AC 78-03
Technical Support Report for Regulatory Action
Review of Past Studies Addressing the Potential Impact of CO,
HC, and NOx Emissions from Commercial Aircraft on Air Quality
by
Philip Lorang
March 1978
NOTICE
Technical support reports for regulatory action do not necessarily
represent the final EPA decision on regulatory issues. They are intended
to present a technical analysis of an issue and recommendations resulting
from the assumptions and constraints of that analysis. Agency policy
considerations or data received subsequent to the date of release of this
report may alter the recommendations reached. Readers are cautioned to
seek the latest analysis from EPA before using the information contained
herein.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air and Waste Management
U.S. Environmental Protection Agency
-------
TABLE OF CONTENTS
Page
List of Tables ii
List of Figures iii
Introduction 1
Review of Past Airport Air Quality Studies 3
Chronology 3
Model Validity 5
Model and Monitoring Results 11
Carbon Monoxide 13
Hydrocarbons 21
Oxides of Nitrogen 30
Discussion 47
Conclusions Regarding Air Quality Violations On
and Near Airports 47
Carbon Monoxide 47
Hydrocarbons 48
Oxides of Nitrogen 48
The Regional Perspective 50
The Planned FAA Air Quality Study 59
Conclusions 62
References 64
-i-
-------
LIST OF TABLES
Table Title Page
1 National Ambient Air Quality Standards 12
2 Results of CO Monitoring at L.A. 14
International
3 Summary of Six Months of Air Quality 16
Observations at Washington National
Airport - Carbon Monoxide
4 Summary of Six Months of Air Quality 24
Observations at Washington National
Airport - Hydrocarbons
5 Summary of Six Months of Air Quality 38
Observations at Washington National
Airport - Nitrogen Dioxide
6 Results of NO and NOx Monitoring at 39
O'Hare Airport
7 Daylight Scheduled Operations 49
as Percentage of Total Schedule
Operations at Five Airports
8 Commercial Aircraft Emissions as 54
Percentage of Total Air Quality
Control Region Emissions
9 Comparison of Aircraft Emissions 55
With Emissions From Other Mobile
Sources - Three Air Quality
Control Regions
10 1973 HC Emissions from Selected Sources 56
in Baltimore AQCR
11 Comparison of National HC Reductions 57
Achievable With Various Control
Strategies, Including Proposed
Commercial Aircraft Controls
12 Comparison of National NOx Reductions 58
Achievable With Various Control
Strategies, Including Proposed
Commercial Aircraft Controls
-ii-
-------
LIST OF FIGURES
Figure Title Page
1 Comparison of Estimates of Aircraft 7
Emissions at Washington National,
1970-1980
2 Comparison of Estimates of Aircraft 8
Emissions at JFK, 1970-1980
3 Comparison of Estimates of Aircraft 9
Emissions at L.A. International,
1970-1980
4 Comparison of Estimates of Aircraft 10
Emissions at O'Hare, 1970-1980
5 Frequency Distribution, Carbon Monoxide, 15
Hourly Averages, L.A. International
Airport
6a Airport CO Concentrations for Baseline 19
Conditions, Atlanta
6b Airport CO Concentrations for Engine 19
Emission Standards, Atlanta
7a Wind Line CO Profiles Under Baseline 20
Conditions, Summer, Atlanta
7b Wind Line CO Profiles Under Baseline 20
Conditions, Fall, Atlanta
8a Hydrocarbon Isopleths in the Vicinity 22
of Los Angeles International: Aircraft
Sources 3-hour Average for 1970
(6-9 am)
8b Hydrocarbon Isopleths in the Vicinity 22
of Los Angeles International: Aircraft
Sources 3-hour Average for 1980
(6-9 am)
9 Calculated Non-Methane Hydrocarbon 23
Concentrations Downwind of Los Angeles
Airport for 1980 with Non-Aircraft
Sources Controlled
-iii-
-------
Figure Title Page
lOa Airport HC Concentrations for 26
Baseline Conditions - Atlanta
lOb Airport HC Concentrations for 26
Engine Emission Standards - Atlanta
11 Airport HC Concentrations for Worst 27
Case Situtation - Atlanta
12 Regional Impact of HC Emissions 28
from Airport Sources Alone Under
Baseline Conditions - Atlanta
13 Regional 3-hour Hydrocarbon 29
Concentrations Under Worst Case
Conditions - Atlanta
14 N0_ Isopleths at Los Angeles 32
International: Aircraft Sources
15 NO- Isopleths at Los Angeles 33
International: Total Emission Sources
16 NO Isopleths in the Vicinity of 34
Los Angeles International: Aircraft
Sources
17 N02 Isopleths in the Vicinity of 35
Los Angeles International: Total
Emission Sources
18 NOx Isopleths in the Vicinity of 36
Los Angeles International: Aircraft
Sources, Annual Average for 1970
19 NOx Isopleths in the Vicinity of 37
Los Angeles International: Aircraft
Sources, Annual Average for 1980
20a Airport NOx Concentrations for Baseline 42
Conditions - Atlanta
20b Airport NOx Concentrations for Engine 42
Emissions Standards - Atlanta
21a Wind Line NOx Profile Under Baseline 43
Conditions, Summer - Atlanta
21b Wind Line NOx Profile Under Baseline 43
Conditions, Fall - Atlanta
-iv-
-------
Figure Title Page
22 Airport NOx Concentrations for 44
Worst Case Situation, All Sources -
Atlanta
23 Mean Annual Predicted Values of 46
NOx - Salt Lake City
24 Projected Aircraft Emissions to 51
the Year 2000 at J.F.K.
25 Projected Emissions to the Year 52
2000 at Los Angeles International
-v-
-------
-1-
INTRODUCTION
The Air Quality Act of 1967 required HEW to investigate and report on
aircraft engine emissions and their impact on air quality. HEW sponsored
a study and reported its findings to Congress as directed. As a result
of the findings, HEW and DOT negotiated a voluntary smoke retrofit
agreement with 31 airlines. The 1970 Amendments to the Clean Air Act
required another investigation of aircraft emissions and also granted
EPA regulatory authority to set standards as needed to protect the
public health or welfare. EPA studied the problem and developed standards
which were promulgated in final form in July 1973. These standards
contained requirements applicable in 1979 for newly manufactured turbine
and piston engines, with more stringent requirements applicable in 1981
for newly certified large turbine engines.
