United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/4-84-011
August 1984
Air
Northeast Corridor
Regional Modeling
Project

Ozone and Precursor
Transport in New York
City and Boston During
The 1980 Field Program

-------
                                                  EPA-450/4-84-011
  NORTHEAST CORRIDOR REGIONAL MODELING PROJECT
Ozone and  Precursor Transport in New York City and
          Boston During  the 1980  Field Program
                                    by

                              Norman C. Possiel*
                      Office of Air Quality Planning and Standards
                            Office of Air and Radiation
                        U.S. Environmental Protection Agency
                     Research Triangle Park, North Carolina 27711

                                   and

                Chester W. Spicer, Philip R. Sticksel, and George M. Sverdrup
                          Battelle's Columbus Laboratories
                               505 King Avenue
                             Columbus, Ohio 43201

                                   and

                       Abdul J. Alkezweeny and William E. Davis
                       Battelle's Pacific Northwest Laboratories
                                  Box 999
                           Richland, Washington 99352
                     U.S. ENVIRONMENTAL PROTECTION AGENCY
                            Office of Air and Radiation
                      Office of Air Quality Planning and Standards
                     Research Triangle Park, North Carolina 27711
            "On assignment from the National Oceanic and Atmospheric Administration

-------
                                    DISCLAIMER


           This report has been reviewed by the Office  of Air  Quality
Planning and Standards, EPA, and approved for publication.  Mention  of
trade names or commercial products is not intended to constitute  endorsement
or recommendation for use.
                                        ii

-------
                                  CONTENTS
                                                                         Page
Disclaimer 	      ii
Fi gures    	       v
Tables     	      vi
Acknowledgments    	     vi i
Executive Summary  	    viii
   1.  Introduction    	        1
   2.  Program Scope  	        3
          2.1  Selection  of Case  Study  Days  	        5
   3.  Definition of Airflow on Study Days  	        7
          3.1  Synoptic-scale Trajectories   	        7
          3.2  Mesoscale  Trajectories  	        7
   4.  Interpretation of  Pollutant and  Meteorological
       Mea suremen ts     	        9
          4.1  Pollutant  Concentrations in  the Urban Plumes 	        9
               4.1.1  Maximum Pollutant Concentrations
                      from Morning Emissions 	        9
               4.1.2  Diurnal Pollutant Concentration  Profiles
                      in  Urban Air Parcels  	       14
               4.1.3  Transport Time and Distance to Peak
                      03  and N02	       19
               4.1.4  Downwind Distance to  Background NOX	       24
               4.1.5  Plumes from Medium Size Cities Between New York
                      and Boston	       26
          4.2  Pol 1utant  Transport 	       34
               4.2.1  Analysis Procedure	       34
               4.2.2  Discussion  of Ozone and Precursor Transport	       35
          4.3  Temporal Changes in Ozone Above the
               Boundary Layer	       43
                                       i i i

-------
         4.4   Comparison of Ozone  Levels at the
               Surface and Aloft  	      49
               4.4.1  Comparison  of Morning 03 Concentrations at
                      the Surface and Aloft	      50
               4.4.2  Comparison  of Afternoon 03 Concentrations at
                      the Surface and Aloft	      54
   5.  Concl us ions	      59
References	      64

-------
                                    FIGURES


Number                                                                   Page


 1       The study  region and the location of monitoring
         stations included  in this  study  	       4
         Time lines of  air  parcels crossing the New York City
         urban area at  0600, 0800, and 1000 EST and diurnal
         ozone concentrations at  sites in the vicinity of parcel
         tracks on  July 22, 1980  	      10

         Time history of air parcel  leading to maximum ozone on
         July 22,  1980  	      18

         Flight track for Boston  on  the afternoon of
         July 24,  1980  	      31

         Ozone distribution aloft downwind of Boston on the
         a fternoon  of July  24,  1980  	      33

         Surface ozone  concentration  isopleths (ppb), 0800 through
         2200 EST,  on June  24,  1980  	      41

         Characteristic ozone  profile types identified by
         Ludwig et al	      45

-------
                                      TABLES
Mumber                                                                    Page


 1       Case Study Days Selected for Analysis	       6

 2       Maximum Pollutant Concentrations Generated  in  Air Parcels
         Crossing New York City and Boston at Selected  Times  	      12

 3       Time, Position, and Concentration Profiles  for Air
         Parcels Generating Ozone Maxima  	      15

 4       Times and Distances from New York City to Maximum Ozone  	      21

 5       Times and Distances from Boston  to Maximum  Ozone  	      23

 6       Times and Distances to Maximum N02 Aloft in the
         New York City and Boston Urban Plumes  	      25

 7       Distance at Which NOX in Plume Approaches Background
         Concentration  	      27

 8       Ozone and Wind Direction at Kent County, RI
         on July 18, 1980 	      30

 9       Transport Regimes on Case Study Days 	      36

10       Average Transported Ozone and Precursors for Corridor  and
         Non-Corri dor Flow Regimes 	      39

11       Temporal Variations in Ozone Aloft 	      47

12       Comparison of  Mid-morning Surface Ozone With Early
         Morning Ozone  Aloft 	      51

13       Comparison of  Afternoon Ozone Concentrations at the
         Surface and Aloft in the New York Area  	   .   55

14       Comparison of  Afternoon Ozone Concentrations at the
         Surface and Aloft in the Boston Area 	      56

-------
                               ACKNOWLEDGEMENTS


           This report was prepared by EPA from the report "Analysis
and Interpretation of Ambient Measurements in  New York and Boston from
the 1980 Northeast Corridor Regional  Modeling  Project" submitted by
Battelle's Columbus and Pacific Northwest Laboratories under Contract
68-03-2958.  Additional analyses conducted by  EPA to supplement the
original submittal have been included in this  report.  The assistance
of various EPA staff in reviewing the report,  and the final  typing by
Zada Nelson, Carol Bradsher and Carole Mask, are greatly appreciated.

-------
                              EXECUTIVE SUMMARY

     This report describes the results of a  1-1/2 year  study  to analyze
portions of the data base obtained during the 1980  Northeast  Corridor Regional
Modeling Project (NECRMP).  The NECRMP data  base was  obtained primarily for
the application of urban and regional  scale  photochemical models to the
Northeast, including the major Corridor cities:  Washington,  DC; Baltimore;
Philadelphia; New York City; and Boston.  Although  the  NECRMP data base pro-
vides data primarily for model application,  the extensive data base is also
suitable for interpreting the meteorological  and chemical processes which
influence 03 formation and transport.
     The 1980 NECRMP data collection program involved extensive air quality
and meteorology measurements at the surface  and aloft.   Surface based meas-
urements available from selected sites in the Northeast Corridor include 03,
NO/NOX, NMOC, wind speed, wind direction, temperature,  solar  radiation, and
hydrocarbon species.  In addition, upper air meteorological measurements were
obtained by rawinsonde soundings, sodar, and pilot balloon  observations.
Instrumented aircraft were operated in the New York and Boston areas to
obtain measurements aloft of 03, NO/NOX, bscat,  temperature,  and relative
humidity.  The protocol for aircraft monitoring  flights required vertical
profiling of the atmosphere upwind of  the city in  the mornings and mapping
of  the urban 03 plumes in the afternoon.  Surface based measurements were
made throughout the summer, while aircraft monitoring was limited  to the
period from mid-July  through mid-August.

-------
     The present study focuses  on  20  days  of the  1980  NECRMP data base.
These case study days were divided between New York  City  (12 days) and
Boston (8 days)  and were selected  to  provide for  analyses of the  following
three meteorological situations:
     0  moderate wind speeds with  wind directions conducive to  interurban
        transport between Corridor cities;
     0  moderate wind speeds with  wind directions from sectors  for which
        interurban transport is unlikely;  and
     0  weak flows and stagnating  or  near  stagnating conditions.
The interpretive effort was designed  to address a number  of questions and
issues relating to 03 formation and transport in  the Northeast  Corridor.
These include:
     1.  Do concentrations of 03 and  precursors tranported into New  York
         City and Boston differ during along-Corridor and non-Corridor
         transport regimes?
     2.  What is the diurnal variation of 03 and precursors in  air parcels
         leading to the maximum ozone concentration  in the urban  plume?
     3.  What is the average transport time and distance  to maximum  03 in
         the urban plume?  What is the typical downwind distance  to  maximum
         N02 in the urban plume?
     4.  What is the typical downwind distance to where NOX in  the urban
         plume  becomes indistinguishable  from background  concentrations?
     5.  Can mid-morning surface 03 measurements be used  to estimate
         early  morning upwind 03 aloft?
     6.  Can mid-day aircraft measurements of 03 be used  to estimate
         surface concentrations for areas between measurement  sites?
      7.  Does 03 aloft,  initially isolated  from  the effects of surface
         emissions and scavenging, change substantially prior  to  the
         dissipation of  the nocturnal inversion when pollutants aloft
         are mixed  to the surface?
     8.   Is there evidence  in  the data of 03  plumes from medium size
         cities such as  Bridgeport, New Haven, and Hartford,  CT or
         Providence, RI?

-------
     A summary of the study and the important findings  pertinent to these
questions are presented below.
     1.  Do concentrations of 63 and precursors  transported into New York
         City and Boston differ during along-Corridor and non-Corridor
         transport regimes?
     The transport of 03 and precursors into New York City and Boston was
examined for the surface layer and for layers aloft.  Although the analysis
was to have categorized each day according to one of three flow regimes,
intepretation of trajectories and pollutant measurements indicates that more
than one type of flow regime occurred on about half of  the case study days.
On such days the direction of transport was rather complex and varied with
time and altitude.  Typically, overnight transport in the surface layer was
along  the Corridor, whereas aloft, above the. nocturnal  inversion, transport
was from areas west of the Corridor.
     Measurements at the surface and aloft were  partitioned and averaged by
flow regime with the following results.  For New York City, morning precursor
concentrations transported into- the urban area at the surface were twice as
high with along-Corridor transport than when transport  was  from outside the
Corridor.  Aloft, precursor concentrations, particularly NMOC, were also much
higher when transport was along the. Corridor.  The analysis indicates that
high morning precursor concentrations transported into  the  New York area with
along-Corridor flow were attributable to overnight emissions  in  the Philadelphia
area.  During mid-afternoon the impact of the Philadelphia  03 plume was typically
observed at one  or more monitoring sites on the upwind  (southwest) fringe of
New York City.

-------
     In Boston,  there was little  difference in  surface precursor concentrations
during along-Corridor versus non-Corridor transport regimes.   However,  the con-
centration of NMOC aloft during along-Corridor  transport was  triple the magni-
tude of aloft concentrations when transport was from outside  the Corridor.
Average 03 aloft transported into Boston during the morning was also higher
with along-Corridor flow (100 ppb versus 76 ppb).   The impact of the New York
City 03 plume on portions of the Boston area was observed during the evening
across the upwind (southwest) fringe of Boston  on  days when transport was
along the Corridor.
     2.  What is the temporal pattern of 03 and precursors in the air
         parcel  containing the maximum ozone concentration in the urban
         piume?
     Mesoscale trajectories and surface measurements were used to determine
the temporal pattern of 03, NO, N02, and NMOC in the air parcel which con-
tained the maximum 03 observed within the urban plume on each study day.  In
New York City, the concentration of 03 decreased as the air parcel traveled
from upwind rural and suburban locations into the urban area in the morning,
then increased rapidly later in the morning as the air parcel departed the
city and the rate of photochemical reactions increased.  In almost every
case,  the maximum 03 concentration in the plume occurred between 1300 and
1500 EST.  (In general, after 1500 EST, UV intensity decreases, and the rate
of dilution overcomes the rate of 03 production such that  the air parcel  03
concentration also  begins to decrease.)  The concentrations of NO, N02, and
NMOC generally increased as  the air parcel approached the  city in the morning.
After  leaving the city, the nitrogen oxide concentration fell, most likely
due to dilution  and chemical conversion to HN03, PAN, and  particulate nitrate.
                                     XI

-------
Unfortunately, there were insufficient NMOC sites  to  judge  the  temporal pattern
downwind beyond the urban area.  Hourly concentrations  of 03, NO, N02, and
NMOC are provided for ten of the case study days in Section 4.1.2.  (Analyses
pertaining to the temporal pattern of 03 and precursors in  the  Boston plume were
very limited since the plume was transported offshore on all  but one case study
day).
     3.  What is the average transport time and distance to maximum 03
         in the urban plume?  What is the typical  downwind  distance to
         maximum N02 in the urban plume?
     Surface measurements of 03, NO/NO?, and NMOC  were  examined in conjunction
with mesoscale trajectories to determine the maximum  pollutant  concentrations
formed in air parcels crossing the urban centers of New York  City and Boston
during the morning when precursor emissions are greatest.   For  New York City,
the results indicate that the highest 03 concerttrations were  associated with air
parcels crossing the city at approximately 0800 EST.   On the  average, air par-
cels crossing the city at 0600 EST yielded maximum 03 of 152  ppb at 1300 EST,
parcels crossing at 0800 EST generated an 63 maximum of 219 ppb at 1400 EST,
and those crossing at 1000 EST showed a maximum 03 concentration of 211 ppb
occurring at 1500 EST.  (Only one day was analyzed for Boston since the urban
plume was transported offshore on the other case study days.)
     The data base was analyzed  to determine the distance  and travel  time
from the center of both New York City and Boston to the location of the peak
03 concentration in  the urban  plume.  For New York City, the average  downwind
distance to maximum  03 measured  at the  surface was 93 km,  with an average
transport  time of just over 5  hours.  The average distance and travel  time  to
peak 03 aloft was  110  km  and 6.5 hours,  respectively.  In  the Boston  area,
aircraft data provided the most  useful  information relative to this  question
                                     xii

