United States Office of Policy, EPA 230/2-89/026
Environmental Protection Planning, and Evaluation February 1989
Agency (PM-221)
<&EPA Photochemical Modeling
Analysis of Emission
Control Strategies in the
New York Metropolitan
Area
-------
PHOTOCHEMICAL MODELING ANALYSIS OF EMISSION CONTROL STRATEGIES
IN THE NEW YORK METROPOLITAN AREA
by
S.T. Rao, G. Sistla and R. Twaddell
Division of Air Resources
New York State Department of Environmental Conservation
Albany, New York 12233-3259
No. CR-814051-01-0
EPA Project Officer: Robin Miles-McLean
Prepared for
U.S. Environmental Protection Agency
Office of Policy, Planning & Evaluation
Washington, D.C. 10460
-------
DISCLAIMER
This report has been reviewed by the Office of Policy, Planning &
Evaluation, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency and the New York State
Department of Environmental Conservation, nor does mention of trade name or
commercial products constitute endorsement or recommendation for use.
(i)
-------
ABSTRACT
Despite a downward trend in the emissions of ozone precursors, like many
urban areas in the United States, the New York metropolitan area continues to
experience high levels of ozone concentrations. Because different types of
Volatile Organic Compounds (VOCs) have different levels of reactivity, deter-
mining which control strategies would be most effective in reducing the ambient
ozone concentration levels is a complex problem. The primary objective of this
study is to evaluate in the New York metropolitan area some of the specific
emission control options, envisioned under the EPA/s post-1987 ozone policy and
various bills before Congress.
The emission control strategies analyzed in this study were evaluated using
the Urban Airshed Model (DAM) with the aerometric data for one of the high ozone
days in 1980. Although the strategies considered here can achieve a reduction
in VOCs and nitrogen oxides (NO ) emissions over the modeling domain by about
A.
53% and 47%, respectively, from the base year, the predicted peak ozone concen-
tration in the New York metropolitan area is still well above the level of the
ozone National Ambient Air Quality Standard (NAAQS). A modeling simulation with
an across-the-board reduction in the VOC emissions over the modeling domain of
80% as well as upwind boundary concentration reduction of 80% from the 1980
level while keeping the NO concentrations at their 1980 levels, indicates that
J\
even this level of VOC emissions reduction is not sufficient to reduce the peak
ozone concentration in the New York metropolitan area below the level of the
ozone NAAQS. However, the model predicts that the peak ozone concentration over
the New York metropolitan area can be reduced to the level of the ozone NAAQS
for the day modeled with the VOC concentration reduction from the upwind
boundary at the 80% level when coupled with an acrossthe-board reduction in the
VOC emissions within the modeling domain by 95% from their 1980 levels.
(ii)
-------
CONTENTS
Abstract ii
List of Tables '. iv
List of Figures vi
Acknowledgements vii
Chapter 1 - Introduction 1
Chapter 2 - Model Application
2.1 Modeling Domain 5
2.2 Modeling Day - August 8, 1980 (JD80221) 5
Chapter 3. - Emission Inventories - Base Case and Control Strategies
3.1 SCOPE BASE Emissions 17
3.2 SCOPE STRATEGY 1 22
3.3 SCOPE STRATEGY 2 22
3.4 SCOPE STRATEGY 3 25
3.5 SCOPE STRATEGY 4 25
3.6 SCOPE STRATEGY 5 32
3.7 Synopsis of the Scope Strategies 32
Chapter 4 - Modeling Results
4.1 Base Case and Strategy Simulations 37
4.2 Strategy to Reduce Ozone Levels to the NAAQS -
SCOPE STRATEGY 5 45
4.3 Discussion 49
Chapter 5 - Summary and Recommendations 57
References 59
(iii)
-------
LIST OF TABLES
Number
2.1 Hourly Highest and Second Highest Ozone Concentrations
Measured on August 8, 1980 (JD80221)
2.2 Hourly Diffusion Break (Mixing Height), Region and
Vertical Cell Top Heights for August 8, 1980 (JD80221) 10
2.3 Vector-Averaged Hourly winds for August 8, 1980 (JD80221) 11
2.4 Metscalar Parameters for August 8, 1980 (JD80221) 12
2.5 Pollutant Concentrations at the Top of the Modeling Region
for August 8, 1980 (JD80221) for Base Case (1980) and
Strategies 15
3.1 Summary of 1980 Emissions OMNYMAP Base over the Modeling
Domain (Tons/Year) 18
3.2 Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions
(G-Moles) for OMNYMAP Base 19
3.3 Summary of Emissions for SCOPE BASE over the Modeling
Domain (Tons/Year) 20
3.4 Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions
(G-Moles) for SCOPE BASE 21
3.5 Summary of Emissions for SCOPE STRATEGY 1 over the Modeling
Domain (Tons/Year) 23
3.6 Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions
(G-Moles) for SCOPE STRATEGY 1 24
(iv)
-------
LIST OF TABLES
Number Page
3.7 Summary of Emissions for SCOPE STRATEGY 2 over the
Modeling Domain (Tons/Year) 26
3.8 Summary of Typical Day (0400 to 2000 hrs) Speciated
Emissions (G-Moles) for SOOPE STRATEGY 2 27
3.9 Summary of Emissions for SOOPE STRATEGY 3 over the
Modeling Domain (Tons/Year) 28
3.10 Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions
(G-Moles) for SOOPE STRATEGY 3 29
3.11 Summary of Emissions for SOOPE STRATEGY 4 Over the Modeling
Domain (Tons/Year) 30
3.12 Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions
(G-Moles) for SOOPE STRATEGY 4 31
3.13 Summary of the Emission Control Strategies Considered for
the New York Metropolitan Area 33
4.1 Pollutant Concentrations for "Clean" Conditions 47
4.2 Conditions for SCOPE STRATEGY 5A - 50
4.3 Conditions for SCOPE STRATEGY 5B - 51
4.4 Peak Ozone Level Over Connecticut Under Various Strategies
for a Selected Meteorological Scenario 55
(v)
-------
LEST OF FIGURES
Number
2.1
2.2
2.3
2.4
3.1
3.2
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8.
4.9a
4.9b
4.10
Model Domain Covering Portions of the States of New Jersey,
Synoptic Weather Pattern for August 8, 1980 (JD80221)
Initial Pollutant Distribution for August 8, 1980 (JD80221) . . .
Diurnal Variation of Pollutant Concentrations at the Southwest
Corner Cell for August 8, 1980 (JD80221)
Summary of Percentage of VDC Reduction in the Tri-State
Region of the Modeling Domain
Summary of Percentage of NO Reduction in the Tri-State
Diurnal Plot of Pollutant Concentrations at the Southwest
Corner Grid Cell of the Modeling Domain.
Areal Distribution of Ozone for OMNYMAP BASE 1980 Simulation. .
Areal Distribution of Ozone for SCOPE BASE 1988 Simulation. . . .
Areal Distribution of Ozone for SCOPE STRATEGY 1 Simulation. . .
Areal Distribution of Ozone for SCOPE STRATEGY 2 Simulation. . .
Areal Distribution of Ozone for SCOPE STRATEGY 3 Simulation. . .
Areal Distribution of Ozone for SCOPE STRATEGY 4 Simulation. . .
Areal Distribution of Ozone Under Different Sensitivity
Areal Distribution of Ozone for SCOPE STRATEGY 5a Simulation. .
Areal Distribution of Ozone for SCOPE STRATEGY 5b Simulation. .
Number of Grid Cells Exceeding the Ozone NAAQS Level Under
Each UAM Simulation
Page
6
8
13
14
34
35
38
39
40
42
43
44
46
48
52
52
53
4.11 Percentage Reduction in the Predicted Ozone Concentrations
Associated with the VOC and NO Emission Control Strategies
Evaluated in this Study 54
(vi)
-------
The authors gratefully acknowledge the technical assistance provided by
Edward Davis (NYSDEC), Norman Possiel (EPA/QAQPS) and Kenneth Schere (EPA/ORD).
This work would not' have been completed without the encouragement and support of
Harry Hovey and Thomas Allen of NYSDEC. Special thanks are extended to
Stephanie Liddle and Linda Stuart for typing the manuscript and to Carol Clas
and Gary Lanphear for their excellent cartographic work.
This work was performed for the U.S. Environmental Protection Agency's
Office of Policy, Planning and Evaluation under the Cooperative Agreement
No. CR-814051-01.
(vii)
-------
CHAPTER 1
INTRDDUCTION
Ozone concentrations in the northeastern part of the United States continue
to exceed the level of the National Ambient Air Quality Standard (NAAQS) for
ozone despite a downward trend in the emissions of ozone precursors in this
region. These exceedances mainly occur during the months of May through
October, the so-called "ozone season," and are found to be region-wide
indicating that ozone is a pervasive air contaminant. Currently, the level of
the ozone air quality standard is exceeded in over 60 urban areas across the
country. As a result, a large portion of the population continues to be exposed
to frequently healthful levels of ozone. Both Congress and the U.S.