Since 1973, EPA has monitored the progress of the technology needed to
meet these existing standards and reviewed the impact of various classes
of aircraft on ambient air quality. As a result, there is now in draft
form an NPRM to amend the existing standards. As proposed in this
draft, emission standards for engines other than commercial turbine
engines would be withdrawn. Standards for gaseous emissions from commercial
turbine engines would be relaxed and delayed, as required by the status
of the control technology and by lead time considerations. In addition,
an earlier proposal for a gaseous emission retrofit program has been
revised and is included in the draft NPRM. This program would require
retrofit of HC and CO controls on all in-use commercial turbine engines
with rated thrust greater than 12,000 pounds. This cutoff point includes
the majority of the commercial turbine engine population.
During the preparation of the draft NPRM, commenters, particularly FAA, >
questioned the air quality justifications for the standard for NOx
emissions from newly manufactured engines and for the HC and CO retrofit
program. In addition, EPA may soon set an ambient air quality standard
for short-term concentrations of NO,,, a possibility that generally was
not considered in the past studies which supported the amendments now
in draft form. FAA has announced that it will conduct a new study of
the air quality impact, to be completed in 1979. EPA will participate
in this study.
Previous air quality studies, of which there have been many, have not
been so conclusive as to resolve all uncertainties about the impact of
commercial aircraft on air quality. It is therefore possible that
present uncertainties will be resolved, or at least modified, by the new
FAA study in a way which contradicts the conclusions which justified the
amendments contained in the draft NPRM. If this occurs, EPA will of
course consider further modifications to the standards in light of the
new information.
-------
-2-
This report is partial documentation of EPA's review of the past air
quality studies. Reference 1 recently reviewed the NCL findings from
past studies. This report reviews the HC, CO, and NCL air quality
justifications for the commercial aircraft emission standards contained
in the NPRM. It discusses which points can be considered settled and
which remain unsettled. And it considers what questions the new air
quality study needs to address in order to help clarify the unsettled
points.
-------
-3-
REVIEW OF PAST AIRPORT AIR QUALITY STUDIES
Chronology
Several agencies and groups concerned with aircraft emissions have
conducted a number of studies addressing the impact of aviation on air
quality. The information gathered or calculated by these results has
included emission inventories for airport-related emission sources, air
quality monitoring data on pollutant concentrations, and predictions for
pollutant concentrations based on mathematical dispersion models. Five
of the nation's major commercial airports (JFK, Washington National,
Atlanta International, O'Hare, and L.A. International) and several
smaller airports have received some degree of attention from one or more
of the past studies.
2
The earliest study was in response to the requirements of the 1967 Air
Quality Act. The contractor for the study attempted to compile emission
inventories for JFK, Washington National and L.A. International airports,
and for all FAA controlled terminals combined, for the year 1967. Lack
of good data on engine emission factors and on operational characteristics
at each airport resulted in wide uncertainty ranges around the inventory
estimates. The study also included a few very simple dispersion calculations.
After enactment of the 1970 Amendments to the Clean Air Act, EPA sponsored
two further studies. The Los Angeles County Air Pollution Control
District (APCD) compiled a detailed, current emission inventory for all
sources at L.A. International and monitored concentrations of CO and
particulates in and around the airport for six months.
4
Northern Research and Engineering (NREC) developed a mathematical
dispersion model and compiled emission inventories for 1970, 1975, and
1980 for Los Angeles International, Washington National, JFK, and O'Hare
airports and their surrounding regions. The dispersion model was used
to predict concentrations of pollutants in and around all four airports
for 1970 and for L.A. International for 1975 and 1980. One specific
goal of the modeling was to identify the contribution of aircraft to
local air quality problems, a task that could not have been accomplished
by air quality monitoring or emission inventorying alone.
In 1972 EPA published its own study on aircraft emissions, dealing with
both their impact on air quality and the technology for their control.
The air quality section of the report was based predominately on the
NREC and L.A. Air Pollution Control District studies. EPA did revise
the NREC results in several places to account for more up-to-date
emission factors than those used by NREC. It also added some additional
calculations combining the results of the two studies. EPA compared
aircraft emission control to other control strategies which states were
considering at the time.
-------
-4-
In its report, NREC had compared its dispersion predictions and emission
inventories to those of the APCD. The comparison showed wide disparities
between the two studies. As the model was used further it became evident
that it was capable of order-of-magnitude underpredictions of pollutant
concentrations. EPA contracted Geomet, Inc. to verify the NREC model
against monitoring data from Washington National. Geomet did not test
the original NREC model against the monitoring data it collected at
National. Instead, Geomet made some immediate, minor improvements to
the NREC model, tested it, and found it underpredicting but by less than
had been reported by others. It then made further, major improvements
in the model and concluded that this version (renamed the Geomet model)
performed acceptably. Geomet's_opinion was later challenged by researchers
at Argonne National Laboratory. The Geomet study only developed the
new version of the airport model. It did not answer the same questions
that NREC had tried to answer with its model. Geomet has since applied
its model to smaller airports in-connection with the preparation of
environmental impact statements. ' '
Another pair of airport air quality models, similar in purpose to the
NREC/Geomet models, was developed by Argonne National Laboratory. The
Airport Vicinity Air Pollution (AVAP) model was developed for the FAA.
Argonne also developed the Air Quality Assessment Model (AQAM) for the
U.S. Air Force. The analytical methods used in AVAP and AQAM are similar
and the two models have given similar results when applied to the same
airport. The two models differ only in that one is tailored to operating
conditions at commercial airports, the other to those at military airports.
The AVAP model was developed and validated using O'Hare and Orange
County airports as example cases. The O'Hare version of the model was
not used to analyze the contribution to air quality deterioration from
commercial aircraft. The purpose of this particular application was
model development only.
Argonne has since addressed the question of air quality impact from
commercial aircraft in two studies, both using AVAP. FAA contracted
with Argonne to use AVAP to study the issue of air quality at Washington
National. The results of the study were published in July 1974. The
National version of AVAP was validated using data from twenty days of CO
monitoring during 1972.
EPA also contracted with Argonne for an air quality study using the AVAP
model. This study, published in January of 1975, used Atlanta's
Hartsville International Airport as a setting for a comparison of severa.1
alternate strategies for reducing airport emissions. The study was not
specifically addressed to the question of the impact of commercial
aircraft emissions on air quality, but since a reduction in engine
emissions was one of the strategies considered in the study it is possible
to use the study's results to help answer that question.
-------
-5--
Th e Air Force and the Navy have also used dispersion models to assess
air quality problems around their air bases. The Air Force made a ten-
base study using Argonne's AQAM model. The Navy used a modified version
of AQAM in a study of NAS Miramar, near San Diego.