-------
since the urban plume was transported  over  the  ocean  on  all  but  one  case
study day.  The aircraft data  indicate that the average  distance to  the 03
maximum aloft was 81 km, with  a  range  of 61 to  107  km.   The  average  transport
time to peak 03 was 4 hours.
     The analysis indicates that maximum surface concentrations  of NO  and N0£
are found within or shortly downwind of the urban center.  For most  cases
studied, the highest concentrations of NO and N02 were associated with the  air
parcels crossing the city at 0800 EST.  Aloft,  the distance  to maximum N02
downwind during mid-day ranged from 23 to 48 km in  New  York  City and 34  to  110
km in Boston.  Estimated transport time to  maximum N02  was 1 to  3 hours  for
both urban areas.
     4.  What is the typical  downwind distance to where  NOX  in the urban
         plume becomes indistinguishable from background concentrations?
     For three days in Boston  and two days  in New York  City, the distance
from the urban area to the point where the  urban plume  NOX concentration
aloft became indistinguishable from the air mass background  NOX  concentrations
were estimated.  This distance ranged from 85 to 165 km.  The estimated travel
time to background NOX ranged from 5.5 to 12 hours.
     5.  Can mid-morning surface 03 measurements be used to  estimate
         early morning upwind 03 aloft?
     Measurements of 03 at the surface and aloft, and sodar-derived mixing
heights were analyzed to determine whether mid-morning  surface 03 measure-
ments during the period when the nocturnal  inversion is dissipating can  yield
information on 03 concentrations aloft prior to inversion break-up.  The analy-
sis indicates  that average surface 03 concentrations upwind  (3-hour average
centered around  the  time of inversion dissipation) do represent early  morning
03 concentrations aloft on many  study days.  However, there  were also  a number
                                    xm

-------
of cases when such an assumption  would, lead to  significant underestimation (or
overestimation)  of the early morning levels aloft.   In  these cases, it appears
that 03 aloft transported into the city had actually increased (or decreased)
between the time of the early morning aircraft  measurements and the time
of inversion dissipation.  This may have been due  to reactions among 03 and
transported precursors or to spatial variations in  03 aloft transported across
the urban area.   Thus, using surface data for estimating early morning con-
centrations aloft should be done  with caution,  particularly in situations where
urban areas are in fairly close proximity or high  concentrations of transported
precursors are expected.
     6.  Can mid-day aircraft measurements of 03 be used to 3Stimate
         surface concentrations between measurement sites?
     Afternoon surface 03 measurements were compared to mid-boundary layer
(~ 800m) aircraft measurements made in the vicinity of  surface monitoring
stations.  The results indicate that, in locations where the atmosphere appears
to be well mixed, the surface and aircraft data agree within 10 to 15 ppb.
However, large differences were, observed between surface and aircraft data in
most comparisons near the edge of urban plumes, and in  conjunction with the
internal boundary layer  produced by sea breeze  wind flows.  These differences
were the result of strong vertical and horizontal  concentration gradients
associated with such  features.  It is concluded that aircraft  data should be
used with caution for estimating surface concentrations in areas  subject  to
sea  breeze circulations  or other mesoscale meteorological  flows,  and in areas
of large gradients at the edge of urban plumes.  However,  mid-day aircraft
data can provide a reasonable estimate of area-wide surface 03 concentrations
away from  the preceding  complex situations.
                                     xiv

-------
     7.   Does  03 aloft, initially isolated from the effects of surface
         emissions and scavenging, change substantially prior to the
         dissipation of the nocturnal inversion when pollutants aloft
         are mixed to the surface?
     Temporal  variations in 63 concentrations aloft, in air parcels isolated
from boundary  layer influences (e.g., emissions, scavenging, etc.) were investi-
gated using aircraft pollutant and meteorological measurements.  The results
indicate that, in most of the cases examined, 03 concentrations aloft were
fairly stable  (within ± 5 ppb) between early morning (0500/0600 EST) and
mid-morning (1000/1100 EST) measurements within an air parcel.  However, 03
production was evident in those air parcels containing comparatively high air
mass NOX concentrations and probably other 03 precursors, as indicated by the
aerosol  content and estimated track of the air parcel relative to upwind urban
areas.
     8.   Is there  evidence in the data of 03 plumes  from medium size
         cities such as Bridgeport, New Haven, and Hartford, CT or
         Providence, RI?
     The surface and aircraft data were examined in  an attempt to identify
03 plumes from other  smaller Corridor cities, such as Bridgeport, New Haven,
and Hartford,  CT,  and Providence, RI.   In general, it was difficult  to
define 03 plumes from  such cities due  to  the relatively high air mass 03
levels, the complexity of airflow patterns, and  the  frequent incursions of
urban plumes from  the major Corridor cities.  Also,  since the monitoring
program was not directed  toward  investigating  these  cities, comparatively
little data were available  for  this  type  of analysis.   However, on  two occa-
sions, there was evidence of  the  Providence 03  plume from the  Boston area
aircraft data.  In the  clearest  example,  03 in  the  Providence  plume  was
20  to 30 ppb higher than  the  air mass  03  concentration  upwind  of  both cities.
                                    xv

-------
                                  SECTION 1
                                 INTRODUCTION

     Investigations  since  the mid-1970s  have demonstrated  that ozone  (03)
is a pervasive contaminant of air over the northeastern United States. 1-5
Studies have shown  that ambient  concentrations exceeding the 0.12 ppm  (120 ppb)
National  Ambient Air Quality Standard (NAAQS) level  for 03 are observed  in many
parts of the Northeast, and concentrations up to  300  ppb have been  observed
downwind of cities  in this highly urbanized corridor.  Also, ozone  concentra-
tions exceeding the  NAAQS  are  frequently observed over a large portion of  the
region, and field study results  have shown that both  interurban and long
distance transport of 03 and its precursors should be considered  in designing
control strategies.
     In order to devise effective and equitable strategies for reducing  the
concentration of 03  in the Northeast,  the Environmental Protection  Agency  (EPA)
has been conducting  a major long-term program with both field measurement  and
modeling components.  The  field  measurement programs  associated with the
Northeast Corridor Regional Modeling Project  (NECRMP) were designed to develop
a data base for regional/urban model verification/application, as well as  to
improve understanding of the chemical and meteorological phenomena  resulting in
high regional 03 concentrations.  The major urban data collection program  of
NECRMP was conducted during the  summer  of 1980.   The purpose  of  that effort was
to obtain a data base for  the  application of  the  Airshed^  urban  photochemical
model to Northeast Corridor cities,  including  Washington,  DC,  Baltimore, New
York City, and Boston.  Regional monitoring  studies were also  conducted  during

-------
1980 as part of NECRMP.  Although NECRMP was  designed  primarily to provide

data for model application, the extensive data  base  is also  suitable for

interpretation of meteorological  and chemical  processes which  influence 03

formation and transport.

     This report describes the results of a  program  designed to address a

number of questions and issues relating to 03 formation and  transport in the

New York City to Boston portion of the Northeast Corridor.   These questions/

i ssues include:

     1.  Do concentrations of 03 and precursors transported  into New York City
         and Boston differ during along-Corridor and non-Corridor transport
         regimes?

     2.  What is the diurnal variation of 03 and precursors  in air parcels
         leading to the maximum 03 concentration in  the urban  plume?

     3.  What is the average transport time  and distance to  maximum 03 in
         the urban plume?  What is the typical  downwind distance to maximum
         N02 in the urban plume?

     4.  What is the typical downwind distance to where NOX  in the urban plume
         becomes indistinguishable from background concentrations?

     5.  Can mid-morning surface 63 measurements be used to  estimate early
         morning upwind 03 aloft?

     6.  Can mid-day, aircraft measurements of 03 be used to  estimate surface
         concentrations in areas between measurement sites?

     7.  Does 03 aloft, initially isolated from the effects  of surface emissions
         and  scavenging, change substantially/ prior to the dissipation of the
         nocturnal inversion when pollutants aloft are mixed to  the surface?

     8.  Is there  evidence in  the data of 03 plumes from medium  size cities
         such as Bridgeport, New Haven, and Hartford,  CT or  Providence, RI?

-------
                                   SECTION 2
                                 PROGRAM SCOPE

     The 1980  NECRMP  urban measurement  programs  in  New  York  and  Boston  included
extensive air  quality and meteorological measurements at the surface  and  aloft.
Surface-based  measurements available  from  sites  in  the  Northeast Corridor
include 03, nitric oxide/nitrogen dioxide/oxides of nitrogen (NO/N02/NOX),
nonmethane organic compounds (NMOC),  wind  speed, wind direction, temperature,
solar radiation,  and  hydrocarbon  species.   In  addition,  30-minute average
mixing heights were derived  from  monostatic  sodar,  and  temperature and/or winds
above the surface were obtained by rawinsonde  soundings and  pilot balloon
observations (pibals).  Measurements  made  aloft  by  instrumented  aircraft  in
the New York and Boston areas include 03,  NO/NOX, light scattering coefficient
(bscat), temperature, and relative humidity.   Surface-based  measurements  were
made throughout the summer,  while aircraft monitoring was limited to  the  period
from mid-July  through mid-August.  A  map  of the  study  area is shown in  Figure  1
along with the location of surface monitoring  sites included in  this  analysis.
     The present study was designed to  analyze 20 days  of the 1980 NECRMP data
base.  These case study days were selected to  provide  for analysis of the
following three meteorological situations:
     (1)  moderate wind speeds with wind  directions conducive to interurban
          transport between  Corridor  cities;
     (2)  moderate wind speeds with wind  direction  from sectors  for which
          interurban  transport is unlikely; and
     (3)  weak flows  and stagnation or near stagnation  conditions.

-------
   5   «i«i
          78*N 75*30«  7?M  71*JOM  71* H  7J*Mt4  7TH  TnOH  7TH  7ITOM  71 M  70*30«
                                                LONGITUDE
atoN  6TH  aa*»M
Figure  1.   The  study region  and the  location  of monitoring stations included in  this study.

-------
     The scope  and organization of this report are linked directly to the
Statement of Work  for  Contract 68-03-2958,  "Analysis and Interpretation
of Ambient Measurements in  New York and Boston from the 1980 Northeast
Corridor Regional  Modeling  Project,"  conducted by Battelle's Columbus and
Pacific Northwest  Laboratories.

2.1  Selection  of  Case Study Days
     At the start  of the program, 20  days were selected for analysis out of
the approximately  3 months  of the 1980  NECRMP  field program.  These study
days were divided  between New York City and Boston, and it was  desired  to
select days on  which the airflow  fit  certain criteria.  These criteria  were
listed earlier and basically describe days  on  which interurban  transport is
(1) likely, (2) unlikely, or (3)  days of near  stagnation.  The  data base for
the entire NECRMP  study was screened  to  identify candidate study  days meeting
the airflow criteria.   Hourly surface wind  speed, wind direction, and 03
concentration data were used, together  with pertinent aircraft  and upper air
wind measurements, to provide a  preliminary classification of the candidate
days into the three airflow categories.   From  the list of candidate days meet-
ing the airflow criteria, final  study days  were  selected using  additional
criteria including the availability  of  aircraft  data, completeness of  the  sur-
face data set, level of pollutant concentrations, etc.  The  New York area  was
the focus of 12 days and Boston  was  the focus  of 8  days.   The  study days are
listed  in Table 1.

-------
              Table 1.  Case Study Days  Selected  for Analysis
                               New York ,Area


June 24   Tue.                 July 21*   Mon.                Augusts*  Wed.
June 25   Wed.                 July 22*   Tue.                Augusts*  Fri.
July 16*  Wed.                 July 24*   Thu.                August 26  Tue,
July 18*  Fri.                 July 31*   Thu.                August 28  Thu,
                                Boston Area

    June 24   Tue.                                      August 1*  Fri.
    July 15*  Tue.                                      August 5*  Tue.
    July 16*  Wed.                                      August 6*  Wed.
    July 17*  Thu.                                      Augusts*  Fri.
*Aircraft data available.

-------
                                  SECTION  3
                     DEFINITION  OF AIR  FLOW  ON  STUDY  DAYS

3.1  Synoptic-Scale Trajectories
            Synoptic-scale trajectories were used in  this analysis  to  determine
the probable upwind source areas of 03  transported into  New  York  City  and
Boston, and the general  downwind direction of urban plume transport.   For-
ward and backward trajectories were prepared for five cities in  the Northeast
Corridor:  Boston; New York City; Philadelphia; Baltimore; and Washington,  DC.
            These trajectories were computed using the ATAD  trajectory model
developed by Heffter et al.?  The trajectory calculations were made from 6-hour
and 12-hour National Weather Service (NWS) upper air wind data averaged through
the layer from 500 to 1500m.  Trajectories were computed in  3-hour  increments,
backward and forward for 48 hours, from all  five cities.  In addition, forward
trajectories using wind data through the layer from 200  to  1000m were  computed
for New York City, Philadelphia, and Washington, DC.  The  start/end time for the
forward/backward synoptic trajectories  was 0700 EST.

3.2  Mesoscale Trajectories
            Mesoscale forward and backward trajectories were computed  for
points within the New York City and Boston urban areas.   Used in the compu-
tation were 6-hour and 12-hour NWS upper air wind data, along with data from
*A revised version of this model is now available from Heffter.8  However,
because the trajectories in this analysis were computed for a fixed layer and
were not used to specify space/timem source/receptor relationships it is unlikely
that using trajectories computed with the revised model would alter the conclusions
of the study.

-------
the special rawinsonde soundings and plbal  observations  obtained as part of
the 1980 NECRMP ambient monitoring program.  The pibal observations were
hourly, whereas the rawinsonde soundings were conducted  at various times
between 0500-1500 EST.  The winds were averaged over the appropriate layers
and interpolated hourly between measurement times to fill  in  missing data.
These data, in turn, were used to produce gridded wind  fields covering  the
area of interest.  The interpolation and trajectory computation  techniques
are based on procedures described by McNaughton, et al .9
            Mesoscale trajectories were computed starting at  0600, 0800, and
1000 EST from each of three locations in the New York urban area and two
locations  in Boston.  Forward trajectories were computed hourly  for 10  hours
and the backward trajectories for 5 hours for these locations and  start times.
            The back trajectories ending at 0600, 0800,  and 1000 EST were used
to represent the flow above the nocturnal surface layer, upwind  of the  urban
area.  The layer chosen for these trajectories was 250m  to 1000m.  The  lower
limit of 250m was selected to avoid the effects of wind  shear near the  ground
during the early morning  hours.  For the forward trajectories, wind measure-
ments were averaged within the layer from 50m to the top of the  boundary layer.
An estimate of the depth  of the boundary layer on an hourly basis  for each  day
was developed from the available temperature soundings.   These estimates rep-
resent the overland, convectively induced mixing heights in the  area of
interest.  For cases of a morning,  surface based stable layer, a mixing height
of 250m was used.

-------
                                SECTION 4



        INTERPRETATION  OF  POLLUTANT  AND METEOROLOGICAL MEASUREMENTS





4.1  Pollutant Concentrations  in the Urban Plume



     Several  characteristics of the  New York  City and Boston urban plumes were



examined as part of this  task.   These include the travel  time and downwind



distance to peak concentrations of 03, NO, and N02 and the  travel time and



downwind distance to where NOX in  the urban plume diminished to  the estimated



air mass background concentration.  In addition, the data were examined  for



evidence of 03 plumes from several medium  size cities between New York City



and Boston.