Environmental Protection Agency are under increasing pressure to develop and
implement effective emission control programs for protecting the public health
and welfare . This study attempts to analyze the effectiveness of a series of
control strategies in mitigating the ozone problem in the urban areas.
The relationship of ozone to its precursors, in fact, the ozone-forming
process in its entirety, must be elucidated before rational and effective
precursor control strategies can be developed. Because the oxidation of
non-methane organic compounds (NMOCs) leads to the formation of ozone, a
reduction in NMOC is expected to reduce ozone production. The efficiency of
ozone reduction, however, depends upon the amount of oxides of nitrogen (NO ).
' - j^
Since different types of volatile organic compounds (VOCs) have different levels
of reactivity, determining which control strategies will be most effective in
reducing the ambient ozone levels is a complex problem. A one ton reduction of
VOCs from mobile sources will not have the same impact on ozone formation as
will a one ton reduction of VOCs from architectural coatings. The "VOC
reactivity" issue has been attracting increased attention because cost-effective
reductions might be achieved by excluding unreactive VOCs both from inventory
and control.
Ozone is not usually emitted directly into the atmosphere, but is instead a
secondary pollutant that is formed over a period of time from a variety of
atmospheric reactants. The magnitude of the ozone concentration in an urban
-------
-2-
area depends upon the transport of ozone and its precursors into the region,
precursors emitted within the region, the rate at which the chemical reactions
take place, and the transport and diffusion of pollutants out of the region. In
order to assess whether "a region will be in compliance with the ozone NAAQS at
some future date, one needs to utilize mathematical models which predict the
complex relationship between the precursor emissions and ozone air quality. The
current generation of photochemical air quality models can with reasonable
accuracy predict the peak ozone concentration downwind of an urban area
resulting from prescribed changes in source emissions. The Urban Airshed Model
(UAM) is one of the grid-based photochemical air quality models which treats the
atmospheric physical and chemical processes in a sophisticated manner. With
this model, it is possible to determine the most effective means for reducing
the ozone concentrations in a large metropolitan area through the application of
control strategies to specific source categories.
The objectives of this study (referred to as the SCOPE Project) are to:
(a) evaluate the impacts of specific control options on ambient ozone
concentrations in the New York metropolitan area, and (b) assess the relative
merit of various control plans in mitigating the ozone problem in other major
urban areas. As part of this investigation, new emission inventories for a
future year have been developed for the New York metropolitan area to analyze
the impact of reductions from such source categories as evaporative emissions
from the use of high RVP gasoline, gasoline refueling emissions, enhanced
inspection/maintenance programs, autobody refinishing and architectural
coatings, alternative fueled vehicles, etc. The,UAM has been applied with each
emission inventory for a selected meteorological scenario. In this study,
controls have been applied on an incremental basis, i.e., an additional control
for each model simulation so the effects of each set of control strategies could
be analyzed separately. Such information can then be used by the regulatory
agencies in their efforts to identify cost-effective ozone control strategies.
Although the control strategies considered in this study can achieve a
reduction in VOC and NO emissions over the New York metropolitan area by about
.X
53% and 47%, respectively, from their corresponding 1980 emission levels, the
predicted peak ozone concentration in the modeling domain is still well above
the level of the ozone NAAQS. A modeling simulation with an across-the-board
-------
-3-
reduction in the VOC emissions over the New York metropolitan area of 80% as
well as upwind boundary concentration reduction of 80% from the 1980 level,
while keeping the NO concentrations at their 1980 levels, indicates that even
X
this level of reduction in the VDC emissions is not sufficient to reduce the
peak ozone concentration over the modeling domain below the level of the ozone
NAAQS. However, a simulation retaining the VDC concentration reduction from the
upwind boundary at the 80% level when coupled with an across-the-board reduction
in the VDC emissions in the New York metropolitan area by 95% from their 1980
levels, reveals that for the day modeled the peak ozone concentration over the
New York metropolitan area can be reduced to the level of the ozone NAAQS.
-------
-4-
(BLANK PAGE)
-------
-5-
CHAPTER 2
MODEL APPLICATION
In examining the various emission control strategies for achieving the
ozone NAAQS in the New York metropolitan area, the modeling domain (see Figure
2.1) encompassing portions of the States of New Jersey, New York and
Connecticut, has been utilized. Thus, the results from this study can be
compared with those of the previous study on the Urban Airshed Model application
2
to this region (OMNYMAP) . The following is a brief description of the model
set-up and its application. Further details on the model design can be found
elsewhere.
2.1 MODELING DOMAIN
The modeling domain (see Figure 2.1) extends 248 km east-west and 200 km
north-south with its southwest corner set approximately at Trenton, NJ and
northeast corner at East Thompson, CT. near the Massachusetts and Rhode Island
border. The grid size was set at 8 km resulting in 31 cells in the east-west
and 25 cells in the north-south directions, respectively. The layer between the
ground and the top of the simulation region was divided into four levels whose
thicknesses were varied during the day as a function of the height of the
mixed-layer.
2.2 MODELING DAY - AUGUST 8, 1980 (JD80221)
Based upon the application of the UAM to the New York metropolitan area for
2
five high ozone days in the 1980 oxidant season, August 8, 1980 was selected as
the candidate day in this study. UAM simulation for this day indicated that the
peak modeled concentration agreed well with the peak measured ozone
concentration. Also, the correlation coefficient between the measured and
predicted concentrations is 0.74, the highest correlation achieved over the five
high ozone day simulations performed under the CMNYMAP study. Further, since a
variety of sensitivity and strategy simulations have been reported with the
July 21, 1980 case in the CMNYMAP study, examination of a different day could
provide additional confidence in the modeling results. Hence, in this study,
-------
< S U L L
J
\
s
"L^
*
A
v+ N
BARREN .'
»6
\.
Sr ?W --
/
/
k
/ U L S T
\
r i
ER
XJO HfCHflELI. J HARIFOHD | ' OL L AN DL->- W IN 0 H A
r T i _jj
IVAN
i .r"1
*
V
Ur ^
NEW LONDON
S MIOOL E SE X
0 H A N G
/
. S S E « / * /ROCKLANO
P U I N A M I
"~~"~~~~!
'.
.WESTCHISUR
HE« HAVEN
\
\ .^MIDDLESEX
O 10 20 30 40 SO Km
-------
-7-
all the analyses were conducted with the August 8, 1980 case. The synoptic
weather pattern, shown in Figure 2.2, consists of a "Bermuda High" with an
extension over the Appalachians and a high near James Bay with a stationary
front that extends to a low over the Lake Superior. The surface winds during
the day were from a south-southwesterly direction with speeds in the range of
3.5 to 5 m/s with maximum surface temperature in the 90-95 F range. The hourly
highest and second highest measured ozone concentrations are given in Table 2.1.
The measured ozone peak value for the day is 246 ppb at Stratford Light House,
CT, with several other monitoring stations in the region reporting
concentrations in excess of the ozone NAAQS.
The meteorological conditions for the UAM simulation are provided in Tables
2.2 through 2.4, respectively. The air quality data, initial and boundary
concentration fields required for the model were estimated from the 1980 ambient
data and are shown in Figures 2.3 and 2.4. The initial and boundary fields
which were representative of the 1980 conditions need to be "modified" to
reflect the future-year conditions under the various emission control strategy
scenarios. This was accomplished, as in the CM₯NMAP study, by scaling the
initial and boundary concentrations of the precursor pollutants to reflect the
changes in the precursor emissions from their 1980 levels. In the case of
ozone, the future-year initial and boundary fields due to changes in the
precursor emissions cannot be estimated easily. The region-top concentrations
of ozone range from 60 to 85 ppb for the 5 days of 1980 modeled in the OMNYMAP
study. With the projected decreases in the precursor emissions, these levels
should probably be in the range of 40 to 60 ppb, or a 20 to 30% reduction from
3 4
the 1980 concentration level consistent with the the suggested estimates ' of
the background concentrations. In this study, as a first approximation, the
assumed reduction in ozone was set at 20% from the 1980 level for all future
strategy simulations. Table 2.5 provides the pollutant concentrations at the
top of the modeling region for the 1980 simulation as well as for the
future-year scenarios.