During May and.. June of 1976, EPA monitored the air quality in Mountain
View, Georgia. This small town is very near the Atlanta Airport.
The dispersion models used in all of the air quality studies did not
account for chemical and photochemical reactions. Systems Applications
Incorporated has attempted to better account for the reactive steps
connecting0aircraft NOx and HC emissions to ambient concentrations of
. , -to
oxidants.
While these air quality studies were in progress, EPA continued to
monitor the development of the technology needed for compliance. As
part of this work, EPA has compiled up-to-date aircraft emission inven-
tories for L.A. International, JFK, and O'Hare airports for the years
1972, 1975, 1980 and 1985.
Model Validity
As a result of the past airport air quality studies, several estimates
of the impact commercial aircraft have on pollutant concentrations near
five major airports are now available. Each of these estimates has been
made with one of three dispersion models: NREC, Geomet, or AVAP. A
discussion of the validity of these three models is necessary in order
to interpret the estimates that will be reviewed shortly.
From a regulatory viewpoint it is desirable to know (1) the spacial and
temporal distribution of pollutant concentrations attributable to
commercial aircraft alone and (2) the concentrations due to other emission
sources. If aircraft emissions alone are enough to cause violations of
the national ambient air standards it is necessary to control them in
order to achieve the standards. If aircraft emissions do not by themselves
cause violations but contribute to them, it may still be that aircraft
emission standards will be a desirable element in an overall control
strategy. The first question is asked first, since it may eliminate the
need to ask the second.
There are three steps in predicting the ambient concentrations of
pollutants attributable to commercial aircraft, each with pitfalls. The
first is to take inventory accurately: the spatial and temporal' distri-
bution of emissions from aircraft must be modeled. Second, the dispersion
of these pollutants over the airport and surroundings must be modeled.
Third, reactions involving the pollutants while they disperse must be
modeled if they occur.
-------
-6-
As mentioned earlier, none of the dispersion models used to date have
attempted the third of these steps. For the reactive pollutants, NOx
(mostly NO when emitted) and HC, the models predict concentrations
either as if no reactions were involved or as if the reactions were
instantaneous.
There are two approaches to assessing the accuracy of the other two
steps. Where possible, the emission inventories for the same airport
can be compared between dispersion models and with inventories prepared
independently, for example those of EPA or the Los Angeles APCD.
Unfortunately, this is difficult to do precisely because the studies did
not all cover the same airports, dates, and hours. The best comparison
possible is in terms of annual emissions from all aircraft using the
given airport. These numbers will include non-commercial aircraft but
this is unavoidable since some studies did not separate the two types.
Non-commercial aviation is generally minor at the airports under consid-
eration.
Figures 1 through 4 show the comparison for the four airports for which
more than one estimate has been made (Atlanta is the exception). It
should be remembered that three factors influence the positions of the
curves: the numbers and types of aircraft and engines using an airport,
the way they operate while there, e.g., taxi-idle time, and the emission
factors for the engines. Obviously, differences in the way past studies
have estimated these factors have resulted in significant differences in
emission inventories. It is not immediately clear which estimate is
correct in each case. However, the estimates from Reference 19 are
based on the most recent, and presumably most accurate, emission factors;
they may not be based on the best airport usage data. The emission
factors used in Reference 4 are the oldest and least reliable. Reference
5, the EPA report, updated the inventory estimates in Reference 4 using
newer data published by the Cornell Aeronautical Laboratory. As can
be seen in the figures, this updating resulted in inventory estimates
closer to those in Reference 19, with the exception of JFK and L.A.
where the improvement was mixed. The inventory estimates from References
12 and 13 were also based on the Cornell emission factors, as were the
inventories for Atlanta (Reference 14, not shown in figures). Generally,
the various predictions agree less for future years than for past years.
The L.A. NOx projection for 1980 from the 1972 EPA report is the most
notable example of this disagreement.
The second approach to assessing the accuracy of a dispersion model is
to compare its predictions for pollutant concentrations with monitoring
data. This serves as a check on the inventory and dispersion steps
combined. Validation against monitoring data is only possible for total
concentrations, and so a model must be capable of predicting concentrations
from all important sources. However, while inventorying aircraft and
other airport sources is relatively straightforward and much the same
for any airport, inventorying emissions from airport emissions is more
difficult and unique for each airport. Each of the airport models has
tried to account for these environ sources.
-------
4ooor
30001.
2000
1000.
1970
Carbon Monoxide
Ref. 5
Ref. 4
Ref. 13
1975
1980
40'
3000U
2000
1000
1970
Oxides cf Nitrogen
O Ref. 13'
Ref. 5
Ref. 4
1975
1980
4000
3000
2000
1000
1970
Hydrocarbons
I
-J
I
Ref. 5
1975
1980
Figure 1. Comparison of Estimates of Aircraft Emissions
at Washington National, 1970-1980
-------
20,000
15,000
10,000
5,000
1970
Oirbon Monoxide
1
f. 4
1975
20,000
15,000|_
£ 10,000
I
5,000!
1980 IS 70
Oxides of Nitrogen
Ref. 4
1975
20,000
15,000
CO
g
10,000
5,000
Hydrocarbons
L
19
Ref
1980
1975
I
oo
I
19SO
Figure 2. Comparison of Estimates of Aircraft
Emissions at J.F.K., 1970-1980
-------
20,000
Carbon Monoxide
15,000
0 Ref. 3
10,000
5,000 [
Ref. 5
Ref. 4
Ref. 26
Ref. 19
20.00C
Oxides of Nitrogen
15,OOC.
> 10,000.
5,000.
20,000
Hydrocarbons
15,000 .
10,000 -
5,000_
I
vo
I
1970
1975
1980
-------
20,000.
15,000
10,000.
5,000.
Cnrbon Monoxide
Ref.
Ref
@Ref. 12
Ref. 19
1970
1975
1980
20,000.-
I
15,000 .
10,000
5,000
Oxides of Nitrogen
Ref. 4
1970
1975
20,000
Hydrocarbons
15,000
« 10,000
1980
5,000
Ref. 12
1970
19
Ref.
1975
O
I
I
"Rso
Figure 4. Comparison of Estimates of Aircraft
Emissions at O'Hare. 1970-1980
-------
-11-
The NREC model for L.A. International underpredicted HC and NOx emission
inventories for the airport itself compared to those of the APCD. It
agreed fairly well on the CO inventory, but still underpredicted CO
concentrations by a factor of about 2.8. NREC attributed this to a poor
model of environ CO sources. Later, Geomet decided that the dispersion
algorithm of the NREC model was itself partly to blame and modified it
in ways that gave consistently higher concentrations. This version of
the NREC model still underpredicted when compared with monitoring data
from Washington National. Geomet thought this was due to faulty inventories
for both airport and environ sources and to the coarseness of NREC's
source point model. Thus, the NREC model has been shown to be incapable
of predicting total pollutant concentrations. What little evidence is
available suggests that the model underpredicted concentrations due to
aircraft sources, assuming its emission inventory was correct.