     4.1.1  Maximum Pollutant  Concentrations  from Morning Emissions



            The objective of this  task was to identify the  maximum concentra-



tions of 03, NO, and N02  associated  with air  parcels crossing the urbanized



portions of New York City and  Boston during  the morning  when precursor emis-



sions are at a maximum.  The analysis procedure included the use of mesoscale



trajectories to estimate  the downwind track  of air  parcels  initially  over  the



city at each of three morning  time periods:   0600; 0800; and  1000 EST.   Iso-



pleth maps of 03 concentrations for  various  times during the  day were used



to confirm the path of the urban plume.   Time histories  of  03 concentration



within each of the three  air  parcels were  estimated  from surface monitoring



data in the vicinity of the trajectory track  using  the  procedure shown  in



Figure 2.  In this figure, the intersection  of the  heavy lines  with  the  03



diurnal profiles represents the 03 concentration  in  the  air parcel  that  had



crossed the city at 0600, 0800, or 1000  EST as noted.

-------
     240
                                                 July 22,1980
      0800    1000
                    1200    1400     1600
                              Tim«, EST
                                         1800
                                                 2000
                                                       2200
Figure 2.   Time lines of air  parcels crossing the  New York City
            urban area at 0600,  0800, and 1000 EST  and diurnal
            ozone concentrations at sites in the  vicinity of
            parcel tracks on July 22, 1980.
                              10

-------
            As  indicated in the example in Figure 2, the air parcels
containing morning  emissions  from New York City on July 22, 1980 were trans-
sported northeastward across  Connecticut during the day and approached the
Boston area by  evening.  The  maximum measured 03 concentrations were 169 ppb,
227 ppb, and 226  ppb along  the 0600, 0800, and 1000 EST trajectories, respec-
tively.  All of these values  occurred in Connecticut.
            Of  the  20 case  study days, 10 were excluded from this task because
the wind flow on  those  days transported the urban plume out over the Atlantic
Ocean away from surface monitoring  sites.  The maximum 03 concentration esti-
mated from morning  emissions  on  the remaining 10 case study days are provided
in Table 2.  Also shown in  this  table are the maximum concentrations of NO
and N02, along  with times  and locations of these maxima.  Unfortunately, the
coverage available  for  NO  and N02 is not nearly as comprehensive as  for 03,
due to the much smaller number of stations which monitored  these pollutants.
In any case, it is  clear  from Table 2  that both NO and N02  typically reach
their peak values within  or shortly downwind of the urban area.  Even though
N02 is being produced  in  the  air parcels during the first hours of  transport,
the rate of production  is  apparently overwhelmed  by the rate of dilution.   In
most cases, the highest concentrations of NO and  N02 are associated  with the
0800 EST air parcel.
            Inspection  of the 03 data  in Table 2  reveals that  the highest 03
maxima are associated  with air parcels departing  the city at 0800 EST.  The
concentrations of 03 precursors  in  air leaving  the  city at  0600 EST  are not
as high initially and  are no  doubt  reduced even more  than  the  later  parcels
by dilution prior  to the onset of  extensive  photochemical  reaction.  The
                                     11

-------
Table 2.  Maximum Pollutant Concentrations Generated in Air Parcels
          Crossing Mew York City and Boston at Selected Times
Time of Air
Parcel Departure, Hax. 03
Date EST ppb
6-24-80


6-25-80


7-15-80


7-16-80


7-21-80


7-22-80


8-1-80
(New >York Source)

8-1-80
(Boston Source)

8-6-80


8-8-80


0600
0800
1000
0600
0800
1000
0600
0800
1000
0600
0800
1000
0600
.0800
1000
0600
0800
1000
0600
0800
1000
0600
0800
1000
0600
0800
1000
0600
0800
1000
140
220
230
160
275
180
160
190
145
180
230
290
175
240
303
170
220
225
122
140
125
110
145
125
125
250
190
140
205
215
Time
EST
1500
1530
1430
1400
1300
1430
1600
1500
1700
1200
1400
1400
1600
1300
1500
1300
1500
1400
1200
1300
1500
1100
1600
1130
1100
1300
1500
1330
1500
1630
Hax. NO
Location ppb
Kent Co.. RI
Mlddletown. CT
Stratford, CT
Mtddletown, CT
Stratford, CT
Stratford, CT
Georgetown, MA
Worcester, HA
Worcester, MA
Mlddletown. MA
Mlddletown, MA
New Haven, CT
Kent Co. , RI
Stony Brook, NY
Stratford, CT
Mlddletown, CT
Mlddletown, CT
Stratford, ,CT
Litchfleld, CT
Litchfleld, CT
l.itchfield, CT
Portsmouth, NH
Gardiner, ME
Sagamore Hill, MA
New Haven, CT
Stratford, CT
Stratford, CT
Kent Co., RI
Kent Co., RI
Kent Co., RI
051
037
005
038
—
005
061
025
009
068
006
Oil
040
Old
Oil
03U
026
004
054
010
002
040
005
005
057
039
002
065
039
003
Time
EST
0600
0800
1000
0600
—
1000
0600
0800
1000
0700
0800
1000
0700
0800
1000
0600
0800
1100
0600
0800
1000
0600
0900
1000
0600
0800
1100
0700
0900
1100
Location
Manhattan,
Manhattan,
'Manhattan,
Manhattan,
—
Manhattan,
Manhattan,
Manhattan,
Manhattan,
Queens, NY
Manhattan,
Manhattan,
Queens, NY
Manhattan,
Manhattan,
Manhattan,
Manhattan,
Queens, NY
Dumont , NJ
Manhattan,
Manhattan,
Max. N02 Time
ppb EST
NY
NY
NY
NY

NY
NY
NY
NY

NY
NY

NY
NY
NY
NY


NY
NY
East Boston, MA
Danvers, MA
East Boston
Manhattan,
Manhattan,
Queens, NY
Queens, NY
Queens, NY
Queens, NY

, MA
NY
NY




070
117
068
059
—
063
073
076
059
079
095
074 '
100
095
078
082
062
069
108
080
040
100
115
035
067
085
076
097
105
075
0600
0800
1000
0600
—
1000
0600
0800
1000
0700
1000
1000
0700
0800
1000
0600
0800
1100
0600
0800
1000
0600
0800
1000
0600
0800
1100
0700
0900
1100
Location
Manhattan, NY
Manhattan, NY
Manhattan, NY
Manhattan, NY
—
Manhattan, NY
Manhattan, NY
Manhattan, NY
Manhattan, NY
Queens , NY
Glen Cove. NY
Manhattan, NY
Queens, NY
Manhattan, NY
Manhattan, NY
Manhattan, NY
Manhattan, NY
Queens, NY
Manhattan, NY
Manhattan, NY
Manhattan, NY
East Boston, MA
East Boston, MA
East Boston, HA
Manhattan, NY
Manhattan, NY
Queens, NY
Queens, NY
Queens, NY
Queens, NY

-------
reduced photochemical  activity is due to the relatively low solar intensity
at this early hour.
            On the average, air parcels leaving New York City at 0600 EST
yielded maximum 03 of 152  ppb at 1300 EST, parcels departing at 0800 EST
generated an 03 maximum of 219 ppb at 1400 EST, and those leaving at 1000 EST
showed a maximum 03  concentration of 211 ppb, with the peak occurring at
1500 EST.  The times of the maxima reflect the combined effects of reaction
time, solar intensity, and dilution.  Reduced solar intensity and continuing
dilution in the late afternoon eventually overcome the effects of increased
reaction time, so that the highest concentrations on a given day usually
occur before 1600 EST.
            On five  of the 10  days analyzed  (June 24 and 25, July 16 and 21,
and August 6), the trajectories  indicate a westerly flow in the New  York area
although the track of the  surface 03 plume was to the northeast across
Connecticut into eastern Massachusetts, consistent with the low level wind
flow.  Examinations  of surface and upper air wind observations in the area
indicate that this situation  may  reflect the combined effects of the onshore
sea breeze circulation near  the  coast,  coupled with topographic channeling of
the low level wind flow northward within the broad valley  from New  Haven to
Hartford, CT.  As a  result of this  flow regime,  the New York City plume on
these days was transported across Long  Island Sound (and adjacent land areas
of coastal Connecticut and Long  Island),  then northeastward into central
Connecticut.  Because the sea breeze  flow  is most pronounced within  400 to
500m of the surface it was apparently  smoothed out  in  the  mesoscale  trajectory
computations which included  winds  through  a  much  deeper layer.
                                     13

-------
            It should be noted that the diurnal profiles for sites along the
urban plume track provide strong prima facie evidence for 03 transport in
urban plumes.  In nearly every case, the peak 03 concentration for each site
occurs at progressively later times along the plume track.  In several  cases,
the 03 maxima for the sites farthest downwind of the urban center occurred in
the late afternoon or evening after the sunlight intensity, the driving force
behind 03 production, had dropped off considerably, supporting the contention
that transport is the prime contributor to the observed peak 03 concentration.

     4.1.2  Diurnal Pollutant Concentration Profiles in Urban Air Parcels
            This task was aimed at determining the time history of 03, NO, N02,
and NMOC concentrations in air parcels which lead to the observed maximum 63
in the urban plume.   In order to determine the time history of the pollutant
concentrations, the path of the air parcel leading to the maximum concentration
was determined from trajectories and isopleth maps as in the previous task.
With this information, monitoring stations in  the vicinity of the air parcel
track were selected to provide the pollutant concentrations within the plume
at times prior to and after the observed maximum concentration.
            As in the previous section, days were excluded from the analysis
if the urban plume was carried out over the ocean or over a land area with
few monitoring stations.  This criterion  resulted in the exclusion of most
Boston study days and some of the New York cases.  However, data from two
additional days  (July 15 and August 1) were included for the New York area
to supplement the analysis.  The summary  of 03 concentration time histories
is given  in  Table  3.  Data on NO, N02, and NMOC are also tabulated, but these
data  are  limited by  the  fact that few of  the monitoring  sites  for these
pollutants are outside  the urban area.
                                        14

-------
Table  3.   Time, Position, and Concentration Profiles  for Air Parcels Generating Ozone  Maxima
Date
6/24/80






6/25/80





7/15/80







7/16/80







7/21/80








Time, EST
0600
0700/0800
0800/0900
1130
1400
1500
1700
0500
0600/0700
0700/0800
1030
1300
1600
0500
0600/0700
0700/0800
0900
1100
1300
1500
1730
0600
0800/0900
0900/1000
1100
1330
1400
1600
1800
0700
0800/0900
0900/1000
1100
1200
1400
1500
1700
1800
Location
New Brunswick, NJ
Linden/Bayonne, NJ
New York City, NY
Greenwich, CT
Stratford, CT
New Haven, CT
Middletown, CT
New Brunswick, NJ
Linden, NJ
New York City, NY
Greenwich, CT
Stratford, CT
Middletown, CT
New Brunswick, NJ
Linden/Bayonne, NJ
New York City, NY
Glen Cove, NY
Stratford, CT
Middletown, CT
Worcester, MA
Georgetown, MA
New Brunswick, NJ
Linden/Bayonne, NJ
New York City, NY
Queens, NY
Stratford, CT
New Haven, CT
Middletown, CT
Kent Co., Rl
New Brunswick, NJ
Linden/Bayonne, NJ
New York City, NY
Queens, NY
Glen Cove, NY
Stony Brook, NY
Stratford, CT
Middletown, CT
Kent County, RI
03, ppb
006
041
031
149
253
170
159
019
033
027
119
276
165
017
012
019
078
135
162
193
142
004
018
033
103
254
291
262
112
032
042
084
133
202
240
303
262
200
NO, ppb
029
015
027
004
—
—
—
___
012
032
005
—
—
010
032
041
003
—
—
—
...
041
020
025
006
—
—
—
---
013
004
008
002
005
—
	
__.
—
N02, ppb
058
043
100
022
—
—
—
...
040
072
022
—
---
032
057
076
045
—
—
—
---
056
054
080
055
—
—
• —
—
031
032
064
046
015
—
	
~_ _
—
NMHC, ppbC
630
955
1890
	
	
	
----
....
420
1620
	
	
	
120
233
930
—
—
—
—

690
293
1125
	
	
	
	
	
300
360
1620
	
	
	
	
	 	 	
	
a Measurement below detectable Unit of the Instrument.

-------
Table  3.  Time,  Position,  and Concentration  Profiles  for Air Parcels Generating Ozone Maxima (continued)
           Date
Time, EST
Location
03,  ppb
NO, ppb
                                           N02, ppb
                                                                                               NMHC, ppbC
7/22/80









8/1/80
(NEW YORK)






8/1/80
(BOSTON)





8/6/80






8/8/80







0600
0600/0700
0700/0800
1000
1100
1300
1400
1500
1700
1900
0600
0600/0700
0700/0800
0900
1000
1200
1300
1500
0700
0800
1000
1200
1400
1600
2000
0500
0600/0700
0700/0800
0900
1000
1300
1400
0600
0700/0800
0800/0900
1030
1200
1300
1600
1830
New Brunswick, NJ
Llnden/Bayonne, NJ
New York City, NY
Glen Cove, NY
Greenwich, CT
Stratford, CT
New Haven, CT
Middle town, CT
Stafford, CT
Sudbury, MA
New Brunswick, NJ
Uinden/Bayonne, NJ
New York City, NY
Dumont, NJ
White Plains, NY
Oanbury, CT
Litchfield, CT
Agawam, MA
Medfield, MA
Boston, MA
Sagamore Hill, MA
Portsmouth, NH
Cape Elizabeth, ME
Gardiner, ME
Penobscot Co., ME
New Brunswick, NJ
Llnden/Bayonne, NJ
New York City, NY
Queens College, NY
Glen Cove, NY
Stratford, CT
New Haven, CT
New Brunswick, NJ
Linden/Bayonne, NJ
New York City, NY
Glen Cove, NY
Stony Brook, NY
Stratford, CT
Kent Co., RI
Easton, MA
005
009
025
064
127
213
227
220
197
134
004
018
032
049
062
120
140
075
058
073
103
127
133
143
082
009
017
021
052
085
249
168
003
034
031
084
143
246
222
095
032
020
034
008
003
—
—
—
—
—
036
027
017
013
—
—
—
—
007
LD«
LD
—
—
—
—
G17
014
030
006
LD
—
—
038
008
032
—
—
—
—
« — —
043
042
071
061
017
—
—
—
—
—
050
055
074
075
—
—
—
—
006
115
023
—
—
—
—
020
038
077
053
051
—
—
037
045
092
—
—
—
—
""
630
300
1298
	
	
	
	
	
	
	
810
495
810
	
	
	
	
	
	
0830
	
	
	
	
— ._
300
975
2168
	
	
	
	
	
578
2068
	
	
	
	
....
         aMeasurement below detectable limit of the instrument.

