pressure centered to the southwest. Figure 37 presents the estimated 36-hour
trajectory of an air parcel arriving in the study area at 1400 EOT on August 19. The
parcel approached from the west and may have passed over the heavily urbanized
areas around Washington D.C., Baltimore, and Wilmington. The speed of the parcel
was slow, thus its residence time over these cities was relatively long increasing its
ozone and precursor transport potential.
Surface trajectories for selected air parcels within the study area were
developed for August 19. They are presented in Figure 38. The air parcel located at
the upwind boundary (Site 2) followed a track to the south-southeast until a stronger,
more southwesterly flow regime became dominant in the early afternoon hours. The
parcel then traveled rapidly toward the northeast. The air parcel passed over the
downtown area at 0700 EOT followed a similar downwind trajectory after approaching
from the northwest. An air parcel at Site 5 at 1700 EDT, at which time an ozone
concentration of 0.11 ppm was observed, appeared to approach from the west over
primarily rural areas although Allentown was generally upwind on this day. This
trajectory is somewhat different than others after 1200 EDT. The flow that advected
the other parcels rapidly toward the northeast after this time was apparently a
98
-------
CN
Q
Z
CM
CM~
CM
H
(/}
2
a
Z
o
)
Z
o
Q
Z
5
a,
D
t-
E
CL
CX
C
Z.
O
a;
t-
z
a)
O
Z
O
O
uJ
Z
O
M
O
rt!
+"*
<4-t
O
<
0)
u
03
i-
J!
L.
Q tf\
O O O
.
—i CM CO .3-
O
UJ
o
o
o
>,
m
m
4->
(fl
a.
o
u
-------
FIGURE 37. Air parcel trajectory beginning 0200 EOT August 18 and ending
1400 EOT on August 19. (•) indicates position every three hours.
100
-------
0
10 Mile
-I - 1
0 10 Km
13 8/18
r
17 SOMERVIllE
1 LANCASTER
MARYLAND
PENNSYLVANIA
i
NEW JERSEY
• ANGORA
BIVALVE
FIGURE 38. Surface trajectories for three air parcels located within the study
area on August 19, 1978. Numbers shown indicate clock hour in
EOT. Times and locations of the air parcels tracked are as follows:
—: 0700 at Site 2 (Upwind Boundary)
: 0700 at South Broad Street (Downtown)
: 1700 at Site 5 (High Ozone Level)
101
-------
mesoscale phenomenon (e.g., modified sea breeze) affecting only the southern portions
of the study area.
Figures 39 and 40 present vertical profiles over Sites 1 and 2 taken during the
early morning. Figure 41 presents vertical profiles over Site 1 taken during the late
morning. The early morning profiles indicate an ozone peak in excess of 0.060 ppm at
an altitude of 1,000 meters. However, average ozone levels in the lower 1,000 meters
are only about 0.040 ppm. Perhaps more important, NO concentrations between 0.01
to .05 ppm, not recorded at the surface, were observed by the helicopter. Simulta-
neous pibal wind data indicate weak flow from the northeast in the lower 500 meters
(NO layer) and strong flow from the west-northwest at 1,000 meters (high ozone
layer). Temperature profiles obtained from the instrumented helicopter indicate a
surface-based inversion up to 250 meters. The flow in the ozone layer is the typical
transport direction, while the flow in the NO layer puts Philadelphia and its precursor
A
sources upwind of the monitoring sites.
A dramatic increase in ozone aloft is evident in a late morning sounding over
Site 1 shown in Figure 41. While surface ozone and oxides of nitrogen concentrations
remain low, somewhat elevated NO concentrations are still evident aloft. These
X
elevated NO concentrations might be due to impacts from an elevated plume or
plumes from sources located west of the Delaware River. The atmosphere above 200
meters appears well mixed at this time (0919 EOT): temperature profiles indicate a
nearly adiabatic lapse rate and pibal data indicates west-to-northwest flow and wind
speeds are in excess of 5 m/s. Acoustic radar data indicate the top of the surface-
based mixing layer is 130 meters. Thus, the surface layer is still isolated from the
transport aloft. Acoustic radar data indicate a rapid rise in the surface-based mixed
layer the following hour and a simultaneous increase in surface level ozone was
recorded.
v
The sequence of events on August 19 suggests the mixing aloft of two distinct air
masses, one containing substantial ozone levels and one containing somewhat elevated
NO concentrations. Following this mixing, a substantial increase in ozone aloft was
recorded. Later in the morning, after mixing to the surface would have occurred,
elevated ozone concentrations were seen at the surface.
102
-------
•—V •»- '-»- -9-
a
aJ
o
o
aJ
O (/)
•8
O
z
Q
IU
Q
Z
5
X)
a;
*->
c
o
E §
2°
+-• ^
> M
.S o
00
as
to
D
O
ID
G
al
Q
al
u-\
O
o
o
OO
o
o
u
O)
oi
05
OJ
CL
S
01
E
Q.
a.
LU
O
Nl
O
o
(M
o
o
o
o
o
o
o
o
o
o
o
o> —.
22
•§§
?! oo .
5 3 _«;
to £
u. c o
a> O *-
O-tl 0.
E P^ "^
O rri C
C *^ *^^
o> ao
•S.^.E
i/i
a;
TJ
o o £
5Ss
8 8-°
W O «i
ax
«*>
a]
OS
t
dsw
aannnv
103
-------
NOLLDHtflQ
Q
UJ c^
O UJ
"^ H
-a- C
o to
O —V
*{N <»
Q
UJ
UJ
a,
to
a
Z
M»4
5*
oo
ON
to
D
a
D
Q
UJ
o to
oj
O
z
UJ
05
05
UJ
a<
•2 S
UJ
I-
. o
•a os
4) -H
12
•S ^
O )
Sg -a
c +- £
O "3 dj
N -j >
o .t; .
Q
UJ UJ
(N tl
St-
)
O
O
o
00
o
o
•3-
O
o
-------
Q
UJ —
o uj
O I-
^pi to""*
O I/)
J 2 1 I I JLYW
NOI1D3HIQ
X
L.O
Q
UJ
UJ
oo
r^
00
-
- CM
O
D
o
o
o
Csl
o
o
O
O
O
o
o
00
o
U
UJ
05
Qi
UJ
a,
O UJ
a:
i-
o g
Z
O
u
UJ ^
g w'c
ttfl—. 'S
o
+•> •«-> CO
O 0)
I*
OJ ._
XJ
w u. c:
~ o o
4> O 5.
O 7; 8
4) .5
-
T) O
• flj U
U
UJ
a;
D
a
(ISVM HAOBV
aaniuiv
105
-------
*.7 HYDROCARBON DATA
The study produced two sets of data on hydrocarbons in ambient air. The first
set consisted of total hydrocarbon and methane data collected at five surface
monitoring stations. These are reported as hourly averages in units of parts per
million carbon (ppm C) (methane equivalent).
In addition, about 135 grab samples were taken into evacuated sampling
containers and analyzed for a sequence of hydrocarbon species by Scott Environmental
Technology. Some of these samples were taken at ground level and some from the
aircraft at various altitudes above the monitoring sites. Two pairs of matched samples
taken mid-way through the program were split, with one set being analyzed at EPA's
Environmental Science and Research Laboratory (EPA-ESRL) in North Carolina. Aside
from this one test there was virtually no replication of samples. Since the unreacted
composition was of primary interest all samples were taken before noon; thus little
photoreaction is expected. For many samples, hydrocarbons which are within the
capability of the chromatographic system were not found in measurable concentra-
tions.
4.7.1 Data Validity
The hydrocarbon monitoring systems used in this program sensed total hydro-
carbon and methane separately. Methane is segregated for four reasons: (1) It is
present in large concentrations (~1.4 ppm) even in unpolluted air; (2) it is quite
unreactive in photochemical processes; (3) it is emitted into urban air from leaks in
natural gas lines; and (4) the air quality standard is stated on a non-methane basis.