-------
-8-
Figure 2.2 Synoptic Weather Pattern for August 8, 1980
(JD80221)
-------
-9-
TABIE 2.1
Hourly Highest and Second Highest Ozone Concentrations
Measured on August 8. 1980 (JD80221)
HOUR OF
THE DAY
1200
1300
1400
1500
1600
1700
1800
- 1300
- 1400
- 1500
- 1600
- 1700
- 1800
- 1900
HIGHEST MONITORING
CONCENTRATION STATION
(PPb)
213 Stratford
246* Stratford
237** Stratford
236 Stratford
197 Stratford
160 Stratford
143 Stratford
2nd HIGHEST
CONCENTRATION
(PPb)
180
170
167
145
141
132
143
**
Highest for the day
Second highest for the day
MONITORING
STATION
Greenwich
Bridgeport
Bridgeport
Stony Brook
Derby
Derby
Middletown
-------
-10-
TABLE 2.2
Hourly Diffusion Break (Mixing Height!. Region and Vertical Cell Top
Heidhts for Auoust 8. 1980 fJD8022H
HOUR OF DIFFUSION BREAK
THE DAY
0400
0500
0600
0700
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
- 0500
- 0600
- 0700
- 0800
- 0900
- 1000
- 1100
- 1200
- 1300
- 1400
- 1500
- 1600
- 1700
- 1800
- 1900
- 2000
(m)
345
345
345
375
405
450
540
700
1020
1400
1400
1400
1400
1170
940
710
REGION TOP
fm)
1000
1000
1000
1000
1000
1040
1100
1160
1280
1400
1400
1400
1400
1400
1400
1400
TOP OF CFTTi (m)
3
345
345
345
375
405
450
540
660
780
900
900
900
900
795
705
630
2
230
230
230
250
270
300
360
440
520
600
600
600
600
530
470
420
1
115
115
115
125
135
150
180
220
260
300
300
300
300
265
235
210
-------
-11-
TABIE 2.3
Vector-Averaged Hourly Winds for August 8. 1980 (JD80221)
HOUR OF WIND SPEED WIND DIRECTION
THE DAY fm/sl ( )
0600 - 0500 3.53 227
0500 - 0600 3.48 234
0600 - 0700 3.77 241
0700 - 0800 3.24 244
0800 - 0900 3.67 231
0900 - 1000 3.82 232
1000 - 1100 3.99 237
1100 - 1200 4.57 238 .
1200 - 1300 5.26 233
1300 - 1400 4.80 235
1400 - 1500 5.09 . 236
1500 - 1600 5.63 236
1600 - 1700 4.49 234
1700 - 1800 5.11 245
1800 - 1900 4.48 239
1900 - 2000 4.84 224
-------
TABLE 2.4
Metscalar Parameters for August 8. 1980 (JD80221)
HOUR OF TEMPERATURE
THE DAY
0400-0500
0500-0600
0600-0700
0700-0800
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
BEIDW
.0049
.0049
.0008
.0008
-.0039
-.0084
-.0099
-.0112
-.0106
-.0101
-.0098
-.0096
-.0093
-.0090
-.0088
-.0088
GRADIENT ( K/n>)
ABOVE
-.0069
-.0069
-.0075
-.0075
-.0063
-.0051
-.0061
-.0070
-.0064
-.0058
-.0063
-.0068
-.0073
-.0078
-.0083
-.0083
EXPOSURE
INDEX
0
0
1
1
1
2
2
2
2
1
1
1
1
0
0
-1
PHOTOLYSIS CONCENTRATION OF
RATE
.0010
.0010
.0848
.2355
.3509
.4247
.4718
.5020
.5234
.5319
.5120
.4733
.4059
.2861
.1125
.0010
WATER VAPOR (PPM)
16065.0
16065.0
16065.0
16065.0
16065.0
16072.0
16072.0
16083.0
16093.0
16603.0
16614.0
16126.0
16136.0
16143.0
16659.0
16659.0
ATMOSPHERIC
PRESSURE (ATM)
0.9813
0.9811
0.9810
0.9810
0.9811
0.9807
0.9807
0.9801
0.9796
0.9791
0.9787
0.9775
0.9771
0.9768
0.9763
0.9761
ro
-------
-13-
20
IS
at
x
10
INITIAL SURFACE DISTRIBUTION OF NO?
FOR 080880 (JD80221)
10 19 20 29
X-AXIS
JO
INITIAL SURFACE DISTRIBUTION OF NMHC
FOR 080880(JO80221)
INITIAL SURFACE DISTRIBUTION OF OZONE
FOR 080880 (J080221)
INITIAL SURFACE DISTRIBUTION OF CO
FOR 080880 (JO80221)
19 (O IS 20 29 SO
19 tO 19 20 29 M
Figure 2.3
Initial Pollutant
Distribution for
August 8, 1980
(JD80221)
JO
-------
100-
90-
m
0. 8O-
Q.
z 70-
O
H 60
^1
ty
»- 50
Z
UJ
0 40-
z
o
0 30-
20
10
OZONE o
45]
O
O 40-
0°
35
0
0 30
o
25
° 20
O
45-
.0
0 5
0 rv 0
NO 2
0 0
0
o ° o o
o o 0
° 0 0 0 0 0
»
2 4 6 8 10 12 44 16 48 20 " 2 4 6 8 40 12 44 16 18 20
900-
0 800-
CQ
Si 700-
O 600-
1-
K 50°"
2 400-
U
| 300
U
200
1OO-
NMHC0o
0 3200
O
0 0 ° ° 28°°
O O Q
0 0 ° 2400-
20OO-
4600
12OO
800-
400
CO ,1
4:
o
f\
0 0
0 °0 °°°0
0 0
n 0
o
o
o
TIME (E.S.T.)
2 4 6 8 4O 42 14 16 18 20
TIME (E.S.T.)
Figure 2.4 Diurnal Variation of Pollutant Concentrations at the Southwest Corner Cell for
August 8, 1980 (JD80221)
-------
-15-
TABIE 2.5
Pollutant Ooncentrations at the Top of the Modeling Region
for August 8. 1980 (JD80221) for Base Case Q9801 and Strategies
Conoentration at the Top
Pollutant of the Modeling Region
(ppb)
Strategies
03 70 56
NO2 66
NMHC* 30 30
CO 20 20
*ppbc
-------
-16-
(BLANK PAGE)
-------
-17-
CHAPIER 3
EMISSION INVENTORIES - BASE CASE AND CONTROL STRATEGIES
The UAM version vised in this study employs the Carbon Bond II (CBII)
Chemical Mechanism to speciate the hydrocarbon emissions. The 1980 emissions
data base for the New York metropolitan area for each of the three states is
presented in Table 3.1 by source category on a tons per year basis. The
speciated emissions summary for a typical model day is provided in Table 3.2.
The details on the development of this database can be found elsewhere.
3.1 SCOPE BASE EMISSIONS
To reflect reality more accurately, changes were made to the OMNYMAP
emissions inventory. The 1988 base inventory, referred to hereafter as the
SCOPE BASE Scenario, was adjusted in the following way: The Stage II gasoline
marketing controls assumed in OMNYMAP to be in place over the New Jersey portion
of the domain were removed, since Stage II is not currently iirplemented. The
Reid Vapor Pressure (RVP) for gasoline was adjusted from 10.0 psi (assumed in
the OMNYMAP study) to 11.7 psi for all gasoline related emissions over the
domain, as this is the vapor pressure of gasoline sold in the New York
metropolitan area. Since no data were available for Publicly Owned Treatment
Works (POTWs) when the OMNYMAP study was conducted, they were not included in
the 1988 OMNYMAP emission inventory. However, for this study, information on
the emissions and locations of POTW sources was obtained from USEPA Region II
for the New Jersey, New York portions of the domain. No such data were
available for Connecticut. Assuming that these sources have a stack height less
than 65 m, they were treated as minor point sources with speciation
characteristics similar to those in the consumer/commercial solvent category.
In this manner, these data were incorporated into the 1988 SCOPE BASE inventory.
The annual emissions (tpy) by state and source category and in terms of
model day summary for the SCOPE BASE are listed in Tables 3.3 and 3.4,
respectively. Even though the inventory shows an increase in the precursor
pollutants emissions from 1980 to 1988 for the major point and area source
categories, the remaining categories, minor and mobile sources, show a greater
-------
TABLE 3.1
Summary of 1980 Emissions (OMNYMAP Base) over the Modeling Domain (Tons/Year)
NEW YORK NEW JERSEY CONNECTICUT MDDRT.TNG DOMAIN
CATEGORY VOC NO VOC NO VOC NO VOC NO
MAJOR POINT SOURCES* 61,488 957 62,349 297 26,089 1,254 149,926
MINOR POINT SOURCES 27,651 2,175 114,978 131,039 9,029 8,586 151,658 141,800
AREA SOURCES 132,883 129,390 149,288 67,716 79,002 22,551 361,173 219,657
i
oo
MOBILE SOURCES 219,482 158,297 117,035 84,124 115,317 101,167 451,834 343,588
TOTAL 380,016 351,350 382,258 345,228 203,645 158,393 965,919 854,971
*Sources with emissions greater than 100 tpy and a stack height exceeding 65m.