-------
            In New York City,  data  from five urban  sites were  used to
characterize 03 and precursor  concentrations in  the air parcel  as it passed
within the urban area.  Two-hour average concentrations were computed for two
urban sites in New Jersey (Linden and Bayonne)  and  for three urban sites in
New York (Manhattan, Astoria,  and Brooklyn).  These average values were used
to partially account for a decrease in the reliability of the  trajectories
for estimating transport within the core of the urbanized area.
            The data for July  22, 1980 were plotted as an example profile
from the data given in Table 3.  The time history plot is shown in Figure 3.
This figure shows the concentrations of NO, N02» and 03 in the air parcel which
generated the highest surface  concentration of 03 on this day.  The time of day
is given along the bottom of the graph and the air parcel position along the
top.  This particular parcel crossed northern New Jersey before 0800 EST and
the concentrations of NO and N02 increased as the air approached the metropol-
itan New York area.  The NO reached a peak at 0800 EST directly over New York
City, whereas N02 appears to have peaked at 0900 EST on the downwind fringe of
the city.  The level of NOX dropped off rapidly after the parcel left the
urban source area.  In contrast, the 03 concentration decreased slightly as  the
air parcel approached New York City in the morning but  increased rapidly after
leaving the city.  The rapid rise in 03 concentration occurred at the time  the
sodar data indicate a rapid rise in the boundary layer.  Morning vertical 03
profiles upwind of New York City indicate  that  the 03 concentration above the
surface based inversion averaged 56 ppb (650 to 1500m).  Hence, photochemical
reactions, rather than mixing of 03 stored  aloft overnight, must have been
responsible for the majority of the 03 generated on this day,  although mixing
                                      17

-------
0.30
                      0.12
                                                                           — 0.10
                                                                           — 0.08
                                                                           — O.OB
                                                                                 <
                                                                                 cc
                                                                                 cj
                                                                                 z
                                                                                 o
                                                                                 CJ

                                                                                  cvi
                                                                                 O
                                                                           — 0.04
                                                                           — 0.02
                                   1200       1400

                                    TIME (EST).hr
1600
1800
2000
    Figure  3    Time history of air  parcel  leading  to maximum ozone
                on  July 22,  1980.
                                      18

-------
of 03 from aloft undoubtedly  accelerated  the  photochemical  reactions  by
increasing the rate of NO oxidation  to  N02-   On  this  day  the maximum  measured
03 concentration occurred at  1400  EST at  New  Haven, CT.   As the air parcel
continued to move downwind throughout the afternoon and into the  evening,
the 03 concentration decreased.   Nevertheless,  this same  parcel was apparently
responsible for the maximum 03 measured at Stafford,  CT (1700  EST), and at
Sudbury, MA (1900 EST).
            In general, the data in  Table 3 indicate  that the  concentration
of 03 decreased as the air parcel  moved into  the urban area in the morning,
due to processes such as chemical  scavenging  reactions.   Ozone then  increased
rapidly later in the morning  as  the  air parcel  left  the city and  the  rate of
photochemical reactions increased.  In  almost every case, the  maximum 03 con-
centration in the urban plume occurred  between  1300 and 1500 EST. After 1500
EST, UV intensity generally decreases,  and the  rate of dilution overcomes  the
rate of 03 production such that the  air parcel  03 concentration  begins to
decline slowly.
            The concentrations of NO, N02, and NMOC  generally  increased as
the air approached the city in the morning.  After leaving the city,  the  NOX
fell, likely due to dilution  and probable chemical conversion  to  HN03, PAN,
and particulate nitrate.  Unfortunately,  there are insufficient MMOC data
to evaluate the time history of hydrocarbon concentrations.

     4.1.3  Transport Time and Distance  to Peak 03 and N02
            To determine  the distance and travel time from New York  City  and
Boston  to  the peak 03 and N02 concentrations measured in  the  urban plumes of
                                      19

-------
each city, information on the location of maximum 03  prepared  in  the previous
section was enhanced with 03 and N02 meaurements aloft.
            The times and distances to the maximum 03 concentrations at the
surface and aloft in the New York area are shown in Table  4.   Because of the
complex flow during stagnation conditions on August 26,  1980,  the transport
time was not computed.  The surface data for three other days  (July 18 and 31
and August 28) were also excluded from the analysis,  because  the  northwest
flow aloft on these days transported the plume out over  the Atlantic Ocean.
Recirculation of portions of the plume onshore in sea breeze  flows later in
the day made it difficult to estimate travel distance and  transport time.
The actual surface maximum may have occurred over the ocean.
            Among the remaining days, the average downwind distance to maximum
03 at  the surface was 96 km, with a mean transport time  of just over 5 hours.
On the other hand, the average distance to peak 03 aloft was  110  km, with a
mean transport time of 6.5 hours.  (Direct comparisons between the surface
and aloft values are somewhat misleading, because different days  were  involved
and day-to-day meteorology varied considerably.)  In  general,  for the  cases
studied, one can characterize the transport time to 03 maxima  in  the New York
City plume as 5 to 7 hours, with a typical downwind  distance  of 100 km.
            For Boston, two difficulties were encountered  in  defining  the
maximum 03 concentration within the urban plume.  First, transport from  upwind
sources was frequently responsible for  the highest surface 03 concentration  in
the Boston area.  Second, on ocean transport days, when the  plume was  carried
offshore, the highest concentration in  the plume may not have been measured,
even at coastal monitoring sites during  sea breeze flows.   These situations
                                      20

-------
            Table 4.   Times and  Distances from New  York  City  to Maximum Ozone
Date
6/24/80
6/25/80
7/16/80
7/18/80
7/21/80
7/22/80

7/24/80
7/31/80
8/6/80
8/8/80
8/26/80

8/28/80
Site
Stratford, CT
Stratford, CT
New Haven, CT
a
Stratford, CT
New Haven, CT

a
a
Stratford, CT
Stratford, CT
Linden, NJ
New Brunswick, NJ
b
03 Max.,
ppb
(Ground)
253
276
291
—
303
227
...
...
...
249
246
188
188
...
Time,
EST
1400
1300
1400
	
1500
1400
	
	
	
1300
1300
1400
1300
	
Distance
90
90
110
—
90
110
—
...
...
90
90
U9
50
—
Transport
Time,
hours
5-6
5-6
4-5
—
5-6
5-6
...
...
—
5-6
4-5
	
...
...
03 Max.
PPb,
Aloftd
...
—
2156
249
...
223^
220
152e
269*
352
242
	

...
Time,
EST
	
	
1530
1420
	
1355
1700
1452
1627
1519
1322
	
	
	
Distance,
...
...
103
51
...
91
150
110
124
126
121
	
...
...
Direction
from
CHy
...
...
NE
SE
—
NE
NE
S
ESE
E
NE
	
...
—
Transport
Time,
hours
...
—
4
6
...
5
7-8
5-6
7
7
4
...
—
...
Location
	
	
near Derby, CT
south of Long Island
	
near Bridgeport, CT
near Middle town, CT
near Barnegat, NJ
south of Westhampton, LI
north of Riverhead, LI
over Long Island Sound
	
	
	
?  Northwest air flow;  maximum 03 over Atlantic Ocean.
D  Recirculatlon air flow; maximum 0? may have occurred over Atlantic Ocean.
c  Distance from Manhattan, NY.
d  Measurement at 700-800m M5L unless otherwise noted.
e  May not be maxima 0->; measurement listed was obtained on furthest downwind traverse of plume.
•  nay not oe maxim* uj; measurement MSI
f  Measurement at 200n  RSL during spiral.
g  Measurement at urban site.

-------
were considered when interpreting the times and distances  to maximum 03 given
in Table 5.  For June 24, July 15, 16, 17,  and August 6 and 8,  the first site
listed is upwind, but in fact measured the  highest concentration in the Boston
area.  These values do not reflect the potential  for  63 generation by emissions
in Boston.  The second site is a coastal  location.  The 03 concentrations
listed for this site were measured during an onshore  flow  and  thus provide a
lower limit estimate of 63 formed in the  Boston plume (higher  concentrations
may have occurred offshore).   On August 1 and 5 transport  from upwind urban
areas may have had a substantial contribution to maximum 63 at surface sites.
Thus, in the Boston area, aircraft data provide a more appropriate estimate
of 03 formation in the urban  plume.  The  aircraft data in  Table 5 show that
the downwind distance to maximum 03 in the Boston plume ranged from 61 to
107 km, with an average distance of 81 km.   The mean  transport time to peak
03 was 4 hours.
            The times and distances to 03 maxima in the urban  plumes of New
York City and Boston are useful pieces of information for  the  development of
control strategies, as well as for investigators who  develop and test urban
models for these cities.  Additional information of particular usefulness to
modelers relates to the transport time and distance to  peak  NOg.  The concen-
tration of nitrogen oxides is especially sensitive to monitor  locations,
because NO and N02 are primary pollutants (i.e., emitted  directly to  the
atmosphere).  Nitrogen dioxide is also a  secondary pollutant formed in the
series of  reactions leading to 03 formation.   In and around  major urban areas,
the  peak N02 concentration is usually observed near or shortly downwind of  the
urban center, because surface monitoring stations are strongly influenced  by
                                        22

-------
                     Table 5.   Times  and  Distances  from  Boston  to Maximum  Ozone
Date
6/24/80
7/15/80
7/16/80
ro 7/17/80
Co
8/1/80
8/5/80
8/6/80
8/8/80
03 Max.,
ppb
Site (Ground)
Medffeld, MA
Cape Elizabeth, ME
Worcester, MA
Cape Elizabeth, ME
Easton, MA
Cape Elizabeth, ME
Easton, MA
Cape Elizabeth, ME
Georgetown, MA
Gardiner, ME
Medfleld, MA
Tewksbury, MA
Easton, MA
Cape Elizabeth, ME
154
097
193
145
127
127
150
143
143
143
159
115
120
112
Time,
EST
2100
1500
1500
1400
1500
1400
1700
1700
1100
1700
1400
1300
2000
1200
Distance
km
(a)
150
(a)
150
(a)
150
(a)
150
42b
(d)
31
(a)
150
Transport
Time,
hours
(a)
7-8
(a)
4-5
(a)
6
(a)
5
2b
lib
(d)
2
(a)
5-6
03 Max.
PPb,
Aloftc
—
208
170
200
197
151
133
1706
232
173
156
Time.
EST
	
1538
1400
1700
1542
1648
1554
1525
1548
1448
1449
Distance,
km
—
(a)
65
(a)
107
(a)
61
104
66
68
93
Direction
from
City
--
SW
NE
SU
NE
SU
NE
NE
NE
NE
E
Transport
Time, Location
hours
...
(a)
2
(a)
4
(a)
4
6
5
4
3-4

near RI border
over Atlantic Ocean
near RI border
over Atlantic Ocean
near RI border
over Atlantic Ocean
over Atlantic Ocean
over Atlantic Ocean
over Atlantic Ocean
over Atlantic Ocean
*  Maximum Oj occurred upwind from Boston and reflects daytime transport  from upwind urban areas.
°  Overnight transport aloft from upwind sources may have resulted In  observed 63 max.
c  Measurement at 700-800m MSL unless otherwise noted.
d  Convergent air flow west of Boston; peak may have contribution from Boston and Providence.
e  May not be max. 03; measurement listed was obtained on furthest downwind traverse of plume.

-------
nearby emission sources.  This phenomenon was demonstrated earlier in Tables
2 and 3, which show that the maximum surface N02 concentration  is nearly
always found within or on the downwind fringe of the urban area.  In order to
determine the time and distance required to generate peak  NOg  by chemical
reaction, it is more appropriate to employ the aircraft monitoring data.
These data are much less sensitive to local surface emissions  and, conse-
quently, are more representative of the chemical dynamics  of photochemical
air pollution.  However, these data suffer from a lack of  measurements close
in to the urban area on several days.  Table 6 lists the time,  distance and
transport time to maximum N02 concentration aloft in the afternoon for the
New York study area.  Nbximum N02 concentrations aloft rangeu  from 30 to 87
ppb, and the downwind distance to peak N02 aloft ranged from 23 to 48 km.
Transport times ranged from 1 to 3 hours.  In addition, on July 31 and August
6 relatively high N02 concentrations were measured far downwind (33  ppb at
102 km and 68 ppb at 125 km, respectively).  In both cases, high  levels of 03
were also present, and it is likely that a portion of the  measured N02 was
actually present in the form of nitric acid and peroxyacetyl nitrate.
            The transport times and distances to maximum N02 aloft in the
Boston area during the afternoon are also  shown in Table 6.  Peak N02 concen-
trations aloft were in the range from 17 to 49 ppb.  The distances to peak
N02 were in the range of 34 to 110 km, with transport times of 1  to  3 hours.

     4.1.4  Downwind Distance to Background NOX
            The purpose of this task was to determine the downwind distance
from  the city  to where the NOX concentration in the urban  plume becomes  diluted
                                     24

-------
Table 6.  Times and Distances to Maximum NCL Aloft in the
          New York City and Boston Urbam Plames
Date
NEW YORK
7/18/80
7/22/80
7/24/80
7/31/80
8/6/80
8/8/80
BOSTON
7/15/80
7/16/80
7/17/80
8/1/80
8/5/80
8/6/80
8/8/80
Time,
EST
1153
1205
1211
1204
1220
1218
1339 .
1543
1459
1500
1401
1425
1335
N02 Maximum,
ppb
87
76
34
30
76
75
20
19
17
23
49
23
18
Distance from
City, km
32
37
23
40
48
41
34
110
62
74
36
41
53
Transport Time,
hours
4.0
2.0
1.0
3.0
2.0
1.5
1.0
3.0
2.0
3.0
2.0
2.0
1.5
                             25

-------
to the point where it approaches the observed air mass background  NOX
concentration.  Unfortunately; there was insufficient spatial  coverage  of
surface NOX sites to permit assessment of this distance.   However,  aircraft
data are better suited to this purpose and were used in this task.   It  is
important, for purposes of this task, that fresh emissions of NOX  into  the
plume downwind of the city be excluded; otherwise, the NOX concentration
could actually increase with downwind distance.  This criterion  was met by
using flights which pass primarily over the ocean, where such emissions are
minimal.  Two afternoon flights in the New York area arid three in  Boston met
all of the criteria for this task and were used in the analysis.  The  urban
plume NOX concentrations from aircraft plume traverses were plotted versus
downwind distance, and a straight line fitted through the points and extra-
polated to the air mass background NOX concentration observed outside  the
urban plumes.  The results of this analysis are shown in Table 7.
            The distance at which plume NOX reaches the background concentrations
(i.e., is indistinguishable from air mass background by the measurement method)
is obviously  dependent on a number of variables,  foremost of which are dilution
rate, reaction rates and wind speed.  No direct measure of dilution or reaction
rate  is available, but the afternoon boundary layer transport speeds estimated
from  the  trajectories are included in the table.'  As expected, the highest
transport speeds are associated with the furthest distance to background  NOX.
The  travel time  for NOX  to reach air mass background concentrations ranged
from  5.5  to 12 hours, and the estimated travel distance was 85 to 165 km.