After Geomet made further changes to arrive at the Geomet model, it
compared the frequency distribution of concentration levels predicted by
the model to the distribution from monitoring data. The Geomet model
did fairly well for the non-reactive pollutants except at the low
frequency/high concentration end of the distribution - where violations
of ambient standards would be expected. The Geomet model was later
criticized for being unable to match measured concentrations on an hour-
by-hour basis.
Argonne's AVAP model has been tested against monitoring data at O'Hare,
Washington National, and Atlanta. The model was found to perform well
enough in predicting total NOx concentrations at O'Hare, a case in which
emissions from non-airport sources did not contribute much background
noise. The model did somewhat less well for CO at National and Atlanta.
For reactive pollutants and ones for which poorly modeled environ sources
were important, the AVAP model has done poorly. Thus, there is evidence
that suggests that the AVAP model can reasonably accurately predict
concentrations of non-reactive pollutants from aircraft, and no specific
evidence that it cannot.
None of the models account for obstructions to dispersion such as buildings,
and should not be trusted to predict concentrations due to such obstructions.
Further, comparisons of predicted NOx to measured NO,, have usually shown
overprediction, since some of the predicted NOx is NO.
Model and Monitoring Results
Table 1 lists the national ambient air quality standards for CO, N0~,
and HC. In addition, EPA may establish a short-term N0« standard in the
near future. This standard, if established, might be in the range of
190 to 320 pg/m hourly average NO-, the range of current World Health
Organization recommendations. The WHO recommendations are that the
hourly average not be exceeded more than once per month.
-------
-12-
Table 1 NATIONAL AMBIENT AIR QUALITY STANDARDS
Pollutant
Carbon monoxide
(Primary and secondary
standards are the same)
Nitrogen dioxide
(Primary and secondary
standards are the sarne)
Hydrocarbons (non-methane)
(Primary and secondary
standards are the sa
Participate matter
Primary standard
Secondary standard
Sulfur dioxide
Primary standard
Secondary standard
Oxidant
(Primary and secondary
standards.are the same)
Standard Description
- 10 milligrams per cubic meter (9 ppm), maximum
8-hour concentration dot to be exceeded more .than
once per year.
- 40 milligrams per cubic meter (35 ppm), maximum
1-hour concentration not to be exceeded more than
once per year.
- 100 micrograms per cubic meter (0.05 ppm), annual
arithmetic mean.
- 160 micrograms per cubic meter (0.24 ppm), maximum
3-hour concentration (6-9 a.m.) not to be exceeded
more than once p;-r y?ar. For use as a guide in
devising implementation plans to meet the oxidant
standards.
- 75 micrograms per cubic meter4 annual geometric
mean.
- 260 micrograms per cubic meter, maximum 24-hour
concentration not to be exceeded more than once per
year.
- 60 micrograms per cubic meter, annual geometric
mean, as a guide to be used in assessing implementa-
tion plans to achieve the 24-hour standard.
- ISO microgrnms per cubic meter, maximum 24-hour
concentration not to be exceeded more than once per
ysar.
- 80 micrograms per cubic meter, annual arithmetic
mean.
- 365 micrograms per cubic meter, maximum 24-hour
concentration not to be exceeded more than once per
year.
- 60 microgrsms per cubic meter, annual arithmetic
mean.
- 260 micrograms par cubic meter, maximum 24-hour
concentration not to be exceeded more than once per
year.
- 1300 micrograms per cubic meter, maximum 3-hour con-
centration not to be exceeded more than once per
year.
- 160 micrograms per cubic meter, maximum 1-hour con-
centration, not to bo exceeded more than once per
year.
-------
-13-
In the summary of results that follows, no mention of a study's results
usually means the study did not yield any useful information on that
topic.
Carbon Monoxide
3
The results of the CO monitoring at L.A. International by the APCD are
summarized in Table 2. The "satellite" locations in the table are
passenger terminals completely surrounded by ramp areas. The ticketing
building is adjacent to an access road and loading area. The table
shows that the one-hour CO standard was exceeded repeatedly at two
satellite terminals and the ticketing building. The monitored violations
at the terminals were probably due to aircraft and ground service vehicles
combined, judging from the locations. Access traffic was probably
responsible for the violations at the ticket building. Figure 5 compares
the frequency distribtuions of the CO measurements from two monitoring
stations. Station 204 was west (generally upwind) of the airport;
station 208 was at one of the passenger terminals. The figure indicates
the frequency of the CO violations and the effect airport emissions had
on raising CO concentrations above those in the upwind areas.
4
The NREC model predicted that aircraft sources alone would cause CO
levels above the standards at the ends of runways at all four airports
and at a terminal at JFK. CO levels at most other points were quite
low. NREC's CO inventory for aircraft sources at O'Hare was about 25%
high compared to other estimates; its inventories for other airports
were in general agreement with other estimates (see Figures 1-4).
The EPA report summarized the CO results from the APCD and NREC studies.
It also looked more closely at a specific monitoring site in L.A. where
a residental area was close to and downwind of a runway. EPA noted that
the APCD monitoring data had shown frequent violations of the CO standard
at this site. It then used a rollback calculation to show that, if the
present contribution by aircraft to measured CO levels at that site were
more than 20%, there would still be CO violations at the site even after
other CO sources were controlled. While the NREC model had not predicted
any violations at the site because of its general tendency to underpredict,
it did suggest that the percent contribution from aircraft was over 20%.
No conclusion should be drawn without first remembering that the NREC
model tended to overpredict the percent contribution from aircraft
because it had much too low an inventory for non-aircraft sources.
While monitoring at Washington National, Geomet recorded violations of
both CO standards at a terminal and at a maintenance area and violations
of the eight-hour standard at another runway and in the auto traffic and
parking area. The summary of Geomet1s monitoring is given in Table 3.
The CO levels in the terminal and maintenance area reached twice the
allowed levels at their maximum. Although Geomet's model was capable of
predicting the aircraft contributions to pollutant concentrations,
Geomet did not report them. Geomet made a general comment that "aircraft
-------
Table 2. Results of CO Monitoring at L.A. International
STA. LOCATION
NO.