Non-methane hydrocarbon (NMHC) data were obtained by subtraction of the methane
concentration from the total hydrocarbon concentration. Although this is the standard
procedure for hydrocarbon monitoring it can lead to large errors in NMHC when
methane constitutes a large fraction of the total hydrocarbon. For example, if the
total hydrocarbon were found to be 2.3 ±0.1 ppm C (~5% error) and the methane was
found to be 2.0 + 0.1 ppm C (~5% error) the subtraction would yield 0.3 + 0.1 ppm C of
NMHC (30% error). Thus, the NMHC concentration found by subtracting the methane
measurement from the total hydrocarbon measurement has a larger percent error than
106
-------
the two measurements have separately. There is added uncertainty because the two
concentrations are measured from separate, not necessarily identical, air samples.
Another problem arises from the units used in the various measurements. Since
the average molecular weights of the mixture of hydrocarbons which make up the total
is not known it is necessary to report NMHC as parts per million carbon (ppm C),
relying on the fact that the flame detector responds approximately according to the
concentration of organically bound carbon in the sample. The speciation analyses were
(except for the samples analyzed by the EPA) all reported as parts per million
compound (that is, by moles). To make a comparison of the hydrocarbon breakdown
with the NMHC derived from the continuous monitoring instrument it was necessary to
multiply the concentration of each hydrocarbon by its carbon number to obtain ppm C,
then to sum all species concentrations. To the extent that many hydrocarbons present
in too small concentrations to be quantified might collectively make a significant
contribution to the NMHC, this sum would fall short of that reported by the continuous
monitor.
The two samples analyzed by EPA-ESRL and Scott are presented below for
comparative purposes. One pair was time-integrated and one-hour samples and one
pair was helicopter grab samples. The Scott samples were converted to ppm carbon to
compare directly with the EPA-ESRL samples. Also shown is the NMHC concentra-
tion reported by the continuous surface monitor for comparison with the integrated
samples. The values were:
Concentration (pprn C)
Date
Aug. 24, 1978
Aug. 24, 1973
Hour (EOT)
0500-0600
0611*
Site
3
14
NMHC
EPA
Analysis
1.58
1.59
Scott
Analysis
0.55
1.65
Surface
Monitor
0.1
—
*helicopter grab samples.
107
-------
The comparisons are not very encouraging and the reason for the discrepancy is
not known. Although the total NMHC reported by EPA and Scott correspond well for
the 0611 grab sample, concentrations of individual compounds do not. The EPA and
Scott samples all showed substantial amounts of two oxygenates (acetaldehyde and
acetone) although the agreement on concentration was not close. These oxygenates
may be artifacts. If so, they must be introduced by the sampling system rather than
the analytical instrument since they appear in both EPA and Scott analyses. These
two oxygenates together accounted for about 0.3 ppm to 0.^ ppm of the 1.58 to 1.59
total shown above so their elimination would not remove the discrepancy with the
hourly average concentration obtained by continuous monitoring.
The Scott gas chromatographic analyses offered only one pair of duplicate
analyses. One of these samples was taken in a glass container and, like so many of the
other samples, showed a high (30 ppb) concentration of acetaldehyde. It also showed
anomalous concentrations of propane and isobutane and, to a lesser extent, n-butane.
Contamination with some kind of liquefied petroleum gas is suspected. The agreement
of this glass container sample with the matching metal sampler was not good.
In view of the inconsistency among surface samples it was not possible to discern
a trend or gradient in comparing aircraft samples with simultaneous ground samples.
Both showed the frequent presence of acetaldehyde and acetone.
From the data it is unclear which system is reporting the more nearly correct
concentrations. Even though the absolute magnitude of NMHC or individual species of
hydrocarbon is unknown, relative values can nevertheless be used to indicate the
location and the nature of hydrocarbon sources. This is demonstrated in the following
section.
^f.7.2 Analysis of Continuous Hydrocarbon Monitoring
The flux of hydrocarbon past the air monitoring station can be used so evaluate
source strength. Flux of NMHC can be estimated by multiplying wind speed by
concentration. A source strength which is constant should yield a constant flux in
spite of variations in wind speed (for comparable vertical mixing). Thus, multipli-
108
-------
cation by wind speed corrects for the greater dilution of hydrocarbon caused by higher
wind speed. Tables 16 through 20 present joint frequency distributions of wind
direction and NMHC flux for the five monitoring sites. Some wind directions are
associated with frequent occurrences of high flux values, which suggests that
important NMHC sources lie in those directions. For example, Site 1 showed high
fluxes from the north-northwest to east-northeast. Site 5 showed high flux values
associated with flows from the north-northeast to east-northeast and south-southwest
to west-northwest wind directions. Site 5 was above Washington's Crossing where
north-northeast to east-northeast flow of pollutants from north New 3ersey or New
York could emanate and south-southwest to west-southwest would come from
Philadelphia.
It is important to note that the wind directions of maximum NMHC flux differ
from the wind directions associated with high ozone readings (see Figures 13 through
17). For example, the quadrant associated with high ozone concentrations at Sites 1,
2, and 3 is southwest while the quadrants associated with maximum NMHC flux at
these same sites are northeast for Site 1, east for Site 2 (secondary maximum from the
south), and southeast for Site 3. These maximum NMHC wind directions put
Philadelphia and the industrialized corridor to the southwest along the Delaware River
upwind of Sites 1, 2, and 3. In contrast, the ozone roses seem to rule out Philadelphia
as a major area of origin. The data thus indicates that high NMHC concentrations at
monitoring sites around Philadelphia are due to short-range transport from the urban
core and are not related to long-range transport.
4.7.3 Hydrocarbon Species Analysis
Neither of the two analytical systems used in this program was especially
designed to detect organic compounds emitted by vegetation, even though isoprene and
the monoterpines (e.g., alpha and beta pinene) are within the C?-C[0 carbon range
specified. With very few exceptions the compounds which were detected are all
hydrocarbons typical of gasoline and/or auto exhaust; that is, they range from C? to
about C|Q and include all the major structural types. The only exceptions are
acetaldehyde and acetone which were present in many samples, often at high
concentration, even when the concentrations of other compounds were low or
109
-------
Table 16. NON-METHANE HYDROCARBON (NMHC) FLUX ROSE FOR SITE 1.
Wind
Direction
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
WSW
w
WNW
NW
NNW
NMHC Flux (ug/m2/sec)
0-300
20
31
23
22
17
2k
2k
25
W
52
W
2k
11
2
7
8
301-600
9
9
11
15
2
3
0
0
2
6
2
1
1
1
1
2
601-900
2
8
5
1
1
0
0
0
0
1
0
0
0
0
0
3
900-1200
2
0
3
2
0
0
0
0
0
0
0
0
0
0
0
0
1200-1500
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
>1500
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
110
-------
Table 17. NON-METHANE HYDROCARBON (NMHC) FLUX ROSE FOR SITE 2.
Wind
Direction
N
NNE
NE
ENE
£
ESE
SE
SSE
S
SSW
sw
wsw
w
WNW
NW
NNW
NMHC Flux (yg/m2/sec)
0-300
31
24
27
25
18
13
13
k
26
If
2*
34
32
30
2k
8
301-600
16
15
21
17
17
13
5
11
1*
17
9
23
20
9
22
2
601-900
7
6
4
9
5
8
3
7
10
3
5
8
5
1
8
3
900-1200
0
1
2
4
5
1
2
1
2
4
0
4
0
1
2
0
1200-1500
0
1
0
2
0
0
0
2
1
1
0
0
0
0
0
0
>1500
0
1
0
1
2
2
0
2
5
1
0
0
0
0
0
1
111
-------
Table 18. NON-METHANE HYDROCARBON (NMHC) FLUX ROSE FOR SITE 3.