      4.1.5  Plumes  from  Medium Size Cities Between New York City and Boston
            Clearly defined 03 plumes have frequently been observed downwind
of major  centers of population and  industry  on photochemically active days.
                                      26

-------
Table 7.   Distance at Which NOX  in Plume Approaches Background  Concentration
        Date
City
            Estimated Distance
Flight          to Background
Number            NOV, km
                Transport
                  Speed
                 km-hr"1
        7/18/80
        7/24/80
        7/24/80
        8/5/80
        8/6/80
New York
New York
Boston
Boston
Boston
   5
  15
  18
  31
  32
 85
165
100
 95
100
 7.0
20.0
18.2
12.5
18.0
                                             27

-------
The contribution of medium size cities to the  downwind 03  burden has been
studied infrequently.  Several medium size cities  are sources of 03 precursors
in the Northeast Corridor, but observation of  03 plumes  from these cities is
likely to be difficult, due to the generally high  background levels of 63 in
the region.  Nevertheless, important insight into  03  formation and transport
processes in the region may be gained by identification  of plumes from smaller
cities, and comparison of their plume characteristics with those from New
York City and Boston.
            It was difficult to define plumes  from small cities using surface
data, because the number and distribution of monitoring  stations was nearly
always insufficient for this purpose.  Aircraft data  are mos^ appropriate for
this task, since complete upwind and downwind  traverses  can  be made over a
relatively short time.  Although a considerable number of  aircraft flights
were conducted during  the monitoring program,  no flights were designed spe-
cifically for small city plume mapping.  As a  consequence, this analysis is
limited to those flights in which small city plumes were observed fortuitously.
            The identification of 03 plumes from smaller cities in the study
region focused on Bridgeport, New Haven, Hartford, and Providence.  Because
no data were collected to address the issue of small  city  03  plumes specfi-
cally, all surface and aircraft data collected during NECRMP, were surveyed
without limiting the analysis to the 20 study  days.  From  this examination two
possible cases of  the  Providence plume were identified  in  the data.

     Case  1 —
     There  is an indication of an 03 plume  from Providence,  on July 18,  1980.
An  examination of  surface 03  concentrations indicates that the 03  level  at  the
                                       28

-------
Kent County site is relatively  high compared to surrounding sites.  The time
series of 03 concentration  and  wind direction at Kent County is shown in
Table 8.  At this site,  03  reached 95  ppb at 1800 EST, while the concentration
remained less than 80  ppb at the  surrounding stations (Attleboro, Providence,
Hartford, and Middletown).   Airflow on July 18 was  generally from the north-
erly quadrant early in the  day.  However, the wind  data  in Table 8 indicate a
shift to southerly flow after 1600 EST.  These data suggest a  recirculation
of air during the day, such that  Providence emissions were first transported
offshore then back onshore  toward the  Kent County  site.   This  indicates that
the Providence plume was the primary cause of the  elevated 03  levels at the
Kent County site.  However, it could be argued that Boston area emissions
provided a contribution even though  the mesoscale  trajectories for July 18
indicate that Boston's plume passed  south of Cape  Cod during the day.
     If we assume that a plume from  Providence caused  the peak values at Kent
County, then the increase in 03 above  the air mass concentration was 30 ppb,
as judged by the concentrations at surrounding  locations.

     Case 2 ---
     The second, and perhaps more definitive, case of  the Providence  plume
is evident on July 24, 1980.  On the afternoon  of this  day,  the Boston  plume
was  tracked to  the south via aircraft.  The  zig-zag flight track  is  shown  as
a solid line on a map of the area in Figure  4.   The 03  concentration along
traverses C-D and E-F is also plotted  on the map.   At  the western-most  edge
of the  traverse E-F, there is a peak in 03 which appears as  a  shoulder  on  the
major peak representing the Boston urban plume.   The Boston  plume  is  also
                                      29

-------
Table 8.   Ozone  and Wind Direction  at Kent County, RI  on July 18,  1980
                                             Hind Direction.
                   Time. EST     Ozone, ppb	Degrees
                      0800          45              340
                      0900          50              345
                      1000          58              350
                      1100          60              345
                      1200          60               50
                      1300          60              315
                      1400          60              335
                      1500          62               35
                      1600          85              100
                      1700          90              170
                      1800          95              210
                                          30

-------
   o,.p*
Figure 4.  Flight track for Boston on the afternoon  of  July 24,  1980.
                                   31

-------
seen centered along traverse C-D but, 1n this case,  there  is  no  shoulder at
the western edge.  The two traverses show that 63 concentrations upwind of
Providence, and outside the Boston plume, are 55 to  60  ppb, while downwind
in the Providence plume 03 is 75 ppb.  These concentrations are  replotted
in Figure 5 as shaded contours.  This diagram shows  even more clearly  the
distinction between the Boston and Providence plumes.
     The mesoscale trajectories for 1000 EST on July 24, also shown  in
Figure 5, indicate the position of the Boston plume  at  1400 EST. Comparison
of the ozone plume with the trajectories suggests that  the Boston plume center
line and the trajectory track are in exact agreement.   If  ve  assume  a  parallel
trajectory from Providence, the plume from this city would be expected in pre-
cisely the location where the shoulder was observed. The  estimated  03 for the
Providence plume on July 24 was 15 to 20 ppb above the  upwind concentration.
     Examination of the NECRMP data did not reveal plumes  from Hartford, New
Haven, and Bridgeport, CT.  The technical problems of identifying such plumes
are compounded by the proximity of cities in this area  and the complex meteor-
ology introduced by the land/sea interface.  The increase  in  03  in  the two
instances when a Providence plume was identified ranged from  15  to 30  ppb.
This range can be compared with the increase in 03 observed  in the  plume of
a city of  slightly smaller size in the Midwest.  A study of  the  Springfield,
IL plume on a photochemically active day in the summer  of 1977 revealed an
increase in 03 of 30  ppb above  the air mass 03 concentration.1^   Thus, the
contribution  to  the downwind 03 burden of these two moderate  size cities was
similar  under these particular  meteorological conditions.
                                      32

-------
Figure 5.   Ozone  distribution  aloft downwind of Boston
           on  the afternoon  of July 24,  1980.
                      33

-------
4.2  Pollutant Transport
     The objectives of this task were:  (1)  to quantify  the  concentrations of
03 and its precursors transported into and  across the New York City and Boston
areas on the case study days; and (2) to identify the transport  direction on
each day and infer the probable source area of 03 and precursors outside the
New York City and Boston urban plumes.   These objectives are  addressed
concurrently in this section.
     For both New York City and Boston, surface and aircraft  data were used
to determine the concentrations of pollutants transported into and across the
urban area during each case study day.   Ozone measurements  downwind of the
urban area, but outside the urban plume were also examined  to evaluate the
homogeneity of air mass 03 concentrations transported across  the region.  The
probable source area for the observed transported concentrations was  inferred
by interpretation of synoptic and mesoscale trajectories, and the surface and
upper air wind data.

     4.2.1  Analysis Procedure
            Analysis of pollutant concentrations upwind and downwind  but
outside the urban plume of New York City was conducted  for  the case  study
days:  June 24, 25; July 16, 18, 21, 22, 24, 31; and August 6, 8, 26  and 28.
In Boston,  the analysis was conducted  for June 24; July 15, 16,  17; and
August  1, 5, 6, and 8.  For these days, pollutant concentrations were examined
for  three specific  time periods  in order that temporal  patterns  of  pollutant
transport might be  observed:  morning  (0600-1000 EST);  mid-day  (1200-1600  EST);
and  evening  (1800-2200  EST).  First, surface winds and the  trajectories were
interpreted  to determine a boundary  layer transport wind direction  for  each
time period.  The  transport wind directions were used to identify  surface
                                     34

-------
sites upwind or  downwind outside the urban plume.  For these sites, average
concentrations of 63,  NO,  N02, and NMOC were computed for each of the three
time periods. Data  were not averaged  for the entire 4-hour period when
large temporal concentration gradients were observed.  Rather, an average
range was computed for the time period.
            The  concentrations of 63,  NO, and N02 aloft were estimated from
aircraft spiral  and  transect data collected during flights in the vicinity
of each city.  Upwind/downwind directions were  determined from trajectories
and upper air winds.  The  spirals were divided  into  two or three vertical
segments depending on  concentration  gradients,  and layer average concentra-
tions were calculated  for  each pollutant.  Since no  afternoon upwind aircraft
spirals were made in the New York area, measurements made during horizontal
flights over northeastern  New Jersey were used  for estimating upwind concen-
trations aloft.   The concentrations  of 03, NO,  and N02 aloft downwind, but
outside the urban plume, were obtained by computing  a  spatial average con-
centration for each pollutant from  portions of  downwind transects  that were
outside the plume.

     4.2.2  Discussion of  Ozone and  Precursor Transport
            This analysis  was  to  have grouped each case study day  into one of
three distinct flow regimes:  Corridor transport;  transport  from outside  the
Corridor; and near-stagnation/recirculation.  The  results, as summarized  in
in Table 9,  indicate that more  than  one  type  of flow regime  occurred on many
of the days.  Often, the direction  of transport into New  York City and
Boston varied by  time of day and/or altitude.   For example,  on  four days,
                                     35

-------
               Table 9.  Transport Regimes on Case Study Days
Urban Area
New York City
Boston
  Day

6/24/80
6/25/80
7/16/80*
                       7/18/80
                       7/21/80*
                       7/22/80'
                       7/24/80
                       7/31/80*
                       8/6/80
                       8/8/80*
8/26/80*
8/28/80*

6/24/80*


7/15/80*



7/17/80*


8/1/80
8/5/80*
8/6/80
8/8/80*
      Flow Regime

Along-Corridor Transport
Along-Corridor Transport
Along-Corridor Transport, except
morning transport aloft from
outside the Corridor

Non-Corridor Transport
Along-Corridor Transport, except
morning transport aloft from
outside the Corridor

Along-Corridor Transport, except
morning transport aloft from
outside the Corridor

Non-Corridor Transport
Non-Corridor Transport/
Recirculation Flow

Non-Corridor Transport
Along-Corridor Transport, except
morning transport aloft from
outside the corridor

Near-Stagnation Conditions
Near-Stagnation Conditions

Morning:  Non-Corridor Transport
Afternoon:  Along-Corridor Transport

Along-Corridor Transport, except
morning transport aloft from
outside the Corridor

Morning:  Non-Corridor Transport
Afternoon:  Along-Corridor Transport

Along-Corridor Transport
Recirculation Flow
Along-Corridor Transport
Morning:  Non-Corridor Transport
Afternoon:  Along-Corridor Transport
 ''Flow regime  varied with height and/or  time  on  these  days.
                                      36

-------
overnight transport into  the  New  York area at  the  surface and aloft up to
several hundred meters, was from  along  the Corridor.   However, at higher
altitudes up to 1500m transport was  from  the west,  beyond the Corridor.  On
such days, a plume of high N02  and NMOC concentrations was  observed during
early morning in the layer of along-Corridor flow  with much lower concentrations
aloft in the layer transported  from  outside  the Corridor.   As the daytime
boundary layer grew, pollutants from both layers were  eventually entrained,
and became mixed with precursors  from surface  emissions.  Thus, on such days
pollutants transported from source areas  both  within and  beyond the Corridor
participated in 63 formation  in the  New York area.
            The estimated concentrations  of  03, NO, N02,  and NMOC transported
into and across the New York and Boston areas  are  summarized by flow  regime
in Table 10.  The vertical partitioning of pollutant concentrations aloft  into
different flow regimes was typically made at an altitude  between 500  and 1000m.
The partitioning on a particular day was  based upon the vertical variation  in
wind direction, vertical  gradients in pollutant concentrations, and  the  track
of layer-averaged trajectories.  Measurements  from sites  affected  by  recircu-
lation  flows (in which pollutants are transported back over areas  that previ-
ously  had been upwind), and days with near-stagnation  conditions  (when upwind/
downwind areas are difficult to define) are not included  in the  table.  Also
excluded are (1) morning 03 concentrations at the surface which may  have  been
affected  by variations in local scavenging affects, and (2) mid-day  precursor
concentrations which were low compared to early morning concentrations.
            For New York City, the data indicate that morning precursor
concentrations at  the surface resulting from along-Corridor transport were
about  twice the concentration on days with transport  from outside  the Corridor.
                                      37

-------
Aloft, precursor concentrations, particularly NMOC,  were also much higher in
layers transported along the Corridor.  These precursors appear  to be the
result of overnight emissions in the Philadelphia  area  or  other  portions of
the Corridor upwind of New York City.  Fairly large  spatial and  temporal var-
ations in morning NO and N02 concentrations were observed  among  monitoring
sites upwind of New York on most days, as indicated  by  the range in average
concentrations for these pollutants given in Table 10.  This likely reflects
the combination of several factors including variations in the dispersion and
transport of overnight emissions within and just above  the nocturnal surface-
based stable layer.  (Spatial variations in NMOC could  not be assessed  since
only one NMOC site was operated upwind of New York.)
            Mid-day 03 concentrations transported into  New York  City with
along-Corridor flow were 15 to 30 ppb higher at the  "close-in" upwind sites
(i.e., sites over northern and central New Jersey 60 to 90 km  from mid-
Manhattan) than at other sites further upwind (beyond -100km).   The elevated
03 concentrations at these "close-in" sites, are likely  the result of emissions
in suburban areas on the fringe of the main Corridor cities.   Also, on  Corridor
transport days, the direct impact of  the Philadelphia 03  plume was typically
observed during mid-afternoon at one  or more of these "close-in" upwind sites.
However, as the Philadelphia plume traveled across the New York  urban area
surface 03 concentrations declined due to reaction with fresh  NO emissions.
            In Boston, there was little difference in the magnitude of  surface
precursors during Corridor and non-Corridor transport regimes.   Aloft,  however,
the concentration of NMOC during Corridor transport was three  times  the magnitude
of NMOC when  transport was  from outside the Corridor.  Also,  incoming 03 aloft
                                      38