-------
TABLE 3.2
Sumnaxv of Tvoical Dav (0400 to 2000 hrs) Soeciated Emissions fG-Moles) fox* OMNYMAP Base
CATEGORY
MAJOR
POINT
SOURCES
NO N02
5,175,182 393,014
NO^
5,568,196
PAR OLE CARS ARO ETH VDC
68,853 1,136 42,280 368 0 112,637
MINOR
POINT 2,663,471 91,041 2,754,512 9,850,952 566,199 861,614 612,787 624,259 12,515,811
SOURCES
9,078,919 754,870 9,833,789 44,929,998 445,118 2,773,712 681,070 571,640 49,401,538
15,684,255 1,742,748 17,427,003 38,577,935 1,818,341 4,253,342 1,818,341 3,020,000 49,487,959
POTAL 32,601,827 2,981,673 35,583,500 93,427,738 2,830,794 7,930,948 3,112,566 4,215,566 111,517,945
-------
TABLE 3.3
Summary of Emissions for SCOPE BASE Over the Modeling Domain (Tons/Year)
NEW
CATEGORY VOC
MAJOR POINT SOURCES
% Change frcm 1980
MINOR POINT SOURCES 13,167
% Change frcm 1980 -52.4
AREA SOURCES 134,031
% Change frcm 1980 0.9
MOBILE SOURCES 105,438
% Change frcm 1980 -52.0
YORK NEW JERSEY
NO VOC NO
61,488 957 62,349
2,175 69,444 128,446
-39.6 -2.0
132,722 160,266 69,945
2.6 7.4 3.3
95,764 66,610 63,599
-39.5 -43.1 -24.4
CONNECTICUT
VOC
850
186.2
7,223
-20.0
79,248
0.3
57,073
-50.5
_HQX_
29,141
11.7
5,104
-40.6
23,246
3.1
64,878
-35.9
MOD^.TMG DTIMATN
VOC NO
1,807*
44.1
89,864
-40.8
373,545
3.4
229,121
-49.3
A.
152,978
2.0
135,725
-4.3
r\>
0
225,913
2.8
224,241
-34.7
TOTAL 252,636 292,149 297,277 324,339 144,394 122,369 694,307 738,857
% Change frcm 1980 -33.5 -16.8 -22.2 -6.1 29.1 -22.7 -28.1 -13.6
-------
TABLE 3.4
Summary of Typical Day (0400 to 2000 hrsl Speciated Emissions fG-Moles) for SCOPE BASE
CATEGORY
MAJOR
POINT
SOURCES
NO N02
5,398,898 419,890
NO PAR
5,818,788 87,393
OLE GARB ARO ETH VDC
4,624 38,756 656 0 131,429
MINOR
POINT 3,122,126 88,163 3,210,289 5,779,210 346,394 566,566 324,334 391,897 7,408,401
SOURCES
8,949,727 733,730 9,683,457 43,552,626 457,694 2,748,050 672,181 592,912 48,023,463 ,
|\)
I
I
10,253,727 1,139,370 11,393,097 17,453,368 969,852 2,098,413 999,702 1,768,691 23,289,954
TOTAL 27,724,478 2,381,153 30,105,631 66,872,597 1,778,564 5,451,785 1,996,873 2,753,428 78,853,247
-------
-22-
amount of reduction, resulting in a net decrease in the total emissions over the
New York metropolitan area. The reductions for the SCOPE BASE Case are in the
amount of about 28% and 14% over the domain for VOCs and NO , respectively, from
a
their 1980 levels.
3.2 SCOPE STRATEGY 1
This strategy looked at the impact of a series of motor vehicle controls.
To analyze the full effect of those controls, the modeling region was assumed to
have a fully implemented federal motor vehicle control program (FMVCP) and a
fully implemented on-board gasoline vapor recovery system with 93% control
efficiency. RVP was set at 9.0 psi. The average emission rate (g/mile) for a
future year, 2005, was estimated using the MOBHE3 model and New York's 1988
mobile source inventory. These future-year emission rates were assumed to be
reflective of a FMVCP and a fully implemented on-board gasoline vapor recovery
system and were applied to the mobile source inventory of SCOPE BASE. The net
effect of STRATEGY 1 was a reduction of about 44% in the VOCs and 23% in the NC»x
from their corresponding 1980 levels. The annual emissions (tpy) for the SCOPE
STRATEGY 1 are listed in Tables 3.5 and 3.6 by state and source category, and in
terms of model day summary, respectively.
3.3 SCOPE STRATEGY 2
The second control strategy analyzed the effect of reduction in emissions
from certain categories of organic solvents in conjunction with the measures
imposed in SCOPE STRATEGY 1 on the mobile source category. Again, modeling from
controls incrementally to Strategy 1 allows assessment of both the relative
effect of Strategy 2 controls and the combined effectiveness of Strategies 1
and 2.
The SCOPE STRATEGY 1 emissions inventory was adjusted to reflect a
specified level of reduction from each of the following sources: a 50%
reduction in (a) ronsumer/conmercial solvents, (b) auto refinishing and (c)
POTWs; and a 65% reduction in the categories of (i) architectural surface
coatings (oil-based) and (ii) traffic marking coatings. Several other source
types that were considered for inclusion under this strategy were hazardous
-------
TABLE 3.5
Summary of Emissions for SCOPE STRATEGY 1 over the Modeling Domain (Tons/Year)
NEW YORK NEW JERSEY CXMJECnCUT
CATEGORY
MAJOR POINT SOURCES
% Change frcm 1980
MINOR POINT SOURCES
% Change from 1980
AREA SOURCES
% Change frcm 1980
MOBIIE SOURCES
% Change from 1980
VOC NO VOC NO VOC
61,488 957 62,349 850
186.2
13,167 2,175 69,444 128,446 7,223
-52.4 0.0 -39.6 -2.0 -20.0
116,948 132,722 143,727 69,945 73,227
-12.0 2.6 -3.7 3.3 -7.3
53,765 60,063 33,966 39,864 29,102
-75.5 -62.1 -71.0 -52.6 -74.8
1K>
X
29,141
11.7
5,104
-40.6
23,246
3.1
40,691
-59.8
MODELING DOMAIN
VOC
1,807
44.1
89,864
-40.8
333,902
7.6
116,833
-74.1
NO
X
152,978
2.0
135,725
-4.3
225,913
2.8
140,618
-59.1
TOTAL 183,880 256,448 248,094 300,604 110,402 98,182 542,376 655,234
% Change from 1980 -51.6 -27.0 -35.1 -12.9 -45.8 -38.0 -43.8 -23.4
ro
co
i
-------
TABLE 3.6
Summary of Typical Day (0400 to 2000 hrs) Speciated Emissions (G-Molesl for SCOPE STRATEGY 1
CATEGORY NO NO NOX PAR PIE GARB ARO ElH VOC
MAJOR
POINT 5,398,898 419,890 5,818,788 87,393 4,624 38,756 656 0 131,429
SOURCES
MINOR
POINT 3,122,126 88,163 3,210,289 5,779,210 346,394 566,566 324,334 391,897 7,408,401
SOURCES
8,949,727 733,730 9,683,457 37,721,815 402,426 2,601,997 663,132 592,912 41,982,282 i
ro
«P*
6,426,606 707,114 7,136,720 8,746,644 507,561 1,114,206 499,787 948,584 11,816,782
TOTAL 23,897,357 1,951,897 25,849,254 52,335,062 1,261,005 4,321,525 1,487,909 1,933,393 61,338,894
-------
-25-
waste treatment, storage and disposal facilities (TSDF's), coke ovens, wood
furniture refinishing, and web offset lithography. However, either the
emissions data were not available or the source categories were not identified
in the modeling domain and, thus, were not included in this analysis. The above
adjustments were applied to the SCOPE STRATEGY 1 inventory, and the resulting
inventory, SCOPE STRATEGY 2 is shown on an annual basis (tpy) and in terms of
speciated summary of a typical model day emissions in Tables 3.7 and 3.8,
respectively. Under this strategy, the overall emission reductions in VDCs is
about 50% from ***»ir 1980 levels with substantial emissions reductions coming
from the New York portion of the modeling domain.
3.4 SCOPE STRATEGY 3
Under this emission control strategy in addition to the controls in
Strategies 1 .and 2, 30% of the light-duty gasoline powered vehicle population
was assumed to be fueled with methanol. Based upon consultations with the EPA
Office of Mobile Sources, Ann Arbor, MI, the estimated reduction in emissions
was 36% and 92% from exhaust and evaporative hydrocarbons, respectively, for
100% methanol-fueled versus gasoline-fueled vehicles. It should be noted here
that the current version of the UAM utilizes the CBII mechanism which does not
explicitly treat formaldehyde (HCHO) emissions, unlike the Carbon Bond IV (CBIV)
or other chemical mechanisms. Thus, under this scenario no changes were made to
the NO or CO emissions inventories and to the speciation characteristics of the
X
VOC's resulting from the penetration of methanol-fueled vehicles into the fleet.
Performing appropriate adjustments to the mobile source emissions in SCOPE
STRATEGY 2, the inventory was prepared and summarized on an annual basis (tpy)
and on a model day basis in Tables 3.9 at 3.10, respectively.
3.5 SCOPE STRATEGY 4
This strategy assumed a stationary source NO "RACT" rule was in place and
all non-mobile NO emissions were reduced by 40% from the SCOPE STRATEGY 3
emissions inventory, resulting in an overall reduction of 47% from the 1980 NO
emissions (See Tables 3.11 and 3.12). This is due to the fact that 59% of the
reductions in NO were achieved from the mobile source category itself under
SCOPE STRATEGY 1.