Wind
Direction
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
WSW
W
WNW
NW
NNW
NMHC Flux (ug/m2/sec)
0-300
65
46
30
23
27
24
18
29
65
129
93
69
56
39
46
49
301-600
6
9
5
6
5
3
5
14
9
12
7
1
5
1
5
1
601-900
3
2
0
4
0
3
5
7
4
4
1
0
0
1
1
3
900-1200
2
1
0
0
0
0
8
2
3
4
2
0
0
0
1
2
1200-1500
0
0
0
1
1
1
5
2
1
0
2
0
1
0
0
1
>1500
0
0
0
0
0
0
3
0
0
1
0
0
0
1
0
0
112
-------
Table 19. NON-METHANE HYDROCARBON (NMHC) FLUX ROSE FOR SITE
Wind
Direction
N
NNE
NE
ENE
E
ESE
SE
SSE
S
ssw
sw
wsw
w
WNW
N\V
NNW
NMHC Flux (ug/m2/sec)
0-300
20
20
40
37
21
19
31
55
101
126
58
37
39
31
67
46
301-600
2
2
4
5
1
5
2
12
22
If
5
1
0
2
1
0
601-900
0
0
0
2
1
0
0
k
3
3
0
0
0
0
0
1
900-1200
0
1
0
0
0
0
0
1
1
14
0
0
0
0
0
0
1200-1500
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
>1500
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
113
-------
Table 20. NON-METHANE HYDROCARBON (NMHC) FLUX ROSE FOR SITE 5.
Wind
Direction
N
NNE
NE
ENE
E
ESE
SE
SSE
S
SSW
SW
WSW
w
WNW
NW
NNW
NMHC Flux (yg/m2/sec)
0-300
61
50
93
79
27
11
13
34
27
53
112
96
66
39
44
81
301-600
0
1
12
14
5
1
2
0
1
17
16
13
8
5
1
1
601-900
1
2
4
5
3
0
0
2
0
3
11
6
2
2
1
2
900-1200
0
1
3
3
0
0
0
1
0
4
5
4
4
2
0
0
1200-1500
0
1
1
0
0
0
0
0
0
1
5
3
0
0
0
1
>1500
0
1
0
4
0
0
1
0
0
0
0
2
3
0
1
0
114
-------
immeasurable. There are no clearly recognized major sources of these two compounds
although acetone has uses as a solvent, and it and acetaldehyde would be expected in
small amounts in exhaust gas or as a product of atmospheric reactions. There is a
likelihood that both are artifacts of the sampling system. A few samples showed
extreme values (several ppm) of ethene with no other anomalies. Most of these excess
ethene samples were from Site 1 but some were aircraft samples. Another anomaly
was found in two samples, excessive amounts of propane and isobutane. These two
samples had nothing apparent in common. Liquified petroleum gas would be the most
probable source of propane but it would usually be accompanied by n-butane.
One sample in particular, aside from the acetone and acetaldehyde found, looked
much like auto exhaust in composition, although quite dilute. It shows paraffins and
aromatics in about the expected propositions, as well as acetylene (10 ppb) and
propene (4.3 ppb) at about the ratio recognized for auto exhaust. However, only a
trace of ethene was found. One sample showed 0.7 ppb of 1.3 butadiene, a common
industrial emission. It is moderately ractive in the atmosphere but not more so than
some detectable olefins.
Three unsaturated hydrocarbons, acetylene, ethene and propene, provide both a
unique marker for auto exhaust and an index of degree of reaction. Only two samples
reported values for both acetylene and ethene and only three samples reported both
propene and acetylene. One sample showed a propene/acetylene ratio of
4.9/78 (ppb/ppb) and an ethene/acetylene ratio of 41/78. Another sample contained
more propene than acetylene but only trace amounts of ethene. In unreacted auto
exhaust the ratio of propene to acetylene should be (based on old data) about 0.20 to
0.25 (by moles). In one sample the ratio was 0.43 and again ethene was listed as trace.
With such a very limited data base, and one so inconsistent, conclusions on the nature
of the sources cannot be very firm.
115
-------
5. PERFORMANCE EVALUATION OF SELECTED MEASUREMENT TECHNIQUES
Ozone and precursors transported into an urban area affect the maximum ozone
concentrations observed downwind of the urban area during the afternoon. Thus, in
developing control strategies for attaining and/or maintaining the ozone standard,
state agencies should take into consideration the impact of transported ozone and
precursors. The first step in determining this impact is to determine the quantity of
ozone and precursors transported into the urban area.
Ozone and precursors could be transported over long distances overnight either
within or above the surface layer. However, near-surface transport of ozone and
nitrogen oxides over any appreciable period during the night appears to be limited by
surface sinks. The effects of the surface layer on hydrocarbon removal rates is, on the
other hand, less well understood. Limited data (U.S. EPA, 1977) indicate that
concentrations of hydrocarbons are significantly reduced in the urban plume as it
travels away from an urban area, but it is not known what mechanisms are
contributing to the reduction or to what extent. This data showed that hydrocarbon
concentrations upwind of urban areas are usually in the 0.1 - 0.3 ppmC range; the data
obtained at the rural surface stations in Philadelphia during the summer of 1978 also
fall within this range.
In contrast, because scavengers and sinks are lacking, high concentrations of
ozone and, possibly, precursors can remain in and above stable layers aloft and be
transported over long distances overnight. Vertical pollutant concentration profiles
taken in Philadelphia show that, in the early morning hours, high levels of ozone are
found above the surface-based inversion. As the inversion is eroded away, ozone is
mixed down to the surface. Surface measurements in rural areas near Philadelphia on
photochemically active days show an increase in ozone concentrations of about
.02 ppm immediately following the breakup of the surface based inversion. Little or
no change in hydrocarbons and oxides of nitrogen concentrations at the surface
monitoring stations is observed following the inversion breakup. Soundings also reveal
that the oxides of nitrogen concentrations in layers aloft are usually about the same as
those at the surface, typically in the range of 0.01 to 0.05 ppm.
116
-------
Calculations using EKMA (U.S. EPA, 1978) suggest that the observed levels of
transported hydrocarbons and oxides of nitrogen would not appreciably increase the
maximum ozone concentration downwind of an urban area. However, the observed
levels of ozone transported aloft and mixed to the surface could appreciably alter the
maximum ozone concentration.
The discussion above implies that the technique used for measuring the transport
of ozone must be able to quantify the level of ozone that is transported aloft
overnight, since this seems to be the primary parameter (in terms of pollutants) that
would cause a significant increase in the maximum ozone concentration in an urban
area. It also implies that if there is a need to quantify the precursors entering an
urban area, measurements of precursors made at a properly located upwind surface
monitoring station should suffice.
The rest of this chapter evaluates the capability of the five measurement
techniques tested in Philadelphia. The evaluation emphasizes the ability of the
technique to quantify the ozone transported aloft.
3.1 SURFACE MONITORING NETWORK
Since ozone trapped aloft overnight is mixed downward to the surface after the
breakup of the surface based inversion, it seems possible that surface measurements
could be used to quantify the ozone transported aloft. It has been suggested that
surface measurements between 1100 and 1300 LOT would be most indicative of ozone
levels aloft (U.S. EPA, 1977). The data collected during the summer of 1978 allowed
this hypothesis to be tested.
As pointed out in the preceding chapter, eight days during the 1978 summer
study were identified to be significant transport days. On five of those eight days,
soundings of pollutant concentrations were made with the U.S. EPA helicopter. Data
collected during the first flight of each day, prior to the onset of significant solar
radiation, should show the amount of ozone transported aloft overnight.
117
-------
In the preceding chapter, it was also pointed out that during those days, Sites 1,
2 and 3 were upwind sites. On four of those five days, Site 4 was also upwind of
Philadelphia. Early morning soundings, however, were taken mostly over Sites 1, 2,
and 3, with only one sounding taken over Site 4 on the transport days.
A comparison of the ozone concentrations aloft with average ozone readings
between 1100 - 1300 EDT at surface stations shows that the average surface ozone
readings are almost always higher than the ozone concentrations aloft. The average of
the differences between surface concentrations between 1100 - 1300 EDT and the
concentrations aloft is 0.014 ppm. Therefore, using the average surface reading
between 1100 - 1300 EDT would possibly overestimate the amount of ozone transport
aloft.