-------
            Table  10.   Average  Transported Ozone and Precursors  for  Corridor and
                          Non-Corridor  Flow Regimes
Morning Precursors, ppb
Transport Regime
NO
Surface3
NO 2
NMOC,
ppbC
Aloft
NO N02
NMOC,
ppbC
Ozone, ppb
Morning
Upwind
Aloft '
Mid- day
Outside Urban Plume
Upwind Downwind
New York City
               LDd-021   014-034
             567
       Corridor Transport

         LD-006     043     285
            089-116   076-089
Boston
                LD-013    LD-019
                LD-009    LD-009
                Transport from Beyond Corridor

             278        LD     LD-013    048
             219
       Corridor Transport

           LD        013     108
                                        070 b
100
            074-084"
085-098
                                                                                                        .  c
                   009
LD-008
  Transport from Beyond Corridor

285        LD     LD-012     036
076
a  0600-1000 EST

b  Excludes measurements on July 24, 1980 which were unrepresentatively low compared to data for the  other six days
   in this category.  Morning upwind 03 aloft was 035 ppb and mid-day 03 was 040  - 049 ppb.  Trajectories indicate
   48 hour transport from Southeast Canada on this day.

c  Only two days in this transport category with mid-day downwind data:  July 24  with 03 of 040 - 050 ppb, and
   August 6 with 03 of 075 - 085 ppb.

d  Concentration at or below the lower detectable limit of the monitor.
                                                         39

-------
during the morning on Corridor transport days averaged 100  ppb, compared to
76 ppb when transport was from outside the Corridor.   It is interesting to
note that the average concentrations of surface NMOC,  and the average concen-
tration of 03, NO, N02, and NMOC aloft are of similar  magnitude upwind of New
York City and Boston during transport from areas outside the Corridor.  This
suggests regional homogeneity in 03 and precursors on  the days when  transport
was not along the Corridor.
            The impact of the New York City 03 plume on the Boston area was
observed at sites upwind of Boston and/or aloft during the  late afternoon or
evening on most Corridor transport days.  The 1-hour maximum surface 03 con-
centration upwind of Boston associated with boundary layer  urban  plume trans-
port ranged from 120 ppb (August 8) to 193 ppb (July 15).  Aloft,  transported
03 during the afternoon was as high as 200 ppb near the Massachusetts - Rhode
Island border on July 16.
            An example of daytime 03 transport from the New York  area to Boston
is shown in Figure 6.  The location of the New York City 03 plume is clearly
identified by the isopleths of surface 03 measurements.  The analysis indicates
a rapid rise in 03 between 1000 EST and 1100 EST along a 75 km wide  band from
the downwind edge of New York City across Long Island  Sound and coastal
Connecticut.  By 1400 EST the plume extended into Rhode Island with  the  highest
concentrations at ~100 km downwind of New York City.   During the  next several
hours  the area of high ozone within the plume progressed northeastward  toward  the
Boston area.  The highest concentrations at the surface decreased rapidly  between
1600 EST and 1800 EST, but then remained  fairly stable at -140 ppb through
2200 EST.  On this day,  transport appears to have been responsible for  the  maximum
03 concentration measured in  the Boston area (154 ppb/2100 EST at Medfield, MA).
                                       40

-------
       
-------
                          Tf*  Trio*
                              LWDITUX:
Figure 6 (continued).  Surface ozone concentration  isopleths  (ppb),
            0800 through 2200 EST, on June  24,  1980.
                                42

-------
4.3  Temporal  Changes in  Ozone  Above  the  Boundary Layer
     The purpose of this  task was to  address the  following  question:
To what extent do 03 concentrations aloft increase between  sunrise  and  mid-
morning when 03 aloft is  entrained rapidly into the boundary layer  during
the dissipation of the nocturnal  inversion?
     Daytime 03 concentrations  within the boundary layer occur due  to a
combination of physical and chemical  processes, including photochemical
production, scavenging, transport, and entrainment from  aloft.  One of the
key parameters for modeling boundary  layer 03 is the magnitude of 03 aloft
available for entrainment as the boundary layer grows in response to daytime
thermal convection.  In many cases, a lack of extensive  measurements requires
that assumptions be made regarding the temporal behavior of 03 entrained into
the boundary layer.  It was the purpose of this task to  investigate the tem-
poral variation in 03 concentrations aloft outside the boundary layer.  Of
interest was whether substantial production of 03 aloft occurs during the day,
as a result of reactions among transported 03 and aged precursors.   Because
the monitoring program did not include experiments to specifically address
this issue, the following analysis is restricted to using measurements obtained
for other purposes.  As a result, only a cursory investigation of this topic
is possible, given  the available data.
     The analysis procedure was designed to identify groups of aircraft
spirals which represent quasi-Lagrangian measurements aloft during the day.
The temporal variation of 03 aloft on a particular day was determined  by com-
paring measurements from these selected spirals.  Evaluating 03 concentrations
                                        43

-------
in a quasi-Lagrangian manner, rather than  comparing  time changes in
concentrations at a particular location,  avoids  the  complicating influence
of variations in 03 aloft due to changes  in  transport.
     All selected groups of spirals containing quasi-Lagrangian measurements
in an air parcel were pairs of early morning (0500/0600 EST)  upwind and mid-
morning (0900/1000 EST) downwind spirals  in  Boston.   For the  spirals in each
group, temperature, dew point temerature,  03, bscat,  and NOX  measurements
were used in defining the vertical  bounds of specific layers  which were iso-
lated from the boundary layer during all  spirals in  the group.   In most cases,
the spirals resembled one of the six characteristic  profile types identified
by Ludwig ^ and shown in Figure 7.  For types "a" and "b," 0^ concentra-
tions were averaged from above the layer of depletion near the surface to
the top of the spiral or an upper level  inversion.   For types "c" and "d," 03
concentrations were averaged vertically,  beginning above the  large gradient at
the top of the boundary layer to the top of the  spiral  or  an  upper level inver-
sion.  For types "e" and "f" (and other patterns with large 03 gradients well
above the boundary layer), the spirals were divided  into several layers and
the 03 data averaged for each layer separately.   The transition  zones of large
03 gradient were excluded from layer averages.  In most cases, vertical bounds
of  the selected layers were associated with changes  in the atmospheric stabil-
ity.  Discontinuities in the temperature/dew point temperature profiles which
marked a change in stability were usually identifiable on  successive  spirals
so  that continuity of the layers was preserved.   However,  in  some cases, the
top of  the  spiral  had  to be  used as the upper bounds for computing vertical
averages when no  stability discontinuity or pollutant gradient was observed
near  the maximum  altitude of the spiral.  Tracking 03 concentrations  within
                                      44

-------
3


2


1
E
.*
1 0
1- (
X
o _
— 3
iu •»
X
2
1
n
r i
1 a.
1
1
i
1
I
1
j
ii/ii
) 40 80 C




b.
_
;
^K-X| 1 1
-
C.

.

mmmzmmmm

?. MIXED LAYER YXX XX
XiXviXYXivXv^v/xiXYXi^XiXvX:!
) 40 80 120



9
\ d°
1
vivi-i^vi-x-Xvivi-x-x-x-:-:-:-:-:-:-:-:-::!:-:-:-:-:-:-
'£. MJ.X. E P. . L A Y E R . xJXxX::;
•x-x-:i-x-x-iix-x-:fx-x-t:-:v:4xii?:iv:-:vi
                                                            e.
                                                                  1
                                                 '£&*f???-:-SJ ABLE LA YE R
                                                     40    80   120    160
                                                             f.
XMIXEDX.
:::ii:::::::::i:x:;:::i^:
       40    80     0     40    80   120        0    40
                       OZONE CONCENTRATION  - ppb
                                                                        1
         80    120   160
Figure 7.   Characteristic ozone profile types identified by Ludwig.
                                                                  11
                                    45

-------
'J i •.(. r<-Uj, fairly homogeneous layers defined by stability discontinuities
minimized the probability that other processes, such as vertical  dilution  and
entrainment from the boundary layer, significantly affected 03 concentrations
aloft.  Thus, any changes are most likely due to photochemical processes  among
the pollutants aloft.
     As a result of following the above procedures, thirteen cases were
identified for evaluating the formation of 63 aloft above the boundary  layer.
The dates, times, dimensions, and 03 concentrations in these layers are pro-
vided in Table 11.  Also provided are the profile type, and the observed
change in layer average 03 (residual).  It is evident from the data that  the
analyses included a wide range in initial 03 concentrations (?6 to 99 ppb).
Yet, only four cases (I, III, IV, and XI) stand out as having a measurable
increase in 03 concentration.  In the remaining cases, 63 varied by H^ 5 ppb
or less, which was within +_ 10% of the initial concentration.  Variations in
this range could be due to uncertainties in the measurement system, rather
than real fluctuations in 63 concentration.
     In order to investigate why 03 increased in certain cases but remained
unchanged in others, NOX data were examined to see if the layers with appar-
ent 03 formation had higher NOX levels.  Although no NMOC measurements were
available, bscat data and back trajectories were used to infer whether the
layers may have received precursors from major anthropogenic  source areas
during the previous  two days.  In addition, observations of sky cover and
temperature were examined from the Boston National Weather Service station
to determine whether local meteorological conditions may have hindered or
enhanced 03 formation.  However, neither variations in  temperature nor cloud
cover could not explain the differences  in 03 concentrations  aloft.
                                     46

-------
Table 11.   Temporal  Variations in Ozone Aloft

City:
I
II
III
IV
V

VI
VII
VIII
IX
X
XI
Date
Boston
7/15/80
7/16/80
8/1/80
8/8/80
8/14/80

7/15/80
7/16/80
7/17/80
7/24/80
7/25/80
8/1/80
Time,
EST
0545
0915
0600
0945
0500
0930
0530
0945
0530
0850

0545
0915
0600
0945
0600
0945
0530
0930
0530
0930
0500
0930
Flight/
Spiral #
4-1
5-1
7-1
8-1
26-1
27-3
33-1
34-1
39-1
40-1

4-1
5-1
7-1
8-1
10-1
11-1
16-1
17-1
19-1
20-1
26-1
27-3
03 Profile Dimension of
Type Layer, m
(Ludwig) Bottom Top
e
e
b
c
e
f
b
c
b
a

e
e
b
c
b
a
a
a
b
a
e
f
Lower Layers
485
500
600
600
300
450
450
450
650
650
Upper Layers
1700
1660
1850
1840
1000
1000
1000
1000
1300
1300
1100
1170
1600
1500
1500
1500
800
800
• 1350
1350
1500
1500

2040
2050
2120
2100
1400
1400
2000
2000
1900
1900
1500
1520
03 Concentration (ppb)
Layer 403 of Layer
Average Peak Average
094
111
093
095
086
124
071
092
039
036

068
067
078
076
079
081
026
024
046
043
048
062
109
124
101
102
105
141
078
095
044
041

072
071
079
078
081
086
032
033
048
045
050
064
-K)17
+00 2
+038
+021
-003

-001
+002
+002
-002
-003
+014

-------
     Examination of the NOX data indicates that the four comparisons  (Cases  I,
III, IV, and XI) in which 03 increased had higher NOX  concentrations  and  bscat
than the other cases.  Also, these four cases were associated  with  trajectories
that either had traveled along the Corridor or across  portions of the Midwest
within 48 hours prior to reaching Boston.  The largest residual  ( +  38 ppb;
Case III) occurred in a layer which the trajectories and surface 63 data  indi-
cate contained the remains of the New York City plume  from the previous day.
The concentration of NOX (mostly NC^) was 20 ppb and b$cat was 2.18 x 103 nf1
within this layer.  In Case IV, NOX and bscat were also comparatively high
(10 ppb and 2.0-3.00 x 103 nrl, respectively).  The 48-hour trajectory path
from northern Indiana and Ohio, eastward across Massachusetts, suggests a
contribution from the upper Midwest to pollutant levels aloft.  In  Case I, the
layer had arrived in the Boston area after passing along the Corridor, but to
the west of New York City, NOX was 7 to 8 ppb and bscat was 1.00 x  103 nT1.
In  the final case with an increase in 03 (Case XI), 03 was initially  fairly
low (48 ppb), and bscat was also low at 1.00 x 103 m"*.  However, the NOX con-
centration was relatively high at 14 ppb.  The trajectories indicate  that this
layer had been part of the daytime bounday layer over central  and eastern
New York State the day before.
     Of the seven cases with no change in 03, all had NOX levels of 7 ppb or
less, with an average concentration of 4 ppb.  Except for Case II,  the mag-
nitude of bscat was  less than  1.00 x 103 m~* and averaged 0.51 x 103  m"*,
indicating that the  aerosol content tended to be much lower than in the cases
when 03  increased.
     The results  of  this analysis  indicate that 03 production aloft was
apparent in  those layers containing  comparatively  high  air mass concentrations
                                      48

-------
of NOX and perhaps  other  03  precursors  (as  indicated  by  the aerosol content
and history of the  layer  relative  to  urban  areas).   It is  likely  that  these
layers were within  the daytime  boundary layer  on  preceding days and, thus,
include the by-products of photochemical  processes  from  boundary  layer emis-
sions.  However, 63 levels were fairly  stable  from  sunrise through mid-morning
in cases in which NOX was comparatively low, and  trajectories  did not  indicate
transport across areas of major anthropogenic  emissions  within 24 to 48 hours.