-------
TABLE 3.7
Summary of Emissions for SCOPE STRATEGY 2 over the Modeling Domain (Tons/Year)
NEW YORK NEW JERSEY (XJNNECnCUT
CATEGORY
MAJOR POINT SOURCES
% Change from 1980
MINOR POINT SOURCES
% Change fron 1980
AREA SOURCES
% Change from 1980
MOBILE SOURCES
% Change frcro 1980
TOTAL
% Change from 1980
VOC NO VOC NO VOC
61,488 957 62,349 850
186.2
13,167 2,175 69,444 128,446 7,223
-52.4 0.0 -39.6 -2.0 -20.0
85,550 132,722 124,388 69,945 63,200
-35.6 2.6 -16.7 3.3 -20.0
53,765 60,063 33,966 39,864 29,102
-75.5 -62.1 -71.0 -52.6 -74.8
152,482 256,448 228,755 300,604 100,375
-59.9 -27.0 -40.2 -12.9 -50.7
NO
29,141
11.7
5,104
-40.6
23,246
3.1
40,691
-59.8
98,182
-38.0
MODELING DOMAIN
VOC
1,807'
44.1
89,864
-40.8
273,138
24.4
116,833
-74.1
481,612
-50.1
NO
X
152,978
2.0
135,725
-4.3
225,913
2.8
140,618
-59.1
655,234
-23.4
I
ro
at
i
-------
TABIE 3.8
Sumnarv of Typical Day (0400 to 2000 tors) Sueciated Emissions fG-Moles) for SCOPE STRATEGY 2
CATEGORY NO NO2 NO^ PAR PIE GARB ARO ETH VOC
MAJOR
POINT 5,398,898 419,890 5,818,788 87,393 4,624 38,756 656 0 131,429
SOURCES
MINOR
POINT 3,122,126 88,163 3,210,289 5,779,210 346,394 566,566 324,334 391,897 7,408,401
SOURCES
AREA
^^ 8,949,727 733,730 9,683,457 29,660,285 402,426 1,956,857 550,077 592,912 33,162,557 ,
ro
6,426,606 710,114 7,136,720 8,746,644 507,561 1,114,206 499,787 948,584 11,816,782
TOTAL 23,897,357 1,951,897 25,849,254 44,273,532 1,261,005 3,676,385 1,374,854 1,933,393 52,519,169
-------
TABLE 3.9
Summary of Emissions for SCOPE STRATEGY 3 over the Model incr Domain (Tons/Year)
NEW YORK NEW JERSEY CONNECTICUT
CATEGORY
MAJOR POINT SOURCES
% Change from 1980
MINOR POINT SOURCES
% Change frcm 1980
AREA SOURCES
% Change frcm 1980
MOBILE SOURCES
% Change frcm 1980
VOC NO VOC NO VOC
iC" A
61,488 957 62,349 850
186.2
13,167 2,175 69,444 128,446 7,223
-52.4 -39.6 -2.0 -20.0
85,550 132,722 124,388 69,945 63,200
-35.6 2.6 -16.7 3.3 -20.0
42,811 60,063 27,046 39,864 23,173
-80.5 -62.1 -76.9 -52.6 -79.9
29,141
11.7
5,104
-40.6
23,246
3.1
40,691
-59.8
MODELING DOMAIN
VOC NO
1,807
44.1
89,864
-40.8
273,138
24.4
93,029
-79.4
X.
152,978
2.0
135,725
-4.3
225,913
2.8
140,618
-59.1
i
ro
00
TOTAL 141,528 256,448 221,835 300,604 94,446 98,182 457,808 655,234
% Change from 1980 -62.8 -27.0 -42.0 -12.9 -53.6 -38.0 -52.6 -23.4
-------
TABLE 3.10
Suntmarv of Tvoical Dav (0400 to 2000 hrsl Sneciated Emissions (G-Moles) for SCOPE STRATEGY 3
CATEGORY
MAJOR
POINT
SOURCES
MINOR
POINT
SOURCES
AREA
SOURCES
MOBILE
SOURCES
NO N02 NO PAR OLE GARB ARD ETH VOC
5,398,898 419,890 5,818,788 87,393 4,624 38,756 656 0 131,429
/
3,122,126 88,163 3,210,289 5,779,210 346,394 566,566 324,334 391,897 7,408,401
8,949,727 733,730 9,683,457 29,660,285 402,426 1,956,857 550,077 592,912 33,162,557 ,
'ro
vo
6,426,606 710,114 7,136,720 6,946,396 339,689 883,891 392,652 752,121 9,374,749
TOTAL
23,897,357 1,951,897 25,849,254 42,473,284 1,153,133 3,446,070 1,267,719 1,736,930 50,077,136
-------
TABLE 3.11
Summary of Emissions for SOOPE STRATEGY 4 over the Modeling Domain (Tons/Year)
CATEGORY
MAJOR POINT SOURCES
% Change from 1980
MINOR POINT SOURCES
% Change frcm 1980
AREA SOURCES
% Change frcm 1980
MOBILE SOURCES
% Change from 1980
TOTAL
% Change frcm 1980
NEW
VOC
-,-__,_,
13,167
-52.4
85,550
-35.6
42,811
-80.5
141,528
-62.8
YORK
NO
X
36,893
-40.0
1,305
-40.0
79,633
-38.5
60,063
-62.1
177,894
-49.4
NEW JERSEY
VOC NO
X
957 37,409
-40.0
69,444 77,068
-39.6 -41.2
124,388 41,967
-16.7 -38.0
27,046 39,864
-76.9 -52.6
221,835 196,308
-42.0 -43.1
CONNECTICUT
VOC
850
186.2
7,223
-20.0
63,200
-20.0
23,173
-79.9
94,446
-53.6
NO
X
17,485
33.0
3,062
-64.3
13,948
-38.2
40,691
-59.8
75,186,
-52.5
MODET.TN
VOC
1,807
44.1
89,834
-40.8
273,138
-24.4
93,029
-79.4
457,808
-52.6
fG DOMAIN
-NQx-
91,787
-38.8
81,435
-42.6 ,
CO
o
135,548
-38.3
140,618
-59.1
449,388
-47.4
-------
TABLE 3.12
Suronarv of Typical Dav (0400 to 2000 hrsl Speciated Emissions fG-Moles) for SCOPE STRATEGY 4
CATEGORY NO N02 NO^ PAR OLE GARB ARO ETH VDC
MAJOR
POINT 3,239.339 251,934 3,491,273 87,393 4,624 38,756 656 0 131,429
SOURCES
MINOR
POINT 1,873,276 52,898 1,926,173 5,779,210 346,394 566,566 324,334 391,897 7,408,401
SOURCES
5,369,836 440,238 5,810,074 29,660,285 402,426 1,956,857 550,077 592,912 33,162,557 ^
i
6,426,606 710,114 7,136,720 6,946,396 339,689 883,891 392,652 752,121 9,374,749
TOTAL 16,909,057 1,455,184 18,364,240 42,473,284 1,153,133 3,446,070 1,267,719 1,736,930 50,077,136
-------
,-32-
3.6 SCOPE STRATEGY 5
Strategy 5 was aimed at determining the level of emissions reductions
required to meet the ozone NAAQS in the region. This calls for an "educated
guess" of the reduction levels needed in the precursor emissions. Based upon
several sensitivity analyses performed with UAM, the required level of reduction
in VOCs was estimated to be 95% from the 1980 level within the domain with no
change in the NO level from the base year. It should be noted that these are
X
across-the-board reductions and are not source category selective as in the
previous control strategies. In the case of initial and boundary concentration
fields, no changes were made to the levels of GO, NO and NO- from these 1980
levels, while those of NMOC and ozone were reduced by 80% and 40%, respectively.
3.7 SYNOPSIS OF THE SCOPE STRATEGIES
The proposed emission control strategy scenarios, listed in Table 3.13,
were assembled to examine their effects on the levels of ozone in the New York
metropolitan area. The first three strategies were incremental in nature and
were designed to assess improvement in the ozone air quality in relation to a
specific variety of controls imposed upon the VOC emissions. Figure 3.1 shows a
summary view of these changes on a state-by-state basis and over the domain.
The fourth strategy was aimed at examining NO controls; a summary of these
a
reductions is shown in Figure 3.2. Finally, a strategy with an across-the-board
reduction in the VOC emissions was evaluated to bring the peak ozone
concentrations in the New York metropolitan area to the level of the ozone
NAAQS.