If we knew when ozone transported aloft is mixed to the surface we would have
an idea of when surface readings would be representative of ozone concentrations
transported aloft. During this study, therefore, an acoustic radar was installed at
Site 1. Records from the radar indicate when the surface inversion begins to be eroded
and when mixing in the vertical is no longer inhibited. Figure 42 is a record taken on
the morning of August 23, 1978. It shows stable atmospheric structure extending from
the surface to 220 m between 0700 - 1000 EDT. From then until 1100 EDT the surface
mixed layer deepens, after which mixing in the vertical is uninhibited.
The acoustic radar provided information on the time when vertical mixing
became uninhibited on four of the five selected days on which transport was
determined to be significant and helicopter data were available. On August 19 and 22,
mixing to the surface of pollutants trapped aloft was thus determined to begin at 1000
EDT, while on August 23 and 24 the mixing began at 1100 EDT.
Surface ozone readings at Sites 1 through 4 were compared against the peak
ozone reading aloft. Ozone data for three time periods were used for the comparison:
(1) the average for the hour following the mixing, (2) the average reading between
1100 - 1300 EDT, and (3) the average reading between 1000 - 1200 EDT. Table 21
presents the comparison. The limited data available for the comparison indicate that
the average reading for the hour following the mixing and the average reading between
118
-------
oo
IS.
CJs
tf>
Ml
§
L*
O
u
-------
&
Sg
E2
II
Is
tu
cxz «
S ai aj
OOcd
US<
~~- — (M TJ rM - — * CM n x oo
o o o o
dO O r^ O
o — < o —
' ^s ?
— — r*j oc
O O 0 0
1 o o o
O O 0
O 0 —
**^ J- 00 O
o o o o
-— . — fvi
-------
1000 - 1200 EOT are both more representative of the ozone level aloft than is the
average reading between 1100- 1300 EOT. The average difference between the
reading immediately following the mixing and the peak ozone aloft is .005 ppm. The
two largest differences occur on August 2k at Site k and on August 19 at Site 1. Using
the 1100 - 1300 EOT averages for these two instances would result in an even larger
difference, hence would be a poorer indicator of ozone levels aloft.
This analysis indicates that surface monitoring stations located upwind of urban
areas can be used to measure ozone transported aloft. Hourly averaged readings taken
during the hour immediately following the completion of the mixing between the
surface layer and the layers aloft are representative of the ozone levels aloft. The
time when mixing is completed is best determined by an acoustic radar, which gives a
continuous record of the atmospheric structure. If the radar is not available, the
average over the two hours during which the mixing is completed (1000 - 1200 EOT in
Philadelphia) would also be a good indicator of ozone levels aloft.
Given that upwind surface monitoring stations can be used to quantify the ozone
transported aloft, the next question is how many upwind stations are necessary. In
other words, is there a significant horizontal gradient upwind in the ozone concentra-
tions aloft?
On the transport days when helicopter soundings are available, ozone concentra-
tions aloft over all the upwind sites are almost identical (within .01 ppm of one
another). Table 21 shows that the only major gradient in ozone concentrations was
observed on August 19, when the average ozone aloft above Site 3 was .071 ppm while
the ozone concentration above Site 1 was .047 ppm and above Site 2 was 0.048 ppm.
Another day when there is a definite difference is August 22, when helicopter data
shows that the concentration aloft increased from .052 ppm over Site 3, to .062 pprn
over Site 2, to .070 ppm over Site 1. At the surface stations, however, Site 2 had a
reading of .053 and Site 3 a reading of .058. No surface data were recorded at Site 1
at that time. The helicopter soundings indicate that, in general, the ozone-laden
plume aloft traveling into Philadelphia is fairly homogeneous, and there being little
variation on most days in ozone concentrations aloft along the upwind boundary of the
urban area.
121
-------
Table 21 also shows that data at a surface station used to represent ozone aloft
compares well with that at another surface station. The largest difference was on
August 19 when Site 1 was higher than Site 2 by .024 ppm.
Therefore, the data indicate that, for Philadelphia, one monitoring station
upwind of the urban core is sufficient to acquire measurements closely indicative of the
ozone being transported aloft. However, one upwind site might not suffice for
measuring ozone transported into other East Coast and Midwest cities. Consideration
should be given to the probability of significant and varying ozone concentrations from
other nearby, upwind urban sources. But, for most isolated Central and Western cities,
one monitoring station would probably be satisfactory.
5.2 AIRBORNE MEASUREMENTS BY DEDICATED INSTRUMENTED AIRCRAFT
The capability of a dedicated instrumented aircraft to measure the ozone trans-
ported aloft is obvious if the equipment is properly calibrated and the operators are
well trained. The soundings taken by the U.S. EPA helicopter provided valuable
information on the vertical ozone profiles. From surface measurements, one can only
infer the ozone levels aloft. Using a dedicated instrumented aircraft, one can directly
measure the ozone concentration aloft. As discussed earlier, it is probably not
necessary to measure precursor levels aloft. Precursor levels can best be measured at
the surface. Thus, the combination of a dedicated instrumented aircraft measuring
ozone and lapse rate and an upwind surface monitoring station measuring ozone,
precursors, and wind would provide all the needed information on the transport of
ozone and precursors into an urban area.
5.3 AIRBORNE MEASUREMENTS BY A PORTABLE INSTRUMENT PACKAGE
The portable instrument package is another direct means of measuring the
concentration of ozone aloft. The reliability of the data obtained was tested by
comparing the data with that collected by the U.S. EPA helicopter. Eight side-by-side
flights were made with the portable instrument package and the helicopter in
Philadelphia. Figure 43 presents a comparison of soundings made with both methods.
The helicopter data is faithfully duplicated by the profile obtained with the portable
122
-------
1
o
o
8
oo
v£>
•Sf
O
O
O
0)
O
M
a; c
£. i-
+3 «;
II
CL 3
UJ
ai
D
O
*•<
a.
123
-------
instrument package. However, the readings from the portable instrument package are
about twice those of the helicopter. The plots shown are representative of the other
seven comparisons. The higher readings from the portable instrument package are
apparently caused by a zero shift of the ozone instrument of that package. There is no
consistent shift in the zero for the eight soundings. For the case presented, the
portable instrument package reads .08 ppm higher than the instrument on the
helicopter. For others, the difference may be .05 ppm. All the readings obtained with
the portable instrument package are higher than those from the helicopter. Thus,
there appears to be a variable, but positive zero shift throughout the side-by-side
tests.
Table 22 shows results of a statistical comparison between soundings for each of
the eight flights. Ozone concentrations at one-hundred meter intervals were used for
the comparison. There is a very good correlation between the portable instrument
package and helicopter data, except for the first two flights where the range of ozone
concentrations was smaller (about .020 ppm for the first flight and 0.015 ppm for the
second flight) and the clustering of all data points around a narrow range of numbers
leads to a poor correlation. Analysis indicates that the portable instrument package
can duplicate the helicopter if the zero shift problem can be resolved.
This zero shift problem was not discovered until after all side-by-side flights
were completed and more thorough testing of the portable instrument package was
done upon return to Pasadena. Those tests show that turning on the zero mode switch
on the instrument does not allow adjustment of the zero; it merely turns the
photornultiplier tube off. To adjust the zero, zero air has to be introduced into the
analyzer by passing ambient air through activated charcoal placed in front of the
sample inlet of the instrument. This ozone removal procedure was followed in flights
made in September when the free lift balloon ozonesonde and tethered balloon
ozonesonde systems were tested. Good agreement was achieved, as will be discussed
later, between the portable instrument package and the two other techniques when the
zero was properly adjusted.
The signal from the ozone instrument used in this study was fairly noisy, a
consequence of the high amplifier gain settings required when Ethychem was used
-------
z
UJ
s
D
a;
i-
UJ
_l
CD
<
f-
a:
§
UJ
X
CQ
Q .
UJ a5
Q UJ
uJ
UJ Qi
ZH
Q
Z Z
o <
U a,
(N
0)
IS
(TJ
1'.
-i
e<4
s
^
! s'
is
s
a.