4.4  Comparison of Ozone  Levels at the  Surface and  Aloft
     The purpose of this  task is to address two questions:
     1.  Do mid-morning surface 03 concentrations during the  dissipation  of
the nocturnal inversion reflect average 63  concentrations  aloft  prior  to
mixing?  That is, can mid-morning surface 03 measurements  be  used to estimate
overnight 03 transport aloft?
     2.  How well do afternoon  03 concentrations aloft  compare with  surface
03 concentrations at nearby stations?  That is, can aircraft  measurements be
used to infer surface 63  concentrations in  areas without 63 monitoring sites?
     Analyses conducted in the  previous tasks  indicate  that a reservoir of air
with elevated 03 concentrations exists above  the nocturnal  inversion  during
many nights.  This layer of 03  has been separated by the inversion  layer from
deposition at the surface and chemical  scavenging by pollutants  emitted near
the surface.  Also, air in the layer containing elevated 03  aloft is  subject
to transport over long distances overnight by winds above  the inversion layer.
As the mixing height rises in the morning,  due to surface  heating,  the 03 aloft
is entrained into the  boundary layer and contributes to concentrations at the
surface.
                                     49

-------
     4.4.1  Comparison of Morning 03 Concentrations at the Surface  and  Aloft
            In order to compare early morning levels of 03 aloft with surface
concentrations after inversion breakup, aircraft and surface 03 data  were
tabulated for the New York and Boston areas as shown in Table 12.  The  table
contains three main sections.  The first section containing five columns,  pro-
vides the date, spiral number, time, layer thickness, and average 63  concentra-
tion aloft.  The second section of three columns lists the three-hour average
surface 03 concentration, centered around the estimated time of inversion
breakup, for ground stations near the spiral site.  The final three columns
list the time of inversion breakup, the 03 concentration aloft after inversion
breakup, and the time of these measurements.  The inversion breakup time was
determined from sodar data and/or vertical temperature profiles.  On  days
when such data were unavailable, the median time of inversion breakup of 0930
EST was used.
            All of the aircraft and surface data were taken from upwind
locations in order to avoid  the confounding effects of 03  formation in the
urban plume.  The early morning average 03 concentrations  aloft were computed
for the layer above the surface layer and  below 1500m (1500m was used as an
approximation for the upper  limit of mixing during late morning).  The late
morning 03 levels aloft in the New York area Were obtained  upwind of the city,
shortly after the aircraft departed  for the afternoon flight.   In Boston, the
late morning  data are layer  averages  through the  boundary  layer obtained from
spirals upwind at the completion of  the mid-morning  flight.  Two of the
Boston  study  days, August 5  and 6, were not  included  in  this task.  August 5
was excluded  because  surface monitoring sites  in  the  vicinity of aircraft
                                       50

-------
Table  12.    Comparison  Of  Mid-morning Surface  Ozone  With  Early  Morning  Ozone  Aloft
                       Early Homing 03 Aloft
Mid-morning Surface 03
1
Spiral
Date Number
1 1
Time, Altitude, 03, 03. Averaging3
EST Meters, HSL ppb ppb Time, EST
Time of Late
Inversion Morning
Location Dissipation, EST 03 aloft Time
New York Area
I 7/16/80 1
II 7/18/80 1
III 7/21/80 1
IV 7/22/80 1
V 7/24/80 1
VI 7/31/80 1
VII 8/6/80 1
VIII 8/8/80 1

0700 600-1500 085 082
073 0900-1100
087
0545 400-1500 119 062
069 0900-1100
0545 500-1500 076 079
114 0900-1100
110
0530 650-1500 056 051
053 0800-1000
053
0530 400-1500 035 037 0800-1000
042
0545 400-1500 060 059 0800-1000
059
0530 400-1500 071 069
076 1000-1200
061
0530 400-1500 067 097
107 1000-1200
088
Boston Area
Fleming ton
Marlboro 1000
Chester
Flemington
Chester 1000
Flemington
Marlboro 1000
Chester
Flemington
Marlboro 0900
Chester
Flemington 0900
Chester
Flemington 0830
Chester
Flemington
Marlboro 1030
Chester
Flemington
Marlboro 1030
Chester

-
065-070 1130
-
065-070 1130
045 1200
075-085 1130
065-075 1130
090-100 1130

       IX   7/15/80    1     0545    485-1500      095    103    1000-1200   Easton
                                                       104               Attleboro
                                                       100
                                                                         Sudbury
       X    7/16/80    1     0600    600-1500      093    105    1000-1200   Easton
                                                       088               Medfleld
                                                       084
                                                                         Sudbury
       XI   7/17/80    1     0600    500-1500      079    073    0900-1100   Easton
                                                       057               Medfleld
                                                       054
                                                                         Sudbury
       XII  7/25/80    1     0530    500-1500      049    054    0800-1000   Medfleld
                                                       050               Attleboro
                                                       045               Sudbury

       XIII 7/31/80    1     0530    500-1500      048    050    0900-1100   Hedfield
                                                       048               Easton
                                                       051               Georgetown

       XIV  8/1/80     1     0500    200-1500      067    100    0900-1100   Easton
                                                       131               Medfield
                                                       098
                                                                         Sudbury
       XV   8/8/80     1     0530    450-1500     071
                                                              0900-1100   Easton
                                                       087               Medfleld
                                                       082               Worcester
                                                                                          1030
                                                                                          1100
                                                                                          0930
                                   0900
                                   1000
                                   1000
                                   1000
                                                                                                         108
                                                                                                         105
                                                                                                         071
                                                  040
                                                  038
                                                  086
                                                                                                                   1015
                                                                                                                   1100
                                                                                                                   1045
                                                            1030
                                                            1030
                                                105-115      1000
                                                            1100
        aStart time of first hour and of last hour In averaging period.
                                                            51

-------
spirals were downwind of Boston as a  result of the  low  level easterly wind
flow.  August 6 was excluded because  there were no  early morning aircraft
flights on this day.  These days were replaced by July  25 and 31 in this task
only.
            In order to evaluate the  use of surface 03  as an estimate of 03
aloft, values of residuals were computed by subtracting the surface 03 concen-
trations at each site from the respective concentrations aloft.  The results
indicate a wide variation in "agreement," ranging  from  exact correspondence
to a 57 ppb underestimation of 03 aloft.  To provide  a  clearer  interpretation
of the residuals, the data set was separated into  two groups.   The  first group
contains the ten cases in which surface 63 levels  were  somewhat homogeneous
(within  +_ 15 ppb) among the upwind sites at the time of inversion  breakup
(Cases I, II, IV, V, VI, VII,  IX, XII, XIII, XV).   The  second group contains
the  remaining five cases in which there were relatively large variations in
the  upwind surface 63 levels (Cases III, VIII, X,  XI, XIV).  Such inhomoge-
neities in 03 may be attributable to spatial variations in:  (1) 03  aloft;
(2)  the time of inversion breakup; (3) the strength of  vertical mixing; and
(4)  03 formation in the boundary layer.
            Focusing on the first group indicates that, except  for  Case II,
average upwind surface 63 concentration at the time of  inversion breakup
provides a reasonable estimate of 03 aloft some three to  five hours earlier.
Excluding Case II,  the average residual (sign ignored),  for  this group was
only 4 ppb indicating good agreement.
             In the  second group, the average absolute residual  values were
fairly large  (~30 ppb), excluding Case X.  An explanation  for the large
residuals in  two of these cases  (VIII and XIV) is evident from  examining  the
                                      52

-------
late morning 03 measurements aloft.   In both cases, 03 levels aloft had
actually increased substantially during the morning.  As seen from the data
in Table 12, surface concentrations at the time of inversion breakup are
consistent with the later  values aloft.  That is, in these cases, the rela-
tively large residuals were apparently not due to the inability of average
surface concentrations to  reflect 63  concentrations aloft, but rather to a
change in the upwind 03 concentration aloft between the  time of the aircraft
measurement and the dissipation of the nocturnal inversion.
            The large residual values and the difference in surface
concentrations in Case XI  may  be due  to the presence of a low level subsi-
dence inversion (600m) which restricted vertical mixing after the dissipation
of the nocturnal inversion over portions of the Boston area.  In both Case X
and XII, the 03 concentration  at 1300 EST had increased  to within a few ppb
of the concentration aloft, indicating that the reason for the difference  in
03 observed earlier may have  been  due to a  spatial variation  in the time of
inversion breakup.  For Case  III,  no  late morning aircraft data are available,
so it is uncertain whether or  not  the relatively  high  surface 03 concentration
at the  time of inversion breakup might  be  the  result of  an increase in  03  aloft.
            Of the 15 cases examined  in  this  task,  12  exhibited close
agreement between early morning 03 aloft and  surface 03  concentrations  during
the  period of inversion breakup.   In  two of these  12 cases,  upwind 03 aloft
changed  substantially during  the  3 to 5  hours  between  the  initial aircraft
measurements and the time of inversion  breakup.   Thus, mid-morning  surface 03
measurements appear  to provide a  useful  estimate  of early  morning 03  aloft,
although care must be taken in using  mid-morning 03 data for this  purpose,
                                     53

-------
particularly for cities such as New York City and Boston  where  urban  areas  are
less than a hundred kilometers apart and fluctuations in  interurban transport
of 03 and precursors can be substantial between early morning  and the time  of
inversion breakup.

     4.4.2  Comparison of Afternoon 03 Concentrations at  the Surface  and  Aloft
            The second aspect of this task concerns comparison  of afternoon
03 concentrations at the surface and aloft.  Of interest  is whether 03 con-
centrations measured aloft by instrumented aircraft accurately represent  the
concentrations at the surface.  To examine this hypothesis, 03  concentrations
aloft from afternoon flights in the study region were compared with concen-
trations measured at nearby surface stations.  Note that  the aircraft meas-
urements used in these comparisons are 20-second average  values, while the
surface data are hourly averages.  In this analysis, pairs of measurements  at
the surface and aloft were obtained for situations when the aircraft  passed
within 15 km of a surface 03 monitoring site.  The data for these comparisons
are given in Table 13 for the New York area and Table 14  for Boston.
            The data indicate that large differences exist in many of the
comparisons.  However, several explanations exist for the observed discrepan-
cies.  As noted, the surface data are  hourly averages, while the aircraft
data are instantaneous values.  This difference in averaging times can in-
fluence  the comparison.  Another  important factor is the large concentration
gradient which exists near the edge of an urban plume.  In a number of cases,
as  indicated in Tables  13 and  14, comparisons were made near the edge of the
plume, where sharp concentration  gradients were observed.  In this region,
small vertical or  horizontal  separations  between the surface site and the
measurement point  aloft can  lead  to  large differences in 03 concentration.
                                      54

-------
Table  13.   Comparison of Afternoon Ozone  Concentrations at
              the Surface  and  Aloft  in the New  York Area
Date
7/18/80








7/22/80















7/24/80




7/31/80













8//6/80








8/8/80






03 Aloft
Tile 	 03
(EST) (ppb)
1126
1143
1253
1430
1551
1558
1805
1812
1814
1132
1142
1144
1205
1212
1316
1324
1359
1427
1431
1617
1654
1701
1743
1753
1759
1202
1211
1509
1522
1525
1142
1148
1203
1208
1229
1309
1318
1407
1721
1743
1755
1800
1809
1815
1155
1223
1248
1255
1302
1309
1451
1526
1549
1152
1213
1218
1222
1245
1251
1303
064
108
105
120
074
059
100
088
083
077
072
078
114
115
099
150
167
176
179
148
161
213
126
079
083
049
041
060
046
047
065
070
108
112
114
070
069
164
082
077
072
070
079
097
071
201
100
162
208
170
152
208
072
104
146
123
164
115
122
170
Surface 03
Time
(EST)
1100
1100
1200
1400
1500
1500
1800
1800
1800
1100
1100
1100
1200
1200
1300
1300
1300
1400
1400
1600
1600
1700
1700
1700
1700
1200
1200
1500
1500
1500
1100
1100
1200
1200
1200
1300
1300
1400
1700
1700
1700
1800
1800
1800
1100
1200
1200
1200
1300
1300
1400
1500
1500
1100
1200
1200
1200
1200
1200
1300
03
(ppb)
074
097
093
070»
068
059
078b
068b
043b
058
090C
127C
118
072d
087
174
213d
226d
092d
060 d
076d
111
125
123C
087
048
048
065
049
040
084
090
100
0708
070»
059
070
US*
070
070
101d
049b
034b
054b
085
171
090
180
249
-------
Table 14.   Comparison  of Afternoon  Ozone  Concentrations at the  Surface and

               Aloft  1n  the Boston  Area
                              Surface 03
   Date
             Time
             (EST)
       (ppb)
        03
       (ppb)
       Time
       (EST)
                                                     Station
                                              Distance from
                                           Surface Station, km
  7/15/80

  7/16/80

  7/17/80
1525

1653

1639
1707
153

159

134
125
128

093

110
116
1500

1600

1600
1700
Medfield,  MA

Medfield,  MA

Medfield.  MA
Medfield,  MA (124  at 1800)
6.3

9.2
                                                                                       5.9
                                                                                       2.6
8/1/80





8/5/80






8/6/80




8/8/80


1357
1420
1423
1424
1432
1600
1355
1621
1657
1700
1704
1704
1707
1315
1326
1332
1538
1611
1516
1551
1555
126
125
128
130
108
151
133
104
125
107
105
105
107
089
101
098
074b
077b
090
076
082
117
112
114
117
113
098
096
100
105
116
100
042
060
111
106
089
075
068
067
075
064
13003
1400*
14003
1400«
1400
1500*
1400a
1600
1600
1700
1700
17003
1700*
1300
1300
1300
1500
1600
1500
1500
1500
Hamilton, MA (134 at 1200)
Hamilton, MA
Danvers. MA (130 at 1300)
Georgetown, MA (134 at 1300)
Manchester, NH
Portsmouth, NH
Georgetown, MA
Worcester, MA
Easton, MA (090 at 1700)
Medfield, MA (109 at 1600)
Sudbury, MA
Water town, MA
Somervllle, MA
Sudbury, MA
Medfield, MA
Quincy, MA
Easton, MA
Tewksbury, MA
Danvers, MA
Sudbury, MA
Tewksbury, MA
11.5
3.9
4.8
8.5
15.3
13.6
1.4
1.9
12.3
7.1
13.5
3.3
8,4
	
—
—
	
—
	
8.0
4.8
 a  Sea breeze or complex surface  flow resulted  In large vertical/horizontal gradients in Oj.

 b  Aircraft In region of large spatial gradient aloft; average concentration as aircraft passed the site.
                                              56