-------
-33-
TABIE 3.13
Summary of the Emission Control Strategies
Considered for the New York Metropolitan Area
Strategy
Type of Controls
Percent Change
from 1980 Emissions
VDC NO
SCOPE BASE
1988 Base Case
KVP set at 11.7 psi
No Stage II emissions from POIWs
included
28
14
SCOPE STRATEGY 1
Mobile Control
Measures
RVP set at 9.0 psi
Fully implemented on-board gasoline
vapor recovery and JMVCP
44
23
SCOPE STRATEGY 2
Control Technology
Measures
50% reduction in commercial/consumer
solvents and auto-refinishing emission
60% reduction in architechtural,
surface coating and traffic marking
coating emissions
50
23
SCOPE STRATEGY 3
Methanol Option
30% penetration of methanol fueled
vehicles in the light duty gasoline
vehicle fleet
52
23
SCOPE STRATEGY 4
NO RACT
Non-mobile NO emission reduced by 40% 52
47
SCOPE STRATEGY 5
What if?
Across-the-board reductions
95
-------
VOC EMISSIONS
1980 Base
emissionsttons/yr) 380,016
382,258
203,645
965,919
60
50-
Percentage
Reduction
40
301
20^
10-
0
SCOPE BASE
SCOPE STRATEGY 1
SCOPE STRATEGY 2
SCOPE STRATEGY 3
NEW YORK NEW JERSEY CONNECTICUT MODELING
DOMAIN
Figure 3.1 Sunroary of Percentage of VOC Reduction in the Tri-State Region of the Modeling Domain
CO
JS.
-------
NOX EMISSIONS
1980 Base
Emissions (tons/yr) 351,350
345,228 158,393 854,971
Percentage
Reduction
60-
50-
40-
30-
20-
10-
0
NEW
YORK
V777Z\ SCOPE BASE
SCOPE STRATEGY 1
SCOPE STRATEGY 4
NEW CONNECT. MODELING
JERSEY DOMAIN
CO
en
Figure 3.2 Summary of Percentage of IIOX Reduction In the Tri-State Region of the Modeling Domain
-------
-36-
(BLANK PAGE)
-------
-37-
CHAPTER 4
MODELING RESUUS
The UAM simulations of anibient ozone air quality were performed for the
emissions inventories assembled with the August 8, 1980 meteorological
conditions and appropriately adjusted initial and boundary concentration fields.
The results of the HAM simulations are presented and discussed in this section.
4.1 BASE CASE AND STRATEGY SIMULATIONS
As noted earlier, the boundary concentrations of the precursors with the
exception of SCOPE STRATEGY 5 were obtained by scaling the 1980 boundary
concentration values with a factor consistent with the emission reduction from
the 1980 level. For example, in the case of SCOPE BASE simulation, the VOC and
NO concentrations were reduced by about 28% and 14%, respectively, from their
X
1980 levels while the the 0 concentration was reduced by 20% from its 1980
level. The diurnal variation of the pollutant concentrations at the southwest
corner cell are shown in Figure 4.1 for all the strategies with the exception of
SCOPE STRATEGY 5. The results of the UAM simulations for the 1980 and 1988 base
cases and for each of the strategies are presented in Figures 4.2 through 4.8.
The 1980 OMNYMAP base case simulation, shown in Figure 4.2, has a double
peak oriented in a southwest-northeast direction with, high ozone concentrations
extending from the northeastern New Jersey-New York area to central Connecticut.
The distinct double peak structure present in the early afternoon hours merges
into a single peak over the northeastern New Jersey-New York area as the day
progresses. While the measured maximum of 246 ppb for this day over the domain
was at Stratford, CT at 1300 hrs, the predicted maximum of 246 ppb occurred at
1600 hrs in the vicinity of New Haven, CT.
The 1980 SCOPE BASE simulation, shown in Figure 4.3, reveals essentially
the same features as those found for the 1980 OMNYMAP case except for a peak
value of 205 ppb over Connecticut. This reduction of about 17% in the predicted
peak ozone concentration from the base case corresponds to a decrease of 28% and
14% in VOCs and NO emissions, respectively, from their 1980 levels. Also,
-------
-38-
1000-
900'
BOO'
§ TOG-
'S
£ 50O
I 400-
G
300-
200-
100-
o 1980 BASE
SCOPE BASE
0 o ° D SCOPE STRG 1
0 A SCOPE STRG 2
o
°°°
O o
.. o o o «
° ° .
DDD*» .*** 2.
^^A« * 0EOE"
A^nDDDAAAA D
AAAAA A
4 6 8 10 12 14 16 18 20
HOUR
100-
90-
80-
1" TO-
SS.
5
8 60-
|
^ *°"
g
C 40-
-------
AREAL OSTRiaUTON OF OZONE
GREATER THAN 123 PP8 ft 60O TOR AUG. 8. I960
a
a
W It M . f»
X-AXIS
AREAL OSTRISUTON OF OZONE
GREATER THAN <2S PP8 O I7OO FOR AUG. 8.1980
II
9
i
I » « it ao
X-AXIS
AREAL OSTRIBUTON OF OZONE
GREATER THAN CS PPB C 1600 FOR AUG. 8.1960
It
w
I
X-AXIS
M
M
X
W
NEW YORK |
CONNCCTICUT
NEW
JERSEY /
10 « zo
X-AXIS
CO
us
Figure 4.2 Areal Distribution of Ozone for OHIYMAP BASE 1980 Simulation
-------
AREAL aSTRIBUTON OF OZONE
GREATER THAN 125 PP8 O t5OO TOR SCOPE BASE
AREAL OISTRIBUTDN OF OZONE
GREATER THAN C9 PPB O I60O fOR SCOPE BASE
!
X-AXIS
AREAL nSTRiaiTON OF OZONE
GREATER THAN C5 PPB O 1700 FOR SCOPE BASE
It
I W W
X-AXIS
I* 20
X-AXIS
I
*>
O
18 »
Figure 4.3 Areal Distribution of Ozone for SCOPE BASE 1988 Sioulation
-------
-41-
there is a general decrease in the areal extent of the concentrations exceeding
the NAAQS level of 0.12 ppm, from the 1980 base level.
The ozone distribution resulting from SCOPE STRATEGY 1, which consists
mainly of lower gasoline RVP along with fully implemented federal motor vehicle
control programs, is shown in Figure 4.4. Wider these controls, the total VOC
reduction was 44% from the 1980 level with no change in the NO emissions from
a
the SCOPE BASE level. The ozone peak over Connecticut has decreased to 171 ppb
or about a 22% reduction from the 1980 level. The peak over New Jersey-New York
is also found to be reduced by about 18% from its 1980 level.
Figure 4.5 shows the ozone distribution resulting from the iirposition of
Control Technology Guidance (CTG) measures in addition to those of SCOPE
STRATEGY 1. The incremental change from SCOPE STRATEGY 1 in the predicted peak
ozone concentration over Connecticut is about 6% corresponding to a VOC
reduction of approximately the same percentage. The incremental decrease in the
peak occurring over New Jersey-New York is about 11%. There is an overall
decrease in the areal extent exceeding the NAAQS level for ozone from the
previous strategy.
In SCOPE STRATEGY 3, in addition to the above VOC and NO changes, the VOC
H
emissions were adjusted for a 30% methanol-fueled auto fleet. This resulted in
an incremental reduction of the total VOCs by about 2%, or a reduction of 52.6%
from their 1980 level with NO emissions remaining at the SCOPE STRATEGY 2,
Jx
level. The incremental effect of these emission reductions on the peak ozone
level, shown in Figure 4.6, over Connecticut when compared with SCOPE
STRATEGY 2, is a reduction of only 3 ppb. However, in relation to the 1980
level, the reduction in the peak ozone level is about 28% as compared with 27%
for SCOPE STRATEGY 2.
789
Recent modeling studies ' ' utilizing chemical mechanisms which explicitly
treat emissions of formaldehyde from methanol-fueled vehicles (MFV) suggest that
there is a decrease in the levels of ozone when compared with conventionally
fueled vehicles (CFV). In these studies, there were reductions in the NO ad CO
H
emissions along with the VOC reductions due to penetration of MFV into the auto
fleet. However, the predicted decreases in the ambient ozone concentrations are
-------
AREAL DISTRIBUTION OF OZONE
GREATER THAN C5 PPB O BOO FOR SCOPE STRG (
to
X-AXIS
AREAL OSTRiairCN OF OZONE
GREATER THAN 125 PPB O 1700 FOR SCOPE STRG I
II
it w it » »
X-AXIS
AREAL OSTRIBUTON OF OZONE
GREATER THAN 125 PPB O 1600 FOR SCOPE STRGl
a
S
II
I I » M tt
X-AXIS
tn
2
M
-fs.
ro
Figure 4.4 Areal Distribution of Ozone for SCOPE STRATEGY 1 Sioulatlon
-------
AREAL DISTRIBUTION OF OZONE
GREATER THAN 125 PPfl O I50O FDR SCOPE STRG 2
IB
M
125
n M 10 to w
X-AXIS
AREAL DISTRIBUTION OF OZONE
GREATER THAN 125 PPB O I70O FOR SCOPE STRG 2
It
» to
X-AXIS
AREAL DISTRIBUTION OF OZONE
GREATER THAN 125 PP8 O 1600 FOR SCOPE STRG 2
!