8
OJ
&
(4
J£
^
CL
L-
4J
Q.
OJ
1
S
VM
8
U
M
C
•o
s
3
no
C
I
0*00— loo^^c^ty^
_j-^fSJ-rt— '--—"—•
isssssli
S^. — i^ a- s£ S.B oo
o o o o o o o
^IsissiS
Ko — rJots^K
1 8 E § g i § 2
p!l trs 00 J) w-x 0
oooooooo
S Sft 8 2 S S *
1 1 ( 1 1 1 1 1
— «OO-*O— •— «O
oO— '-S-J-CiO^-^-OO
"^"^^r^oooo^ON
1
u
<
§
_m!_m(N^_
^^&^,0^^^
— ' — — • CM fN r*4 rsi rs
^•^.— •w"^.-**--*. ^^
125
-------
instead of pure ethylene in the reaction cell. Thus, the profile obtained is not as
smooth as that from the helicopter or the ozonesonde. If bottled ethylene were to be
used, the signal-to-noise ratio would improve three-fold. However, the use of ethylene
gas in an aircraft would require recertification of the plane in the "restricted"
category, which defeats one of the desirable features of a portable instrument
package. As noted earlier, the instrument used here does not satisfy EPA reference
method designation specifications when operated with Ethychem gas.
Based on the data gathered to date, it is uncertain whether the portable
instrument package can be used to measure ozone aloft, as simply as was originally
hoped. Further testing of the portable instrument package is recommended.
5.4 SOUNDINGS BY OZONESONDE BENEATH A FREE LIFT BALLOON
As explained in Section 3.6, the modified ozonesonde removes SO- before it
analyzes air samples. Also, NO interference is minimal. Thus, the ozonesonde
measures ozone concentrations rather than total oxidants.
The modified ozonesonde also measures temperature. If a theodolite is used to
track the balloon, upper air wind speed and direction can also be determined. An
example of data from a sounding is shown in Figure 44. An advantage of this system
over others is that it can attain higher altitudes. During the tests, the ozonesonde was
easily tracked beyond 5,000 m.
Because of logistical problems, this system was not sent to the field prior to the
helicopter's departure from Philadelphia. Thus, side-by-side comparison with the heli-
copter was not possible. However, this system was tested simultaneously with the
portable instrument package and the tethered balloon ozonesonde in September.
A total of eleven soundings were made. Four of them were discarded because
review of the pre-flight calibration results afterwards indicated that the response of
these ozonesondes was non-linear. Two of the successful soundings were made within
an hour of soundings made with the tethersonde; the portable instrument package was
also flown within an hour of one of those soundings. On two other occasions, the
126
-------
T T
NOLLD3HKI QNIA
«o ~
0(7;
- D
UJ
2 a
G
z
u
00
f*.
ON
05
UJ
tfl
2
01
a,
Ot/)
S §
:i
S O
o
—r—i r—t r—»• O
I §
aaniinv
127
-------
portable instrument package was flown within an hour of the release of the
ozonesonde.
Table 23 presents ranges, means and standard deviations of the ozone concentra-
tion obtained by various methods. Data points were taken at 100 m intervals. The
range of the ozone concentrations observed during each sounding is limited, usually
remaining within .02 ppm of the ground-level value. Since the accuracy of the
instruments tested is about 4^.01 ppm, the data collected do not lend themselves well
to correlations.
Figure 45 shows that the ozone profiles track each other well. If the soundings
were taken within thirty minutes of each other, the absolute differences of their
means were equal to or less than .010 ppm, as shown by the second and third
comparisons presented in Table 23. For the first comparison, the soundings were also
made within thirty minutes. However, the portable instrument package readings are
consistently lower, by about .024 ppm, than those from the ozonesonde. There is no
obvious reason for the difference. The two comparisons between the free lift balloon
ozonesonde and tethered balloon ozonesonde were made near noon and in the late
afternoon, with the tethered balloon ozonesonde soundings about an hour later than the
free lift balloon soundings. Around noon, ozone values should increase with time.
Consistent with this, the tethered ozonesonde data show that, on the average, readings
are higher than those from the earlier free lift balloon sounding. Around 1500 EDT
and thereafter, the ozone concentration begins to decrease. The tethered ozonesonde
sounding, which was made later than the free lift balloon ozonesonde sounding, shows
average concentrations about .020 ppm lower, which is again consistent with the
assumed behavior.
Ozone concentrations observed by the ozonesonde before release are always
within +;.01 ppm of the surface station reading, indicating that ozonesonde readings
are accurate to within the specifications of the instrument.
The free lift balloon ozonesonde is thus a good candidate for measuring ozone
aloft. One major advantage of this system is its ability to measure concentrations to
128
-------
UJ
UJ
u-
<
UJ
UJ
to
UJ
Q
Z
UJ
Z
o
N
O
CO •
UJ
QO
uj <
Q y.
0.5
<
J-
H t/)
ON
""*
ITS ^o rsi
(N — ' — •
O O O
fN — i Q
o o S
— rv oo
00 vO d-
o o o
000
o o o
1 ' '
sss
1*1 fx 00
o o o
1 t 1
(^ a- Ji-
o o o
o o o
-* — • o
1^ rv 0
ON OS 0>
•o a.
SI « 3
S
8*&
_
OO O
(M (N
o o
OO O
fM (N
O O
^•o
o c
o w
TJ
fc-8
fl
1
c
o
-Q
1) "^
l-
Jf ^
s§
§8
O 00
ss
s§
c*\ 00
O O
1 t
00 —
°.°
0 0
o o
CM "^
fx r.
OS (^
flj
"O
-D &
S "8
5s3
129
-------
01
O
00
"
u
<
a,
H
Z
ol
5
D
Oi
te
01
rx fc O
o wo a,
00
al
Q
O
I
O
o
O
H
o to LL,
O
O
o
o
CM
O
O
°. E
J:
J
02
H
Z
al
U
•^ z
-? o
u
o)
z
o N
-5 O
•u
i
(D
O
N
O
l
rO 4)
~ s.
is
L- (U
a. So
o 5
N a.
o ^
o
o
o
'^ t:
5
U rt
01
Qi
D
O
aannnv
130
-------
higher altitudes than all other systems considered. The disadvantage is that once it is
released the system cannot be retrieved and reused.
It was learned that the calibration data should be carefully examined before
launching the ozonesonde. If the instrument is determined to be non-linear it should
be discarded and not launched.
5.5 SOUNDINGS BY QZONESQNDE BENEATH A TETHERED BALLOON
This system is similar to the ozonesonde beneath a free lift balloon. Outputs
from a tethered ozonesonde sounding are plotted in Figure 46. A comparison between
this and the free lift balloon system has already been presented in the preceding
section, showing good agreement between the two systems.
On August 24, a tethered ozonesonde sounding was made immediately following a
sounding by the U.S. EPA helicopter. The tethered sounding was made up to 350 m
AGL, while the helicopter vertical spiral ranged from 1,800 m AGL to about 100 m
AGL. Thus the overlap was only about 250 m. Comparison of the data from the two
systems for the 250 m indicate that measurements from the tethered ozonesonde are
higher by about .045 ppm, although the trends are identical. The reason for this
positive bias of the ozonesonde is unknown.
Four comparative soundings with the tethered ozonesonde and the portable
instrument package were made in September. Table 24 presents the comparison
results. Data used from each sounding were taken at 100 m intervals. The first two
comparisons indicate that the zero shift problem of the portable instrument package
was apparently not corrected at the time of the test. The later comparisons indicate
that the two systems reported almost identical information.
The data'show that the tethered ozonesonde accurately measures ozone concen-
trations aloft when the instrument is carefully calibrated. An obvious advantage of
the tethered ozonesonde is that it is reusable. The major drawback (in its present
configuration) could be its limited vertical range, at least with the size of balloon used
here. The balloon and launch system are designed for ascent to 1,000 m above ground
131
-------
V L VW V\
NOLLD3HK1 QNIA
\\ vv vv
Q
UJ
UJH
l
V
X
-.&>
' i
— .. 00
2S Sg
Q
UJ
a,
V)
•Si
"O <4-l
-------
<
X
OJ
Z
uj
cD
LLl
Q
O
(X)
UJ
Z .