KAY (from 5/10/70)
20h F.A.A. VOR Site West End
209 CcmMar.d Post, East End
201 Control Tower, Admin. Bldg.
203 Satellite 2
202 Satellite 2
208 Satellite 7
207 Satellite 7
205 Ticketing Bldg., Mo. 7
206 Ticketing Bldg., No. 7
APCD MOliTTOHING STATIONS
76 Southwest Coastal (Lennox)
1 Central (Dcwntown L.A.)
JUNE
20li F.A.A. VOR Site West End
209 Command Post, East End
201 Control Tower, Admn. Bldg.
203 Satellite 2
202 Satellite 2
208 Satellite 7
207 Satellite 7
205 Ticketing Bldg,, Mo. 7
206 Ticketing Bldg., No. 7
APCD HONITORIKO STATIONS
76 Southwest Coastal (Lennox)
1 Central (Downtown L.A.)
JULY
20U F.A.A. VCH Site West End
209 Command Post, East End
201 Control Tower, Adjr.in. Bid p.
203 Satellite 2
202 Satellite 2
20$. Satellite 7
20? Satellite 7
205 'iicko-Ung Uldg., No. 7
206 Picketing Uldg., K;o. 7
APCD KONTTORIvt; STATIONS
7o "O;ithv:c£;t Coantal (Lennox)
1 "'.Ttrnl (Downtown L.A )
(a) "0" (.! :.side; »1"~ libido.
CARBON MONOXIDE
One-Hour Avg.
« pp:n
Range Arith.
Oa
0
0
0
Ia
0
T
0
I
0
0
0
0
0
0
T
0
T
0
I
0
0
oa
0
0
0
Ia
0
I
0
1
0
0
run.
1
1
1
1
2
2
b
5
2
1
1
1
2
1
2
1
2
2
2
1
1
1
1
1
2
1
1
3
2
2
-
nax.
19
21
21
15
35
22
63
70
35
35
6
16
12
26
79
1.8
90-
66
18
2U
5
8
10
*0
110
2L|
75
51
20
25
wean ill
3.0
3.1i
li.8
3.7
6.0
5.3
11.1
13.6
6.2
5.5
2.2
3.2
It. 6
6.2
9.3
14.8
11.3
13.3
6.2
5.0
1.7
3.3
li.li
U.6
9.8
5.1
9.8
33.1
6.1
5.6
STA. LOCATION
NO.
AUGUST
20lj F.A.A. VOH Sita VJcsb End
209 Cor,-iwnnd Post, East End
201 Control Toucr, Admin. Bldg.
203 Satollito 2
202 Satellite 2
200 Satellite 7
207 Satollito 7
20> Ticketing Bldg., No. 7
206 Ticketing Bldg., Mo. 7
AITI) KOSITORINU STATIONS
76 Southv.'oct Coastal!. (Lennox)
1 Central (Downtown L.A.)
SEPPEMP.ER
20li F.A.A. VO", Site West End
209 Coranani Post, East tod
.201 Control Tover, Admin. Bldg.
203 Sat 0111-10 2
202 Satellite 2
208 Satellite 7
207 Satollito 7
205 Ticketing Bldg., No, 7
206 Ticketing Eldg., Mo. 7
APCD xoniToraro STATIONS
7o~ South-Host Coastal (Lennox)
1 Central (Dovmtovm L.A.)
October
20U F.A.A. VOR Site West End
209 Command Post, East End ,
201 Control Tower, Adr^in. Bldg.
203 Satellite 2
202 Satellite 2
208 Satellite 7
207 Satellite 7
205 Ticketing Blri^., No. 7b
206 Ticketing Bide., No. 7
APTD MONITORING STATIONS
lL SouliM.-ijrit Coastal (Lennox)
1 Centvr.l (DOT ml cun L.A.)
(n) "0"-- (.utuicloj "I"-- -Inside.
CARBON MOIIOXIDS
Ono-Hour Avg.
- ppm -
Range Arith.
Hin. :i-,x. Mo.-in
Oa
0
0
0
1s
0
I
0
t
0
0
0
u
0
0
1
0
1
0
I
0
0
0*
0
0
0
Ia
0
).
0
I
0
0
1
3.
1
1
1
1
3
3
2
1
1
1
2
2
2
2
3
3
1
1
1
3
2
3
1
3
2
U
1
1
12
u»
Hi
29
Uil
20
71*
132
23
27
18
31
27
1(6
79
27
39
66
liT
l|i
2Ij
23
27
2u
;0
68
37
35
51
92
3!i
,30
2.2
3.6
5.1
5.5
10.9
5.6
10.1
15.8
6.6
h.O
3.9
5»o
U
.2
10,3
f.o
.2
>
12.6
f. n
1 A
Jj.O
.9
5.9
7.1
,U
10.14
7.7
.3
18.5
15.0
.0
6.0
-------
40,
Ambient
Standard
35 PPM
CL,
20!
.o|-
0.1
-£
Figure 5
FREQUENCY DISTRIBUTION, CARBON MONOXIDE, Hourly Averages
L.A. International Airport
(Source: Reference 3)
.'SL'fir-f^ffTKAmrm
X
X .-
|x'
x y
»' A
\
10
30 50 70
Percentile
90
X
X ,13
/
y
99
99.9
Ul
I
-------
-16-
Table 3. Summary of Six Months of Air Quality Observations
at Washington National Airport -
Carbon Monoxide
Number of
Pollutant Observations
Carbon Monoxide - 1-hour
Station
1 runway '321
2 hangars 1 830
3 runway 2713
4 ramp 3975
5 parking Iot4137
6 runway 1 599
7 maintenance4G61
8 unspecif iedl 574
Total Obs. 21,260
Carbon Monoxide - 8-hour
Station
1 runway 293
*"* "OP
L hangars ''<--
3 runway b 1 9
4 ramp 923
5 parking lot 1004
6 runway 381
7 maintenance 920
8 unspecified 352
Msdiai:
Value
\
values s mg/
1.70
1.11
1.42
3.40
2.50
1.28
2.80
1.70
values, mg/
2.14
1.12
1.45
3.84
2.66
1.33
3.16
1 . 64
'lean
Value
m3
2,43
1.52
2.10
4.43
2.99
1.52
4.97
1.84
r.i'*
l! !