I
I
to
NCW YORK I
CONNECTICUT
ID
X>
CO
X-AXIS
Figure 4.5 Area! Distribution of Ozone for SCOPE STRATEGY 2 Simulation
-------
MEAL DISTRIBUTION OF OZONE
GREATER THAN 125 PPB £ 600 FDR SCOPE STRG3
to
19
I
«0
I « it to » :
X-AXIS
AREAL DISTRIBUTION OF OZONE
GREATER THAN CS PPB O 1700 FOR SCOPE STRG3
»O
M> 19 20 29
X-AXIS
AREAL OISTRIBUTDN OF OZONE
GREATER THAN 125 PPB O WOO FDR SCOPE STRG3
19
10
19
I
>
to
10 IB M
X-AXIS
25
20
19
in
'x
4
to
NEW YORK
CONNECTICUT
IO 15 20 2S
X-AXIS
JO
Figure 4.6 Area! Distribution of Ozone for SCOPE STRATEGY 3 Simulation
-------
-45-
quite sensitive to the assumed initial and boundary conditions and to the amount
of formaldehyde emitted by the MFV. For example, with no emission of
formaldehyde from MFV the reduction in the ozone levels ranged from 1% to 36%,
while with a MFV exhaust consisting of 10% formaldehyde, the reduction in ozone
was predicted to be in the range of only 0% to 13%.
To assess the changes expected from reductions in NO emissions on the
Ji
ozone levels, all NO sources with the exception of the Mobile Source Category
X
were reduced by 40% in SCOPE STRATEGY 4. The results of this simulation are
shown in Figure 4.7. Even though the peak values over Connecticut and
New Jersey-New York area show reductions of about 7 to 8 ppb from SCOPE STRATEGY
3, there is an increase in the concentration levels in the New YorkConnecticut
corridor of White Plains, Greenwich, and Bridgeport by as much as 15 to 20 ppb.
4.2 STRATEGY TO REDUCE OZONE LEVELS TO THE NAAQS - SCOPE STRATEGY 5
In the above simulations, the various emission control strategies
considered were aimed toward reducing the peak ozone concentration to the level
of the ozone NAAQS over the domain. The UAM results reveal that even with a
projected reduction of 53% and 47% in the precursor emissions of VOCs and NO
X
from the 1980 levels, the peak ozone level over Connecticut can be reduced by
only 32% from its 1980 level; the peak predicted concentration over the modeling
domain is still well above the level of the ozone NAAQS. Hence, model
sensitivity simulations were needed to assess the level (s) of emission reduction
required to reduce the peak ozone concentration over the region to the level of
the NAAQS for ozone.
These sensitivity simulations included (a) no emissions, (b) clean influx,
and (c) clean initial conditions. For these simulations, the SCOPE STRATEGY 3
data base was utilized along with the pollutant concentrations listed in Table
4.1 to represent "CLEAN" conditions. The ozone distribution resulting from the
simulations are shown in Figure 4.8.
The simulation with "no emissions" in the domain indicates that there are
two localized areas, one exceeding the NAAQS over New Jersey-New York area, and
the other approaching the NAAQS level in the northeastern part of Connecticut
-------
AREAL DISTRIBUTON OF OZONE
GREATER THAN 125 PPB ft 600 FOR SCOPE STRG4
to
« M 10. » :
X-AXIS
AREAL DISTRIBUTION OF OZONE
GREATER THAN <2S PPB ft 1700 FOR SCOPE STRG4
w ao
X-AXIS
AREAL DISTRIBUTION OF OZONE
GREATER THAN 125 PPB ft I6OO FOR SCOPE STRG4
<6
M
I « M »
X-AXIS
ZO
19
ui
x
It W YORK |
CONNECTICUT
10
2O
X)
X-AXIS
Figure 4.7 Area! Distribution of Ozone for SCOPE STRATEGY 4 Simulation
-------
-47-
TABLE 4.1
Pollutant Concentrations for "Clean" Conditions
POLLUTANT CONCENTRATION (POb)
o3 o.i
N02 2.0
NO 1.0
NMHC* 5.0
CO 20.0
*ppbc
-------
AREAL DBTRieUTON OF OZONE
CHEATER THAN 60 ffO ft tfcOO FDR SCOPE STRG3
I
It
I
X-AXIS
AREAL OSTMBUTDN OF OZOtf _
CHEATER THAM 80 H>8 ft 1600 fDH SCOPE STRC J
andlllim
I
n w
ANEAL ttSTRlBUTDN OF O2OK _
GREATER THAN BO PP8 ft ttOO TOR SCOPE STRC)*
»
1
I
K-ANM
to
I
>
NEW VOMK
CONNCCTICUT
JERSEY
I » K>
-A»S
n
-AXIS
to n to
Figure 4.8 Areal Distribution of Ozone Under Different Sensitivity Conditions for
SCOPE STRATEGY 3
-------
-49-
indicating the effect of the transport of ozone and its precursors into the
domain as well as the influence of the initial concentrations at the start of
the simulation. The simulation with no transport ("clean influx") shows ozone
exceedances over the northeastern portion of Connecticut with the remaining area
of the modeling domain well below the NAAQS, while the simulation with "clean
initial conditions" shows a pattern very similar to the SCOPE STRATEGY 3 (see
Figure 4.6) but with a reduced level of peak concentration over Connecticut.
From these simulations, it is evident that either the emissions in the domain by
themselves with no influx of pollutants or no emissions in the domain but with
influx of pollutants could lead to exceedance of the ozone NAAQS in the region.
Therefore, both precursor emissions within the domain as well as influx of ozone
and its precursors into the region need to be reduced in order to attain the
ozone NAAQS over the New York metropolitan area.
Based upon this premise, the next simulation, SCOPE STRATEGY 5A, was set
with the conditions listed in Table 4.2; the resulting UAM prediction is shown
in Figure 4.9a. With the exception of the northeastern portion of Connecticut,
the remaining portions of the domain are well below the NAAQS level. Thus, it
appears from this simulation that a fine tuning of the emissions may bring the
entire domain to within the NAAQS level for ozone.
Thus, in SCOPE STRATEGY 5b, the VOCs were reduced further by another 15% or
a total of 95% from the 1980 level with other conditions set similar to those of
SCOPE STRATEGY 5a. The modeling conditions for this case are listed in
Table 4.3. The ozone distribution resulting from this simulation, shown in
Figure 4.9b, indicates that the entire modeling domain is below the NAAQS level,
with a peak value of 122 ppb occurring in the northeast corner of the State of
Connecticut.
4.3 DISCUSSION
In this study, four emission control strategies were investigated to assess
their role in the reduction of the ozone levels over the New York metropolitan
area. The strategies that were considered were designed incrementally in order
to provide information regarding the effects of each of the control strategies
on the ozone levels on a relative basis. Figure 4.10 shows a diurnal plot of
the number of cells exceeding 125 ppb of ozone for the four SCOPE STRATEGIES as
-------
-50-
TABLE 4.2
Conditions for SCOPE STRATEGY 5A
EMISSIONS: 1980 VOCs REDUCED BY 80% ACROSS-THE-BOARD
1980 NO UNCHANGED
X
INITIAL CONDITIONS: CO, NO, NO - NO CHANGE FROM 1980
NMHC REDUCED BY 80% FROM 1980 LEVEL
OZONE REDUCED BY 40% FROM 1980 LEVEL
BOUNDARY CONDITIONS AT THE SURFACE LAYER:
POIUJTANT INFLUX NOT TO EXCEED: OZONE = 58 ppb, NO = 27 ppb
NMHC = 176 ppbc, NO = 13 ppb,
CO = 2300 ppb
CONCENTRATIONS AT THE REGION TOP:
OZONE = 40 ppb NO = 6 ppb
NMHC = 30 PPBC NO = 3 ppb, CO= 20 ppb
-------
-51-
TABLE 4.3
Conditions for SCOPE STRATEGY 5B
EMISSIONS: 1980 VOCs REDUCED BY 95% ACROSS-THE-BOARD
1980 NO UNCHANGED
INITIAL CONDITIONS: NMHC REDUCED BY 95% FROM 1980 LEVEL
OZONE, CO, NO, NO, SAME AS SCOPE STRATEGY-5A
ft
BOUNDARY CONDITIONS: SAME AS SCOPE STRATEGY-5A
CONCENTRATIONS AT THE REGION TOP:
SAME AS SCOPE STRATEGY-5A
-------
AREAL DJSTRIOJTBN OF OZONE
GREATER THAN 70 PPB G I60O FOR SCOPE STRG So*
AREAL DISTRIBUTION OF OZONE
GREATER THAN 7O PPB O> I6OO FOR SCOPE STRG 5b*
3
i
tft
to
29
M
3
I
19
to
« With 80% reduction in
VOtt faom 1960 to* K/K
Mduccd oppnifiriolcty
122
100-
VOC* Imm 1960 teMl
BCMnttwSTRGSa
Figure 4.9a
to w
V-AXIS
Figure 4.95
to t§
X-AXIS
X
I
19
10
NEW YORK
CONNECTICUT
NEW /
JERSEY
Figure 4.9a Area! Distribution of Ozone for
SCOPE STRATEGY 5a Simulation
Figure 4.9b Areal Distribution of Ozone for
SCOPE STRATEGY 5b Simulation
«0 I* 2O 25 M
X-AXIS
-------
CONNECTICUT (Total No. of Cells 208)
160-
- 120r
A
wt
*0>
0 80-
o
. 40-
o
z
0
A
B
__
C
D
E
13
DOMAIN
320-
280-
240-
m 2OO-
0 160-
H-
o
O *£fJ"
2
80-
JB /«V
40-
n
A
B
C
a.