O tu
NO
O<
UJ
a,
<°
aF
°§
o-O
O -J
7-
o
cQ
o
UH
(N
ffl
l_ 1/5
fll IM .
fs^S
E a, o «J
Z "o
* *-*
£ 33 c
o>
OJ U
t; c <"
~~* n^ "^ (tj
O a; O O
u/ '
H
O
00
3.
.3- rO ir»
— c O O
. . .
Ox O -3-
OX CO OO
— < -H O
1 1 1
-d- OX ox
Ox o u"\
— i — < O
000
O O O
VD fv. O
.3- — i — i
CvJ (N (N
OO Ox OX
O
00
O
o
^
o
Ox
O
*
00
o
1
\D
XO
o
*
-------
level. As indicated by data from the other airborne techniques, the ozone structure
above 1,000 m is significantly different from that in the lower 1,000 m. The tethered
ozonesonde's inability to reach above this level would limit its use for measuring ozone
concentrations throughout the mixing layer. A larger balloon would provide a greater
altitude capability, at the expense of ease in ground handling.
-------
6. COST EVALUATION OF SELECTED MEASUREMENT TECHNIQUES
As pointed out in Chapter 2, a fixed station costs about $40,000 to $60,000 to
install and furnish with monitoring systems, and about $2,000 to $5,000 per month to
operate manually. The cost of outfitting and operating an instrumented aircraft is, of
course, very high. An initial investment of about $70,000 is necessary. Operating cost
would be about $100 to $250 per flight hour, excluding technical labor hours required
to operate the instruments onboard and reduce the data.
Thus, the costs and effectiveness of operating a portable instrument package
aboard an aircraft and the ozonesonde beneath free lift and tethered balloons are of
considerable interest.
Experience in utilizing the free lift ozonesonde, tethered ozonesonde, and
portable instrument package during the summer of 1978 indicates that their costs
would be as presented in Table 25. The initial cost includes the purchase of
equipment. The operating cost includes consumable equipment and supplies for each
flight. Labor hours include set-up, operation, and data reduction for each flight. The
final product common to all three systems is a plot of ozone and other parameters
versus altitude above ground level.
The initial cost for the free lift ozonesonde electronics package is approximately
$5,000, about $4,000 more for a theodolite if wind speed and direction data were
desired. Acquiring wind speed and wind direction data would also add about five labor
hours to each flight. Initial costs for the tethered ozonesonde are twice as high as for
the free ozonesonde due to the price of the tethered balloon and the more
sophisticated sensor/transmitter which includes wind speed and wind direction.
Operating costs besides labor include only the cost of helium. Labor hours are
somewhat reduced since processing of the wind speed and wind direction data is
automated. In addition, the labor hours per flight can be reduced if more than one
flight per day is done, since set-up time for the tethersonde amounts to about two to
three hours. Set-up is included in the estimate shown in Table 25.
135
-------
Oi
2
UJ
Z
O
N
O
a
z
5
1/5
<
UJ
S
tt!
O
Q
O
H
LU
s
tt-
o
H
m
O
U
(0
1
4-*
•21
£ o
i- m
Q fc
Ui)
fc
jC
o
u
x
to
-j <7>
nj - —
'2 O
H»
~U
•o
-
u (iT
(/l 3
tn in
i— 4)
2 3 U
i- «
O Q^
CL,
•
1
a.
00
>
MH
O
•»->
IT)
O
XI
rtj
—I
X)
0)
I
(0
136
-------
The initial cost for the portable instrument package is $7,500 which includes the
ozone analyzer, temperature sensor and recorder. Operating costs per flight (one
hour) are about $50 for aircraft, pilot, and required analyzer gas. Labor time is about
nine man-hours. A suitable NO/NO analyzer, just recently available, and recorder
would increase the initial cost by about $7,000. Labor hours would increase by about
two to three hours per flight if NO/NO analysis was performed.
A
Thus, the initial investment is the greatest for the tethered ozonesonde and for
the portable instrument package with an NO/NO analyzer. Operating costs are
A
greatest for the free lift ozonesonde, mainly because the instrument package which is
lost on each flight.
Only one person is needed to operate any of the three systems and the level of
technical expertise required for operation is low for all systems. The operator must
have knowledge of instrument calibration, however, and all systems require access to a
calibrated ozone source prior to flight.
137
-------
7. RECOMMENDATIONS ON MEASUREMENT TECHNIQUES
Of the five measurement techniques evaluated, surface measurements at fixed
sites, airborne measurements by dedicated instrumented aircraft, and soundings by
ozonesonde beneath a free lift balloon have proven suitable for measuring directly or
providing information on ozone concentrations aloft. Selection of the technique by an
air pollution control agency would depend on the funding available, the intended use of
the data, and the expertise of the agency. An expansion on this point will be presented
later.
It is uncertain, based on the data collected thus far, whether the portable
instrument package would serve the purpose of measuring the transport of ozone aloft.
Since the package is easy to maintain and has lower operating costs than any of the
three methods discussed earlier, it would be worthwhile to perform more tests on the
system to be sure of its applicability.
The tethered balloon ozonesonde system tested here could not obtain information
beyond 1000 m above ground level. Since significant ozone structures often exist
above that elevation, its applicability is somewhat limited. A larger balloon would, no
doubt, be able to carry the system to the desired elevation. However, the use of a
larger balloon would require technicians experienced in large balloon launches and also
an additional investment of about $5,000. Thus, the basic attractions of the
system - ease of operation and low cost - would be diminished.
As mentioned earlier, fixed stations can be used to determine ozone aloft if it
can be clearly known when mixing is complete between the surface and layers aloft.
One way of obtaining that information is with an acoustic radar, which gives a
continuous record of the changes of the atmospheric structure. The hourly average
concentration immediately following the mixing is found to be most representative of
the concentration aloft. The use of an acoustic radar, along with a surface monitoring
station upwind of an urban area, would be a practical approach for quantifying
transport into an urban area. This combination would provide continuous information
138
-------
for purposes of routine surveillance and input to models. This approach is within the
technical capability of all air pollution control agencies currently maintaining air
quality monitoring stations. Initial cost of such a station (with an acoustic radar)
would be about $40,000 to $70,000 and the operating cost would be about $2,000 to
$5,000 per month.
A dedicated instrumented aircraft is the best airborne measurement platform.
However, the approach is probably beyond the financial and technical means of many
local and state air pollution control agencies. Also, it is not practical for routine
.surveillance. It is, however, the best approach to quantify ozone levels aloft on days
that are forecast to have significant transport and/or high ground-level ozone
concentrations. Thus, this is a recommended approach for obtaining inputs to models
and special research studies for agencies that can either afford to instrument an
aircraft or hire consultants that can provide instrumented aircraft.
The free lift balloon ozonesonde system is one that most air pollution control
agencies would no doubt be able to afford and have the technical capability to operate.
Similar to the dedicated instrumented aircraft, this approach would not be practical
for routine surveillance. It would be an appropriate technique to measure ozone
concentrations aloft for input into models and for studying vertical ozone distributions
during ozone episodes.
139
-------
8. REFERENCES
Adams, D.F., and R.K. Koppe (1969): Instrumenting light aircraft for air pollution
research. J. Air Poll. Control Assoc., 19, pp. 410-415.
American Public Health Association (1972): Methods of Air Sampling and Analysis.
Washington, D.C. Work performed under U.S. Environmental Protection Agency
Contract No. 68-02-0004.
Anderson, 3.A., D.L. Blumenthal, and G.3. Sem (1977): Characterization of Denver's
urban plume using an instrumented aircraft. Denver Air Pollution Study - 1973,
Proceedings of a Symposium, Volume II. U.S. EPA, Environmental Sciences
Research Laboratory, Office of Research and Development, Research Triangle
Park, North Carolina (EPA-600/9-77-001), pp. 3-33, February.