2.59
1.152
2.10
4.£0
3.02
1 ,53
5.18
1.87
98th Percentile
Value
10.0
9.46
6.26
17.1
9.10
4.86
69.5
6.80
9.97
9.33
5.74
12.8
7.76
4.44
70,0
6.10
Maximum
Value
19.4
16.3
10.3
60.1
19.4
9.35
88.1
11.4
12.0
9.63
7.06
21.6
13.6
7.29
83.3
8.10
Total Obs. 4,920
-------
-17-
and airport emissions represent, for Washington National Airport, only
a very small fraction (of the order of a few percent) of the total
pollution concentration burden, except on few occasions." The locations
of the monitored violations seem to indicate that aircraft sources were
important in at least some of the areas with monitored violations,
however.
Argonne did not monitor CO concentrations in the course of its model
development work at O'Hare. It only predicted CO concentrations for the
few cases for which it also compared NOx predictions and measurements.
The highest of these CO predictions was 8.5 yg/m at the terminal area,
below either CO standard. Argonne did not report what fraction of this
was attributable to aircraft. It should be noted that the Argonne
predictions did not account for the effect of obstructions to dispersion.
Nor did the Argonne modelers select and make predictions for worst case
conditions. It is thus possible that CO violations do occur at O'Hare.
Argonne both monitored and predicted CO levels as part of its study of
Washington National. Argonne's measurements did not cover as extended
a period as did the Geomet measurements. Of the data it did take,
Argonne reported only that for open areas near runways. Data from near
the terminal building were rejected on the grounds that they could not
be used in a validation test of a model which assumed no obstructions
were present. In its monitoring near the runways, Argonne did not
observe any CO violations even though it used sites similar to those
where Geomet had measured violations.
Argonne's model did not predict CO violations at the auto parking area
and terminal area, two sites where Geomet had actually observed violations.
Argonne did predict that aircraft emissions alone would cause violations
of both CO standards at points near the end of runways, under conditions
ranging from average to worst-case. High levels due to all airport
sources were more extensive. If the worst conditions occurred for eight
hours at one time, the total CO level would exceed the standard over
portions of the airport and some areas adjacent to it.
14
The study of the Atlanta airport by Argonne for EPA did not observe
any violations of the CO standards in its monitoring data and did not
predict any violations, even under worst-case conditions. Although
Atlanta ranks as a major airport, its CO inventory for aircraft sources
is only about a third of that at L.A. International, JFK, or O'Hare.
Atlanta's CO inventory is about twice that of National's, but is spread
over a larger land area. These may be reasons why no violations were
observed or predicted at Atlanta.
The Atlanta study does provide some information that is useful in
interpreting violations at other airports, however. Figures 6a and 6b
give an indication of the spatial distribution of CO levels due to
aircraft. Figure 6a is a baseline that includes both aircraft and other
-------
-18-
sources. Figure 6b represents the same conditions but with a 62% reduction
in aircraft CO emissions. A comparison of the two figures shows that
aircraft emissions predominate in areas slightly downwind of runways.
CO concentrations outside the airport boundary are hardly affected by
the large reduction in aircraft emissions. Although the figures do not
show it clearly, CO levels at the ramp area during the summer decreased
40% with the 62% decrease in aircraft emissions, indicating that about
65% of the level in that area is due to aircraft, under summer wind
conditions. Figures 7a and 7b show that the CO level due to the airport
as a whole drops off very rapidly with distance from the airport.
During summer the CO level from airport sources drops by a factor of
five in the first two kilometers past the airport boundary; in the fall
the decrease is slower but still fast. This suggests that at airports
with CO problems attributable to aircraft, the problems will be localized.
9
The Geomet study of the Salt Lake City Airport predicted that, in an
airport with approximately one-tenth the aircraft CO emissions as those
of the largest commercial airports, the only violation due to aircraft
CO emissions was near the end of a runway. On an annual basis, CO
levels from airport sources at points outside the airport boundaries
were predicted to be at most a few percent of the eight-hour standard.
The Salt Lake City model did not have enough resolution to predict
concentrations within the airport itself, except for the runway site
already mentioned.
The air bases studied in the Air Force project were all small in
comparison to the large commercial airports. Their CO emissions averaged
about one-eighth those of O'Hare, for example. The report from the
study was mainly concerned with the effects of the air bases on areas a
few kilometers and more away. At five kilometers, annual average C0~
levels attributable to military aircraft were no higher than 40 yg/m ,
i.e., less than a percent of the eight-hour CO standard.
In summary: Violations of CO standards have been observed at L.A.
International and Washington National. These violations have occurred
in areas that suggest aircraft emissions were important contributors.
Violations due to aircraft alone have been predicted at these two
airports and at O'Hare and JFK in addition. Observed and predicted
violations have been limited to areas near the ends of runways and at
terminal ramp areas. Aircraft contribute to violations beyond the
airport boundaries in some cases, but in general the CO levels due to
aircraft drop rapidly away from congested areas on the airport.
-------
ALL CONCENTRATIONS IK uG/M3, 1-HOUR AVERAGE
ALL CONCENTRATIONS IN VG/K3, 1-HOUR AVERAGE
2200
(J200
\ »w
\600 ointci.'
Figure 6a. Airport CO Concentrations for Baseline Conditions, Atlanta
^500
Figure 6b. Airport CO Concentrations for Engine Emission Standards
Atlanta
-------
500
40°
ca
B 300
S I
S S 200
o ?5
'
O
CD
CC
100
n
-14 -12
228°(SW)
-20-
~T
-10 -8 -6 -4
UPWIND
-2
6 8 10
DOWNWIND
12 14
48°(NE)
DISTANCE FROM AIRPORT LOCATION POINT (ALP).km
Fig. la. Wind Line CO Profiles Under Baseline Conditions, Summer, Atlanta
a)
00
n)
M
a)
M
g
1100
1000
900
'I 800
° 700
re
S 600
z
o
uj 500
o
P 400
o
CO
a
300
200
100
0
1
I I I. I 1 I I I
ALL SOURClsy
-.1 ......L
I
I
I
-14 -12 -10 -8 -6
I7°(NNE) UPWIND
6 8 10 12 14
DOWNWIND 197° (SSW)
-4-2024
DISTANCE FROM AIRPORT LOCATION POINT (ALP), km
Fig. 7b. Wind Line CO Profiles Under Baseline Conditions, Fall, Atlanta
-------
-21-
Hydrocarbons*
NREC predicted that HC concentrations attributable to aircraft alone in
the 6 am to 9 am period were above the national standard at points on
each of the four airports modeled. At Washington National, the exceedances
were by about a factor of three and limited to the areas around the end
of runways. At JFK, they were very much higher, up to a factor of 40,
and occured at terminals, runways, boundaries, and areas as much as 5
kilometers downwind. L.A. and O'Hare were in between these two extremes.
For all four airports, other sources contribute towards large concentrations
also, so that predicted violations from all sources covered most of the
airport.