-------
Comparisons made near  coastal areas during a sea breeze flow were likely
influenced by the presence  of a  shallow internal bounday layer, which iso-
lated surface sites from the impact of a plume aloft, or prevented high
surface concentrations from mixing upward.   In some of the the Boston cases,
vertical wind speed shear may have resulted  in a faster travel time for
upwind transport aloft, such that the surface impact was an hour or so later
than that measured by  the aircraft.  Also, in comparisons of measurements made
late in the day (generally  after 1700 or 1800 EST), surface concentrations
may have been depleted by local  scavenging as the nocturnal stable layer
began to form and isolate 03 in  the  surface  layer  from 03 aloft.  The cases
where these phenomena  affect the surface/aloft comparisons are noted in Tables
13 and 14.
            Considering the above factors, the 85  cases in the tables were
divided into two groups for the  evaluation.   Group  I contains  those cases
where the atmosphere is likely  to be well mixed, and both measurements  (sur-
face and aloft) were not on the  fringe  of  the urban  plume.  Group  II contains
those cases in which complicating features,  as  described above,  were observed.
For Group I - New York (35  cases),  the  average  residual  (sign  ignored)  between
surface and aloft was 10 ppb.   In 63 percent of the Group  I cases, 03 aloft
exceeded 03 at the surface.  In  contrast,  the average  residual  for  Group  II  -
New York (25 cases) was 49 ppb.   In  most of  the Group  II cases 03  was much
higher aloft, due to depletion  near the surface or as  a  result of  low  surface
concentrations in the sea breeze onshore flow.
            For  Group  I - Boston (17 cases), the average  residual  of 15 ppb
was somewhat higher than in New York and,  again, 03 aloft  exceeded surface
                                      57

-------
concentrations in most (71 percent) of the cases.   In  Group  II  - Boston (8
cases), the average residual was double the value  of Group  I.
            One conclusion which can be drawn  from these  comparisons is that
aircraft data are most useful for estimating surface concentrations during
mid-day when the boundary layer tends to be well mixed, and  in  areas away
from the gradients associated with urban plumes and sea breeze  flows.  How-
ever, in some monitoring circumstances, the existence  of  such gradients will
not be known a priori , and their influence will be difficult to avoid.  As a
consequence, the use of aircraft data to estimate  fixed point surface 03
concentrations must be viewed with caution in  the  following  situations:
(1) in areas with complex wind flow patterns-  (2)  in the  vicinity  of an urban
plume; and  (3) when the column of air from the surface to the height of the
aircraft measurement  is not well mixed.  On the other  hand,  aircraft measure-
ments appear to be valuable  for estimating surface 03  concentrations in the
absence of  these cdmplicating situations.
                                         58

-------
                                  SECTION 5
                                 CONCLUSIONS

     It was the  purpose of this study to investigate various aspects of ozone
formation  and transport in the New York City and Boston portions of the
Northeast  Corridor.   At the outset of the study eight questions were posed
relative to this topic which were addressed through numerous analyses.
     The conclusions  of the study pertinent to these questions are presented
below.
     1.  Do concentrations of 03 and precursors transported into New
        York City and Boston differ during along-Corridor and non-
         Corridor transport regimes?
     For New York City, morning precursor concentrations  transported into the
urban area at the surface were twice as high with along-Corridor transport
than when  transport was  from outside the Corridor.  Aloft, precursor concen-
tratins, particularly NMOC, were also much  higher when  transport was along the
Corridor.   The analysis indicates that high morning precursor concentrations
transported into the  New  York area with along-Corridor  flow were attributable
to overnight emissions in  the Philadelphia  area.  During  mid-afternoon  the
impact of the Philadelphia 03 plume was typically observed at one or more
monitoring sites on  the  upwind  (southwest)  fringe of New  York City.
     In Boston, there was little difference in  surface  precursor concentrations
during along-Corridor versus non-Corridor  transport regimes.  However,  the con-
centration of NMOC aloft during along-Corridor  transport  was  triple the magni-
tude of aloft concentrations when  transport was  from outside  the Corridor.
Average 03 aloft transported  into  Boston during  the morning was also  higher
with a long-corridor flow (100 ppb  versus 76 ppb).  The  impact of the  New  York
                                      59

-------
u>/ o. plunu- on portions of the Boston area  was observed  during  the evening
across the upwind (southwest) fringe of Boston on days  when  transport was along
the Corridor.
     The analyses indicate that the concentration of 03 and  precursors  trans-
ported into New York City and Boston can vary substantially  depending upon
transport direction.  Vertical variations in  transort direction overnight were
often associated with large gradients in morning pollutant concentrations aloft.
Also, on occasion, transport of high 03 concentrations  from  major upwind cities
resulted in 63 exceeding the NAAQS in portions of the Boston and  New York areas.
     2.  What is the diurnal variation of 63  and precursors  in  the air  parcels
         leading to the maximum 63 concentration in the urban plume?
     In New York City, the concentration of 03 decreased as  the air parcel
traveled from upwind rural and suburban locations into  the urban  area in the
morning, then increased rapidly later in the  .morning as the  air parcel  departed
the city and the rate of photochemical reactions increased.   In almost  every
case,  the maximum 0.3 concentration in the plume occurred between  1300 and
1500 EST.  The concentrations of NO, N02, and NMOC generally peaked as  the
air parcel crossed  the city  in the morning.  The diurnal varition of 03 and
precursor concentrations are  provided  for 10  days in Section 4.1.2.
     3.  What is the average  transport time and distance to maximum
         03  in  the  urban plume?  What  is the typical downwind distance
         to maximum N02 in  the urban plume?
     For New York City, the  analyses indicate that the highest 03 concentrations
were associated with air parcels crossing the city at approximately 0800 EST.
On  the average, air parcels  crossing the city at 0800 EST generated an  03
maximum of 219  ppb  at  1400  EST.  The average downwind distance to maximum  03
was ~100 km.  The average  transport  time to  peak 03 was 5 to 7 hours.
                                      60

-------
     In  the Boston  area, aircraft data provided the most useful information
relative to these  questions  since the  urban  plume was  transported over the
ocean on all  but one  case  study  day.   The aircraft data indicate that the
average  distance to the 03 maximum  aloft was 81 km, with a range of 61 to
107 km.   The average  transport time to peak  03 was 4 hours.
     At the surface,  maximum N02 concentrations occurred in and immediately
downwind of the city. Aloft, away  from the  effect of  local sources, maximum
midday N02 concentrations  were measured 23  to 48  km downwind  of New York
City, and 34 to 110 km downwind  of  Boston.   For both cities,  the transport
time to maximum N02 ranged from  1 to 3 hours.
     4.   What is the  typical  downwind distance to where NOX in  the urban
         plume becomes indistinguishable  from  background concentrations?
     For three days in Boston and two days  in  New York City,  the distance  from
the urban area to  the point where the urban  plume NOX  concentrations aloft  be-
came indistinguishable from the  air mass  background NOX concentrations were
estimated.  This distance  ranged from 85  to 165 km.  The estimated travel  time
to background NOX  ranged  from 5.5 to 12 hours.
     5.   Can mid-morning  surface 03 measurements  be used to estimate  early
         morning upwind  03 aloft?
     The analyses  indicate that  average upwind surface 03  concentrations  (3
hour average centered around the time of  inversion  dissipation)  do  provide  a
meaningful estimate of early morning 03 concentrations aloft  in  many  cases.
However, there were also a number of cases when  such  an assumption would  lead
to signficant underestimation (or  overstimation)  of the  early morning  levels
aloft.   In these situations, it appears  that 03  aloft transported  into  the
city had actually increased (or decreased)  between  the time of the  early
                                     61

-------
morning aircraft measurements and the time  of inversion dissipation.  This
may have been due to reactions among 63  and transported precursors or to
spatial variations in 03 aloft transported  across the  urban area.  Thus,
using surface data for estimating early  morning  concentrations aloft should
be done with caution, particularly in situations where urban areas are in
fairly close proximity or high concentrations of transported precursors are
expected.
     6.  Can mid-day aircraft measurements  of 03 be  used  to estimate
         surface concentrations between  measurement  sites?
     In locations where the atmosphere appears to be well mixed,  surface and
aircraft data agree within 10 to 15 ppb.  However, it  was observed that strong
vertical and horizontal gradients confound  the use of  aircraft data for esti-
mating surface concentrations in the vicinity of urban plumes and sea breeze
ci rculations.
     7.  Does 03 aloft, initially isolated  from the  effects of surface
         emissions and scavenging, change substantially prior to  the
         dissipation of the nocturanl inversion  when pollutants aloft
         are mixed to the surface?
     The data examined indicate that, in most cases, 03 concentrations aloft
were fairly  stable (within ±5 ppb) between  early morning  and mid-morning
measurements made within an air parcel.   However, 03 production was evident
in  those air parcels containing comparatively high air mass NOX concentra-
tions and  probably other 03 precursors,  as indicated by  the aerosol content
ansd estimated  track of the air parcel  relative to upwind urban areas.
     8.   Is  there evidence in the data  of 03 plumes from medium  size
          cities such as Bridgeport,  New Haven, and Hartford,  CT  or
          Providence, RI?
     In  general,  it was difficult to define 63 plumes from such  cities  due
 to the  relatively  high air mass 63  levels, the complexity of airflow  patterns
                                     62

-------
and the frequent incursions of urban  plumes from the major Corridor cities.
Also, since the monitoring  program was not directed toward investigating  these
cities, comparatively little data  were available for this type of analysis.
However, on two occasions,  there was  evidence of the Providence 03 plume  from
the Boston area aircraft data.  In the clearest example,  03 in the Providence
plume was 20 to 30 ppb higher than the mass 03 concentration upwind, outside
the Boston plume.
                                       63

-------
                                  REFERENCES


 1.   Cleveland, W. S.; Kleiner, B.; McRae, J. E.; and Warner,  J.  L.,
     "Photochemical Air Pollution: Transport from the New York City Area
     into  Connecticut and Massachusetts," Science, 191,  179  (1976).

 2.   Wolff, G. T.; Lioy, P. J.; Meyers, R. E.; Cederwall, R.  T.;  Wight, G.  D.;
     Pasceri, R.  E.; and Taylor, R. S., "Anatomy of Two  Ozone Transport
     Episodes in  the Washington, DC to Boston, Massacusetts  Corridor," Environ.
     Sci.  and Tech., _U, 506  (1977).

 3.   Spicer, C. W.; Joseph, D. W.; Sticksel, P. R.; and  Ward, G.  F.,  "Ozone
     Sources and  Transport in  the Northeastern United States," Environ. Sci.
     and Tech., .13, 975 (1979).

 4.   Ludwig, F. L. and Shelar, E., "Ozone in the Northeastern United  States,"
     EPA-910/9-76-007, U.S. Environmental Protection Agency,  Research Triangle
     Park,  North  Carolina (1977).

 5.   Wolff, G. T.; Lioy, P. L.; Wight, G. D.; Meyers, R. E.;  and
     Cederwall, R. T., "An Investigation of Long-range Transport  of Ozone
     Across the Midwestern and Eastern United States," Atmos. Environ.,  11,
     797  (1977).                                                       ~

 6.   Reynolds, S. D. and Reid, L. E., "An Introduction to the SAI Airshed
     Model  and  Its Usage," Systems Applications, Inc., Report EF  78-53R  (1978).

 7.   Heffter, J.  L.; Taylor,  A. D.; and Ferber, G. J., "A Regional-Continental
     Scale Transport, Diffusion and Deposition Model,"  NOAA Tech. Memo.
     ERL  ARL-50 (1975).

 8.   Heffter, J.  L.; "Air Resources Laboratories Atmospheric Transport and
     Dispersion Model (ARL-ATAD)," NOAA Tech. Memo.  ERL ARL-81 (1980).

 9.   McNaughtpn,  D. J. and Powell, D. C., RAPT--The Pacific Northwest
     Laboratory Regional Air  Pollutant Transport Model:   A Guide," Report
     prepared  for the U.S. Department of Energy under contract DE-AC06-76RLO,
     1830  (1981).

10.   Spicer,  C. W.; Joseph, D. W.; and Sticksel, P. R., "An Investigation of
     the  Ozone  Plume  from a Small City," J. Air Poll. Control Assoc., 32,
     278  (1982).

11.   Ludwig,  F. L.,  "Assessment of Vertical Distributions of Photochemical
     Pollutants and  Meteorological Variables  in the Vicinity of Urban Areas,"
     EPA-450/4-79-017,  U.S. Environmental Protection Agency, Research Triangle
     Park, North  Carolina  (1979).
                                      64

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-450/4-84-001
4. TITLE AND SUBTITLE
Northeast Corridor Regional
Precursor Transport in New
the 1980 Field Program
2. 3. RECIP
5. REPOf
Model inq Pro.iect — Ozone and Augu
York City and Boston During 6-PERFC
7.AUTHOR c^ Poss1el) EpA. c> w_ Sp1cerj p> R; stickse'8'PERFC
and G. M. Sverdrup, BCL; and A. J. Alkezweeny and
W E Davic BNL
9. PERVOR'MING dRGANTlZATION NAME At
J.S. Environmental Protectic
Dffice of Air Quality Planni
Monitoring and Data Analysis
Research Triangle Park, NC
•JO ADDRESS 10. PRO
)n Age!3cL , J A24A
ng and Standards n CON
Division (MD-14)
27711
12. SPONSORING AGENCY NAME AND ADDRESS 13. TYP

14. SPOf
15. SUPPLEMENTARY NOTES
lENT'S ACCESSION NO.
=IT DATE
St 1984
)RMING ORGANIZATION CODE
3RMING ORGANIZATION REPORT NO.
3RAM ELEMENT NO.
2F
TRACT/GRANT NO.
E OF REPORT AND PERIOD COVERED
MSORING AGENCY CODE

16. ABSTRACT
This report describes the results of a study to analyze portions of the Northeast
Corridor Regional Modeling Project (NECRMP) ambient data base collected in the New
York City and Boston urban areas. The study includes (1) an examination of the
ozone, oxides of nitrogen, and hydrocarbon concentrations transported into each city
and downwind in the urban plumes on 20 high ozone days; (2) the temporal changes in
ozone above the boundary layer; and (3) the relationship between surface ozone con-
centrations and mid-boundary layer values.
17.
a. DESCRIPTORS
Air Pollution
Ozone
Nitrogen Oxides
Hydrocarbons
Transport
18. DISTRIBUTION STATEMENT
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN END
NECRMP
New York City
Boston
Aircraft measurem
Urban plume
19. SECUHITY CLASS (This
20. SECURITY CLASS (This

ED TERMS c. COSATI Field/Group
ents
Report 1 21. NO. OF PAGES
80
pagei 22. PRICE
EPA Form 2220-1 (Rev. 4-77)    PREVIOUS EDITION is OBSOLETE

-------