Baboolal, L. B., M. I. Smith, D. W. Allard, L. G. Wayne, and 3. W. Mortz (1975): A
climatological and air quality characterization and air quality impact assessment
for various future growth alternatives in the Santa Ynez Valley. (AV FR 509,
AeroVironment Inc., Pasadena, California).
Blumenthal, D.L., T.B. Smith, W.H. White, S.L. Marsh, D.S. Ensor, R.B. Husar, P.S.
McMurry, S.L. Heisler, and P. Owens (1974): Three-dimensional pollutant
gradient study — 1972-1973 program. Technical report submitted to California
Air Resources Board, Sacramento, California. (MRI 74-FR-1262, Meteorology
Research, Inc., Altadena, CA), November.
Chadwick, R.B., K.P. Moran, R.G. Strauch, G.E. Morrison, and W.C. Campbell (1976):
Microwave radar wind measurements in the clear air. Radio Science, 11, 10, pp.
795-802, October.
Cleveland, W. S., B. Kleiner, 3. E. McRae, and 3. L. Warner (1976): Photochemical air
pollution: transport from the New York City area into Connecticut and
Massachusetts. Science, 191., pp. 179-181.
DeHay, W. (1974): Sutro tower atmospheric boundary layer experiments, user's
manual. San 3ose State University, Meteorology Department, San 3ose,
California, pp. 64.
Edinger, 3.G. (1975): Acoustic sounding of the lowest one and one-half kilometers over
Los Angeles. Paper presented at 16th Radar Meteorology Conference, Houston,
Texas. American Meteorological Society, pp. 253-256, April.
Edinger, 3. G., M. H. McCutchan, P. R. Miller, B. C. Ryan, M. 3. Schroeder, and 3. V.
Behar (1972): Penetration and duration of oxidant air pollution in the South
Coast Air Basin of California. 3. Air Pollu. Con. Assoc., 22, pp. 882-886.
Environmental Science and Technology (1976): K), p. 730 (news note on DaVinci II
flight).
140
-------
Forrest, J., S.E. Schwartz, and L. Newman (1979): Conversion of sulfur dioxide to
sulfate during the Da Vinci flights. Atmos. Environ., 13, J_, pp. 157-167.
Grimes, B. A., and D. A. Pasquini (1978): Second audit of northeast oxidant transport
study. RTI/1487/73-01F. Prepared for U.S. Environmental Protection Agency,
EPA Contract No. 68-02-2725, August.
Hancock, C. W. (1978): Philadelphia oxidant study, 1978, helicopter platform data
report. ESG-TR-78-15, Northrop Services, Inc., Las Vegas, Nevada. Submitted
to U.S. Environmental Protection Agency, Contract No. 68-03-2591.
Heinsheimer, T.F. (1977): ATMOSAT, a new measurment platform for air quality
monitoring. 3. Air Poll. Cont. Asso., 27, I, pp. 530-533.
Leone, I.A., E. Brennan and R.H. Daines (1968): The relationship of wind parameters
in determining oxidant concentrations in two New Jersey communities.
Atmospheric Environment 2, pp. 25-33.
Lyons, W. A., and H. S. Cole (1976): Photochemical oxidant transport: mesoscale lake
breeze and synoptic-scale aspects. J. App. Met., 15, pp. 773-743.
MacKay, K.P. (1977): Ozone over San Francisco - means and patterns during pollution
episodes. San Jose University, Department of Meteorology, Report No. 77-01,
August.
Millan, M.M. and R.M. Hoff (1978): Remote sensing of air pollutants by correlation
spectroscopy - instrumental response characteristics. Atmos. Environ., 12, 4,
pp. 853-864.
Miller, D. F., W. E. Wilson, and R. G. Kling (1971): A versatile electrochemical
monitor for air quality measurements. 3. Air Pollu. Cont. Assoc., 21, pg. 414.
Pasquini, D. A., and B. A. Grimes (1978): Audit of the northeast oxidant transport
study. RTI/1487/78-01F. Prepared for U.S. Environmental Protection Agency.
EPA Contract No. 68-02-2725, November.
Pasquini, D. A., and L. T. Hackworth (1978): Audit of the northeast oxidant transport
study. RTI/1487/69-0IF. Prepared for U.S. Environmental Protection Agency.
EPA Contract No. 68-02-2725, August.
Petterson, S. (1956): Weather Analysis and Forecasting. Volume I; Motion and Motion
Systems. McGraw-Hill Book Company, Inc.
Peters, G., C. Wamser, and H. Hinzpeter (1978): Acoustic Doppler and angle of arrival
wind detection and comparisons with direct measurements at a 300 m mast. J.
App. Met., \7_, 8, pp. 1171-1178, August.
Post, M.J., R.L. Schwiesow, R.E. Cupp, D.A. Haugen, and J.T. Newman (1978); A
comparison of anemometer- and lidar-sensed wind velocity data. 3. App. Met,,
J7, 8, pp. 1179-1181, August.
141
-------
Russell, P.B., E.E. Uthe, and F.L. Ludwig (1974): A comparison of atmospheric
structure as observed with monostatic acoustic sounder and lidar techniques. 3.
Geo. Res., 79, 36, pp. 5555-5566, December.
Russell, P.B., and E.E. Uthe (1976): The Mt. Sutro tower: a unique facility for marine
boundary layer studies and remote sensor development. Paper presented at Navy
Workshop on Remote Sensing of the Marine Boundary Layer, Vail, Colorado, 9-11
August. Stanford Research Institute, Atmospheric Sciences Laboratory, Menlo
Park, CA.
Russell, P.B., and E.E. Uthe (1978): Acoustic and direct measurements of atmospheric
mixing at three sites during an air pollution incident. Atmos. Environ., 12, 5, pp.
1061-1074.
Saltzman, B. E., and A. F. Wartburg (1965): Absorption tube for removal of interfering
sulfur dioxide in analysis of atmospheric oxidant. Anal. Chem., 37, pp. 779-782.
Saucier, W. 3. (1955): Principles of Meteorological Analysis. The University of
Chicago Press.
Scott Research Labs (1970): Atmospheric reaction studies in the New York City area,
Volume 1. Program design and methodology, data summary and discussion.
Prepared for the Coordinating Research Council, Inc. and Air Pollution Control
Office.
Stephens, E. R. (1977): Chemistry and meteorology in an air pollution episode. 3. Air
Pollu. Cont. Assoc., 25, pp. 521-524.
U.S. Environmental Protection Agency (1973): Investigation of high ozone concentra-
tion in the vicinity of Garrett County, Maryland and Preston County, West
Virginia. Publication No. R4-73-019. Office of Research and Monitoring.
U.S. Environmental Protection Agency (1977): Uses, limitations and technical basis of
procedures for quantifying relationships between photochemical oxidants and
precursors. EPA 450/2-77-021a.
U.S. Environmental Protection Agency (1978): Procedures for quantifying relation-
ships between photochemical oxidants and precursors: supporting documenta-
tion. EPA-450/2-77-021b. Monitoring and Data Analysis Division, Office of Air
Quality Planning and Standards, February.
Wanta, R. C., W. B. Moreland, and H. E. Heggestad (1961): Tropospheric ozone: an
air pollution problem arising in the Washington, D.C. metropolitan area. Monthly
Weather Review, 89, pp. 289-296.
White, W. H. 3. A. Anderson, D. L. Blumenthal, R. B. Husar, N. V. Gillani, J. D. Husar,
and W. E. Wilson, Jr. (1976): Formation and transport of secondary air
pollutants: ozone and aerosols in the St. Louis urban plume. Science, 194, pp.
187-189.
142
-------
Wolff, G. T. P. J. Lioy, G. D. Wight, R. E. Meyers, and R. T. Cederwall (1977):
Transport of ozone associated with air mass. Paper presented at the 70th annual
meeting of the Air Pollution Control Association in Toronto, Ontario, Canada,
20-2& 3une.
Yap, D., and Y. S. Chung (1977): Relationship of ozone to meteorological conditions in
southern Ontario. Paper presented at the 70th annual meeting of the Air
Pollution Control Association in Toronto, Ontario, Canada, 20-2* June.