EPA revised the NREC estimates and constructed HC isopleths for L.A.,
Figures 8a and 8b. EPA's 1970 HC inventory was high by about 30% while
the 1980 inventory was close to the most recent estimate (see Figure 3).
In both cases, the area experiencing violations from aircraft sources
alone is predicted to extend well into the adjoining residential community.
As mentioned earlier, HC is a reactive pollutant and is not harmful
until it has reacted with NO- and light to form oxidants. EPA made a
simple dispersion calculation to determine whether the HC would react
before dispersing to such low levels that the product oxidants would not
be of concern. The results are shown in Figure 9. The curves are drawn
for 1980 and assume that non-aircraft HC sources have been controlled.
The wind velocity is about 5 km/hr. With an assumed reaction time of 3
hours, the figure shows that aircraft sources alone would produce
oxidant concentrations below, but close to, the ambient standard. With
the higher HC emissions of 1970, the oxidant standard would have been
found to be violated by aircraft sources alone. As it is, even in 1980
aircraft emissions contribute to a severe oxidant problem downwind of
the airport.
Geomet took HC measurements at three sites at Washington National. The
results are summarized in Table 4. The maximum values for the 6 am -
9am period are so high that even if aircraft contributed only a few
percent their contribution would be enough to violate the HC standard.
The median and mean HC values indicate that high HC is a very frequent
problem at National.
12
Argonne reported HC measurements at O'Hare up to 20 times the standard,
during one-hour periods throughout the day. Levels well outside the
airport were about 5 to 10 times the standard. At O'Hare, aircraft
accounted for 70% of airport HC emissions, so a very substantial part of
the higher levels on the airport were due to aircraft.
* Throughout this report the distinction between total hydrocarbons and
reactive hydrocarbons has been ignored. Past studies have not been
mutually consistent as to which they reported.
-------
-22-
*"*~~ ~"'r r /V *"*^T*."^" *,.'. "'v. ".' i '..-.. 11 .'J .',;'. ;:,;/' I " __ >
1 -- ''-O._-.-. :- "fl-r/.-C ' h'*'* "* '"«''-' ' T--".' Jill' '''' ';'-"' >:" -- I ]
V 'K'-^f' i l---i="'Y"-|'-"j ---j ]||;:ti^|;&Jc:::
HiM£U^^
'. :
3V
Figure 8b.
(numbers ir
HVD^nr.ARBON I50PI.Llti5_ULJLt|iJLLUILllljaLJJ}SL^r1iU^tKLLR_'l£Ei2li/lki__MMM£LJ5UB£Ii
3-Hr Avcr.K,e for 1?30 (6-9 AM)
-------
600
500.
400,
m
c
to
C
o
1-1
o
o
300.
200
100,
Figure 9
CALCULATED NON-METHANE HYDROCARBON CONCENTRATIONS
DOWNWIND OF LOS ANGELES AIRPORT FOR 1980
WITH NON-AIRCRAFT SOURCES CONTROLLED
Meteorological Conditions Used:
Wind from West
Stability Class 3
Wind Speed 1.5m/sec
Mixing Height = 200m
Total, All Sources
NAAQS =160 ug/m3
Total Airport
Aircraft Alone
oiL
o
Airport East Boundary
(3.2 km)
8
16
20
24
DOWNWIND DISTANCE, KILOMETERS
-------
-24-
Table 4. Summary of Six Months of Air Quality
Observations at Washington National
Airport - Hydrocarbons
Number of Median M^nn 98th Percentile Maximum
Pollutant Observations Value Value Value Value
Non-Methane Hydrocarbons - 1-hour valuG:., yg/m
7788 9564
1828 4440
2461 4813
Non-Methane Hydrocarbons - 6-9 a.m. values,
Station
2 hangars ] 04 465 933 8677 9063
3 runway 73 HI 320 2552 2934
6 runway 28 1229 1187 1974 1974
Total Obs. 205
Station
2 hangars
3 runway
6 runway
Total Obs.
1922
' 2129
1605
5,657
655
166
1238
1258
321
1 205
-------
-25-
The predictions made by Argonne using the AVAP model at Washington
National follow the pattern already described. With moderate to heavy
aircraft activity and average to poor ventilation, air entering the
airport boundary has HC levels about 5 times the level of the standard.
It picks-up more HC as it passes over the airport and leaves with 10 or
more times the standard at the downwind boundary. Dilution occurs
downwind of the airport, but the HC level is still above the standard a
mile away, the limit of the model's range. On the airport itself, both
runways and terminals have high concentrations due to aircraft.
The Atlanta study again provided the most easily visualized results.
All the HC results for Atlanta are based on model predictions. There
were no HC measurements made. Figures lOa and lOb are taken from
Reference 14 and are similar to Figures 6a and 6b which have already
been explained. Aircraft HC emissions are 70% lower in Figure lOb than
in lOa. Aircraft HC emissions at Atlanta are about one-third those at
the largest commercial airports. Under baseline conditions, Figure lOa,
HC concentrations exceed the 160 y g/m level on the runways and terminal
area, and in the parking areas. When aircraft HC emissions are reduced
70%, the changes in concentration levels indicate that aircraft alone
were causing in excess of 160 y g/m over both runways and terminal
during part of the year. In summer, when the wind blows runway emissions
into the terminal area, aircraft are responsible for a concentration of
over 1000 y g/m at the terminal (peak valves are not shown on the
figures but are given in tables in the report). In the fall, this
number is 238 yg/m .
Under worst case conditions, with poor ventilation and high airport
activity, HC levels exceed the standard over the entire airport and into
the surrounding areas, Figure 11. However, the Argonne report indicates
that most of the HC levels in this case are attributable to emission
sources upwind of the airport. The terminal.,area is an exception.
There, aircraft are responsible for 450 yg/m of the 1430 yg/m total.
Figure 12 shows the impact of airport HC emissions on the area around
the airport during normal baseline conditions. Only the town of Hapeville
is predicted to experience HC violations due to airport sources. As
aircraft account for only 70% of the total airport HC inventory, even
Hapeville's predicted problem is not solely attributable to aircraft.