A
__.
B
C
D
E
,
F
14
^^^^ ^^"^^
A
I
B
C
D
E
F
A
"~"-'~
B
1
C
D
E
F
A
B
C
D
E
F
15 16 17
A - OMNYMAP BASE 1980
( Total No. of Cells 667 } g - |COPE BASE (1988
0
E
^l
A
B
MHH
C
D
E
F
A
B
C
D
E
F
A
-^MH
B
C
D
E
F
A
D - SCOPE STRG 2
E - SCOPE STRG 3
F - SCOPE STRG 4
B
wn
C
D
E
F
13
14
16
17
15
HOUR
Figure 4.10 Number of Cells Exceeding the Ozone NAAQS Level Under Each UAM Simulation
-------
-54-
well as the 1980 and 1988 Base cases. The range of iinprovement in terms of the
decrease in the number of grid cells exceeding 125 ppb over Connecticut varies
from 37% to 60% depending upon the hour in consideration. Given the same
meteorological conditions for all the simulations, the results indicate that the
maximum areal extent exceeding 125 ppb was during 1700-1800 hrs, while the peak
ozone concentration over the domain was found to occur at 1600-1700 hrs. The
effects of these strategies on the ozone peak over Connecticut are summarized in
Table 4.4.
In Figure 4.11, the percent change in the predicted ozone level as a
function of reduction in the VDCs resulting from the strategies evaluated in
this study is shown for a monitoring location (Bridgeport/Stratford) as well as
for the peak concentration over the Connecticut region. With increasing
reduction in VOCs, the predicted improvement in the ozone concentration level at
a given location, for example, Bridgeport/Stratford, is significantly greater
than what would result if the predicted peak value over Connecticut is
considered. At 80% VOCs reduction, the concentration at Bridgeport/Stratford
was reduced from 236 ppb to 31 ppb or a decrease of 88% while the peak predicted
ozone concentration over Connecticut decreased from 246 ppb to 155 ppb, or a
reduction of only 37%. These results suggest that the relative merit of the
emission control plans in reducing the ambient ozone concentrations must be
evaluated in terms of peak ozone concentration over the domain as opposed to
concentration changes at a given receptor location. These simulations
demonstrate that the DM is a very useful tool for evaluating the spatial and
temporal characteristics of the ozone concentrations as a function of the
precursor emissions levels.
-------
-55-
TABIE 4.4
Peak Ozone Level Over Connecticut Under Various Strategies
for a Selected Meteorological Scenario
Ozone Percent Change
Strategy Concentration fppb) from 1980 Base
1980 Base Case
1988 Scope Base
Scope Strategy 1
Scope Strategy 2
Scope Strategy 3
Scope Strategy 4
Scope Strategy 5
245
205
191
179
176
168
122
-
16
22
27
28
31
50
-------
-56-
o
00
90
80
GJ 70
tJ
I
60
50
O
K
O
LJ
Ct
UJ
30-
20-
1980 Predicted Peak Concentration
Over Connecticut 246 ppb
1980 Predicted Maximum Concentration
at Bridgeport/Stratford 236 ppb
A
O
o
.j A. Bridgeport/Stratford Ct.
5-10-1 O Peok over Ct.
10 20 30 40 50 60 70 80 90 100
PERCENT REDUCTION IN THE VOCsFROM 1980 OVER THE DOMAIN
Figure 4.11 Percentage Reduction In the Predicted Ozone Concentrations
Associated with the VOC and NOX Emission Control Strategies
Evaluated in this Study.
-------
-57-
CHAPTER 5
SUMMARY AND RECCMMENDATIQNS
In this study, four emission control strategy simulations were performed
with UAM utilizing the aerometric data for one of the high ozone days in 1980 to
examine their effectiveness in meeting the ozone NAAQS over the New York
metropolitan area. These control strategies, which were incremental in design,
were aimed at reducing VOC emissions from specific source categories. Given the
VDC emission reductions in the range of 28 to 53% from their 1980 level for
these strategies, the UAM simulations show a decrease in the peak ozone level
over Connecticut of 18 to 28% from its 1980 level. In addition, UAM predicted
an overall improvement in the areal extent of the modeling region exceeding the
ozone NAAQS level of 0.12 ppm. However, it was found from a one-day simulation
of these strategies that the peak ozone concentration in the New York
metropolitan area is still well above the level of the ozone NAAQS.
Given the aerometric conditions prevailing on August 8, 1980, Strategy 5
was aimed at assessing the level of reduction in precursor emissions required
for the region to be at or below the ozone NAAQS level. Strategy 5 simulation
results indicate that a reduction of 95% in the VOCs from their 1980 levels over
the domain together with 80% reduction in the concentrations, of ozone precursors
at the upwind boundary can bring the modeling domain to the level of the ozone
NAAQS. It should be noted that these VDC reductions are across-the-board and do
not reflect technology-based or source-specific type controls. Thus, additional
modeling analyses of innovative VOC emission controls both within and outside
the domain are clearly needed to demonstrate the attainment of the ozone NAAQS
over the New York metropolitan area.
In this study, the usefulness of a grid model such as UAM in simulating
spatial distribution of the ozone concentrations is demonstrated by
consideration of the predicted peak over the region of interest versus
concentrations at a fixed monitoring location. The results show that given 80%
reduction in the VDCs from the 1980 level there is a decrease in the predicted
concentration by 80% at a receptor location whereas there is a decrease of only
37% in the peak ozone concentration from the 1980 ozone level over the modeling
domain.
-------
58-
It should be noted that in this study severed assumptions regarding the
chemical mechanism, speciation characteristics, upwind emissions, and future
levels of concentrations of both the precursors and ozone were invoked. Also,
it should be recognized that although the emissions inventory assembled for this
study is based upon best available information, there is an uncertainty
associated with the accuracy of the projected emissions. Rirther, the version
of the UftM used in this study employs the CBII chemical mechanism, and the
emissions from the point source were assumed to be uniformly distributed in the
cell at plume height. An improved version of the UAM which includes the CBIV
chemical mechanism and treats the point source emissions and the advection
process in a more realistic manner must be considered for enhancing the
scientific credibility of the analysis of the relative merit of the various
emission control options in reducing the ambient ozone concentrations. Also,
model-nesting, for example - interfacing the urban-scale model with the
regional-scale model - needs to be explored for developing the detailed input
data bases required for the UAM simulation of the emission control strategies.
In addition, model simulations for the other high ozone days in 1980 must be
performed to determine whether the above control strategies can reduce the peak
ozone concentration in the New York metropolitan area to the level of the NAAQS
and to identify the strategies for achieving compliance with the ozone NAAQS.
-------
-59-
References
1. Urban Ozone and the Clean Air Act: Problems and Proposals for Change,
Office of Technology Assessment Staff Paper, April, 1987.
2. Rao, S.T., "Application of the Urban Airshed Model to the New York
Metropolitan Area," EPA-450/4-87-011, May, 1987.
3. G.T. Wolff, P.J. Lioy, R.E. Meyers, R.T. Cederwall, G.D. Wright,
R.E. Pasceri and R.S. Taylor, "Anatomy of Two Ozone Transport Episodes in
the Washington, D.C. to Boston, Massachusetts Corridor," Env. Sci. Tech.
11, 506, 1977.
4. G.T. Wolff and P.J. Lioy, "An Empirical Model for Forecasting Maximum Daily
Ozone Levels in the Northeastern U.S.," Jour, of Air Poll. Cont. Assoc.,
28, 1034, 1978.
5. P. Lorang, USEPA, Office of Mobile Sources, Ann Arbor, MI, personal
communication, October, 1987.
6. R. Baamonde, USEPA Region II, New York, NY, personal comntunications,
September, 1987.
7. R.J. Nichols, J.M. Norbeile, "Assessment of Emissions from Methanol-Fueled
Vehicles: Implications for Ozone Air Quality" presented at the 78th Annual
Meeting of the Air Pollution Control Association, Detroit, MI, 1985.
8. J.N. Harris, A.G. Russell, and J.B. Milford, "Air Quality Implications of
Methanol-Fuel Utilization," SAE Technical Paper Series 881198, presented at
Future Transportation Technology Conference and Exposition, San Francisco,
CA, 1988.
9. G.Z. Whitten, N. Yonkow, and T.C. Myers, "Photochemical Modeling of
Methanol-Use Scenarios in Philadelphia," EPA 406/3-86-001, March, 1986.
88-1-164
------- |