-------
APPENDIX A
The results of the first (July 18-26, 1978), second (July 31-August 4, 1978),
and third (September 5-9, 1978) audits of the Northeast Oxidant Transport Study are
contained in Tables A-l, A-2, and A-3. Table A-l contains the nitrogen oxides audit
data, Table A-2 contains the ozone audit data, and Table A-3 contains the meteoro-
logical audit data for the network as a whole. The AeroVironment hydrocarbon
analyzers were only audited during the third period. Audit results are given in Table
-------
>
a;
g
T3
3
<
(O
*
•s
o
c
4>
ff
is
>>
"S
4->
wn
a
s
nt
i_
H
•a
o
**
$
0)
5
g
J
rd
£
I
g
S
central
c
o
U
1
<
§
s
1
U
ct
K
X
?
<
'o
8
w)
V
u
M
V
ft
«;
N
X
•a
c
<
1
o
1
z
i
i!
51
5
ll
1
1
l
so fv.
S'S
Jld
1
1
i
<
z
3 »>
S 0
— 0
Franklm Institute
City of Philadelphia
I-
G i O
J'd
1 '
1
1,'i
-lo
1
si«
o 1 o
-1°
1
ol
rv Ch
Si8
J|o
i
i
i;§
ojo
I
1
<*l
m M
?l§
O t O
1 '
1
S[2
O 1 O
-"I?
1
1
ol
r-» i r*j
§1 o
• °.
-.0
1
3l £
°i°.
-!•?
i
Hockessm, Delaware
AeroVironment
(N 1(N
S'8
-lo
I
1
ssL
u*i |O
o jo
-,o
1
sL
§l§
-i?
i
i
!
sL
gi§
-lo
1
1
f\ 1
(M > C«>
*^\ .0
o |o
"1°
1
Nl
1^ 1 o
K 8
o' 'd
1
|
ol
I'l
-|o
1
1
•> 1
(N 9*
N. 1 O
o 1 o
-1°
1
}
si*
o 'o
-Id
1 '
1
Downingtown, Pennsylvania
AeroVironment
o
00 OS
S|§
O jO
I
00 |
-* 'oo
'o
Tl*^
-'o
1 '
1
<
Z
EMSL Helicopter
Wilmington, Delaware
v*t
t-
1
1
5
1
§
-------
c
m
T)
'x
O
(0
c;
.c
-f-k
i_
o
Z
3
a
c
0
+J
fl
•»-•
V
(H)
C,
O
,~4
•3
0
flj
CO
C
a.
c.
Q^
CXi
O
O
O
d
o
VO
oo
d
Z
^
Z
ON
O
d
cO
ON
•A
^
1
1
1
Lancaster, Pennsylvania
State of Pennsylvania
1
1
CM
0
O
O O o
21 Zi
— H
o
co
CO0"*
ZZd
ZZZ
^ ^ ^
ZZZ
•Cj" f^ -^
o — « o
o o o
d o d
i
J*- 00 ON
fXk ~-< ^H
00 VO Jd-
ON ON vo
odd
1
Somerville, New Jersey
Ancora, New Jersey
Bivalve, New Jersey
State of New Jersey
ON OO
0 0
o o
d d
ro f\
r^* ON
(N —i
ON O
d -*
Z Z
^ ^
Z Z
O f"\
o o
o o
d d
ff^ f\J
•*J* -^t*
—I OO
ON ON
d d
Broad and Spruce
Franklin Institute
City of Philadelphia
ON — ( (S — i O ON
O O O O — O
O O O O O O
O O O O O O
1
oo r>^ .3- if\ o >A
CM o i^ vo oo r-x
PO ^D CO VO CO OO
ON O — « 00 — i OO
0 — i — « 0 -< 0
jj- rvl .3- -H
O O O O
0000
ddddg<
•* 1^ VO OO
CO *A CM ON
oo oo oo i-^
oo hx vo r^ Q <£
d d d d Z Z
CM -3- — i J3- — i
-H O O CM i/N.
O O O O O
d d d d d 1$
i i Z
— < o o t^> ^
>A ON
-------
O
Z
£
+j
VH
O
x>
fd
I
Jl)
2
**
3
•o
S
c
»•*
v)
Station Location
1
X
a!
a
a
o
fcO
X
B;
3:
a
o
t/i
X
1
a
H
a
*
**
g
Affiliated Organizat
i
2
i
8
z
^^
2;
Z
Z
o
0
o
o
n
Lancaster, Pennsylvani
State of Pennsylvania
ZZZ
c£ t£ c£
zzz
oioi
ZZO
go:
zzo
zzz
zzz
zzz
zzz
zzz
zzz
_. _ o
zzz
— _ O
zzz
1
1
Somerville, New Jersey
Ancora, New Jersey
Bivalve, New Jersey
State of New Jersey
1
ZZ
tK
ZO
ZZ
ZZ
ZZ
«
zz
zz
z z
zz
zo
o-
z z
O-
z z
1
1
Broad and Spruce
Franklin Institute
City of Philadelphia
1
ZZZZZ
OS o£ O Qi
zzzoz
oozoo
ooiSo
**|||
Qi Oi O!
ZZOOZ
O O
zoozo
oiooo
zzzzz
zzozz
zoo zz
z oozz
1
M 1
c 2 '
* e >. 1
Hockessin, Delaware
Downtngtown, Pennsylv
Collegeville, Pennsylva
Woodstown, New Jerse
Ringoes, New Jersey
AeroVironment
1
i i i i i i t i X
tt
< O -.&** o
-------
Table A-4. REGRESSION ANALYSIS OF BECKMAN 6800 GAS CHROMATOGRAPH
RESPONSE ON AUDIT CONCENTRATION.
Station Location
Affiliated
Organization
Hockessin, DE
Downington, PA
Collegeville, PA
Woodstown, NJ
Ringoes, N3
AeroVironment
THC Channel
m
0.9672
0.9066
0.8403
0.8900
0.9312
b
0.021
-0.219
0.155
-0.094
0.331
r
0.9999
0.9943
0.9869
0.9995
0.9982
Methane Channel
m
0.9581
0.7442
0.9987
0.8693
0.5193
b
-0.063
-0.028
-0.210
0.072
0.342
r
0.9999
0.9991
0.9992
0.9999
0.9984
m - Regression slope
b - Regression intercept
r - Regression correlation coefficient
148
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/4-79-039
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE _ . . . ,, , .
Ozone and Precursor Transport into an Urban Area -
Evaluation of Approaches
5. REPORT DATE
December 1979
6. PERFORMING ORGANIZATION CODE
7'^Vctiae'|S)W. Chan, Douglas W. Allard, and Ivar Tombach
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
AeroVironment, Inc.
145 Vista Avenue
Pasadena, California 91107
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-3027
12. SPONSORING AGENCY NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report evaluates five techniques for measuring the transport of ozone and
precursors into an urban area. These techniques were tested in Philadelphia during the
summer of 1978. The data collected in the field program indicate that,, in general,
advection of ozone aloft is the main route by which pollution of photochemical
interest is transported into Philadelphia. Transport of ozone along the surface and
transport of oxides of nitrogen and non-methane hydrocarbons, both aloft and along
the surface, are minimal. Thus, the recommended techniques must primarily be able
to quantify the ozone transported aloft. Of the five techniques, three were
determined applicable for quantifying the ozone transported aloft. These three
techniques are: (1) fixed surface monitoring upwind; (2) a dedicated instrumented
aircraft; and (3) a free lift balloon ozonesonde system. The selection of one of these
techniques by an air pollution control agency would depend on the agency's technical
expertise, funding available, and the intended use of the data.
17.
ixt r WORLDS
f A^TAtYSIb
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATl Field/Group
Air Quality Monitoring
Air Sampling Networks
Ozone
Nitrogen Oxides
Hydrocarbons
Meteorology
Pollutant Transport
Ozonesondes
Aircraft Samplinlg
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
164
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDI TION is OBSOLETE
------- |