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
 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
 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 
                                                                                            o .t; .
      Q
      UJ  UJ
      (N tl
        St-
        
-------
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
    
  1. 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

  2. -------
          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
    

    -------