However, under worst case conditions airport sources are responsible for
HC violations up to 14 kilometers downwind, Figure 13. Even if the
curve for airport sources in Figure 13 is moved downward 30% to account
for the fact that not all airport sources are aircraft, violations due
to aircraft alone extend about 7 kilometers downwind. The windspeed
under the worst case conditions is 7 kilometers/hour, so airport HC
emissions stay concentrated at high levels for two hours and aircraft
emissions for one hour. This indicates that the possibility of oxidant
formation cannot be ignored. And as long as other HC sources upwind of
the airport remain uncontrolled, aircraft HC emissions are very definitely
-------
ALL CONCENTRATIONS IN i.G/M3, 1-HOUR AVERAGE
Fig. lOa. Airport )IC Concentrations for Baseline Conditions - Atlanta
ALL CONCENTRATIONS IN pG/M3. 1-HOUR AVERAGE
Fig. lOb. Airport KC Concentrations for Engine Emission Standards - Atlanta , |
-------
ALL CONCENTRATIONS IN yG/M3, 1-HOUR AVERAGE
»,
WIND
DIRECTION
i
f 5
Fig. 11. Airport HC Concentrations for Worst Case Situation, Atlanta
-------
ALL CONCENTRATIONS IN uG/M3, 1-HOUR AVERAGE
WIND DIRECTION
S \ -*'-^-»l75
'CLAYTON co.
FAYEmE CO. I !\
^""i
ij
00
I
JONESBORO
SUMMER
FALL
Fig. 12 Regional Impact of KG Bnissions from Airport Sources Alone Under Baseline Conditions - Atlanta
-------
1250
!000
750
500
o
o
250
NAAQS
T
HC, WORST CASE
3-HOUR AVERAGE
{S-S A.M.)
BASELINE
TOWING
ENGINE STANDARDS
ALL SOURCES
AIRPORT SOURCES ONLY
160 yg/nT
0
I
I
I
I
I
I
I
8
10
12
14
16
-14 -12 -10 -8-6-4-2 0 2 4 6
DISTANCE, kilometers
Fig 13 Regional"3-Hour Hydrocarbon Concentrations Under Worst Case Conditions - Atlanta
-------
-30-
contributing significantly to oxidant violations. Note that under the
worst case used in the Argonne study, Atlanta itself is upwind of the
airport and so the airport emissions do not effect the most heavily
populated areas. Argonne chose that wind direction so that conditions
at the airport itself would be worst. It did not claim that equally
poor conditions with wind towards the city did not occur.
9
The Salt Lake City study predicted that HC violations due to all airport
sources combined were possible at sites on and near the airport. HC
levels of about 400 yg/m were predicted at these sites. Although the
study report did not say what fraction of airport HC emissions were due
to aircraft, the fractions from other airports indicate that aircraft
alone would cause HC levels above 160 yg/m at these sites. Airport
induced HC levels at sites away from the airport were generally very
small compared to those at the airport itself.
HC emissions from aircraft at the Air Force base (Williams) with the
highest HC inventory were 1540 tons/year, compared to emissions of 5000
to 10,000 tons/year at the major commercial airports (depending on the
source and year of the inventory). HC concentrations 5 kilometers
downwind of this base reached an annual 22 yg/m due to emissions from
the military aircraft. It is plausible from this that HC levels might
reach 160 yg/m or more at similar distances from the commercial airports,
especially for single days under adverse conditions.
In summary; Measured HC concentrations at airports commonly greatly
exceed the 160 yg/m level. The predictions are that during unfavorable
conditions aircraft alone can cause excessive HC levels up to several
miles downwind of the airports. Under these conditions other HC sources
are also important and thus aircraft contribute to HC problems that
already exist over wide areas. Since HC is a reactive pollutant that is
not harmful itself, the conclusions about HC levels do not necessarily
imply an air quality problem. However, there are indications in some
cases that aircraft HC emissions may react to form oxidants while still
concentrated above the 160 yg/m level. Without doubt, the combination
of aircraft and other sources causes oxidant violations.
Oxides of Nitrogen
Under the assumption that the NO to NO,, reaction was instantaneous,
NREC predicted that annual average N0_ levels from aircraft at several
sites near runways at the four airports it studied would be in the range
of 70 to 200 y g/m . N0~ levels from all sources were predicted to cause
annual averages near 100 yg/m over much of the airports. Except for a
few anomalous^looking values, total annual NO- levels were not predicted
over 300 yg/m . NREC also calculated the maximum 24-hour averages at
each airport* Sites at each airport again had predicted levels near
100-200 yg/m , due to aircraft alone, with a few isolated higher points.
24-hour average levels from all sources were higher, ranging from 100 to
1000 yg/m on the airports. The NREC tables showed some large concentration
differences at nearby receptor points, which suggest that the source
point model used in the predictions was too coarse.
-------
-31-
NREC constructed NCL isopleths for L.A. International only, Figures 14-
17. Given the known problems with NREC's non-aircraft inventory,
Figures 15 and 17 should not be trusted much to indicate anything other
than that aircraft on the runways cause N0~ levels comparable to those
caused by the cars in the access road and parking area. Outside the
airport itself, the NO- level attributable to aircraft is no higher than
10% of the ambient air quality standard.
EPA revised NREC's NOx inventory for L.A. upward to incorporate newer
emission factors and constructed new isopleths. These are shown in
Figures 18 and 19. The EPA's 1980 emission inventory was too high
compared to more recent projections, so Figure 19 should be doubted (see
Figure 3). Figure 18 estimates that the residential area near the
airport experienced NOx levels between 20 and 50 y g/m due to aircraft
in 1970. These levels have probably increased since then, but not as
much as shown in Figure 19.
EPA also constructed an isopleth for O'Hare for 1980. EPA's inventory
for this airport and year was reasonably close to its more recent
projection (see Figure 4). The isopleth showed that NOx levels from
aircraft range up to the 100 yg/m level outside the airport boundaries.
EPA realized that assuming an instantaneous conversion from NO to N0~
resulted in overestimates of N0? c -- -- - - -> - -- - - ^-
isopleths for NOx rather than NO,,.
Geomet monitored N0_ levels at three sites at Washington National for
six months. The monitoring data, summarized in Table 5, indicate that
the annual N0? standard was violated at all three sites. The maximum
and 98th percentile one-hour concentrations are high enough to indicate
repeated violation of any likely short term NO- standard. Unfortunately,
there is no way to determine what fraction of these concentrations is
attributable to aircraft.
In developing the AVAP model, Argonne monitored both NO and total NOx at
O'Hare. Most measurements were for NOx, but simultaneous NO and NOx
measurements were taken at a few sites. One-time NOx measurements at
the airport perimeter ranged from 50 to,,540 yg/m . Measurements within
the airport ranged from 100 to 660 yg/m .
/
Table 6 shows the measurement results for those sites where both NO and
NOx were measured. Very few measurements were made compared to the
monitoring at Washington National. The measurements show that little of
the ambient NOx was NO at the sites outside the airport, while more was
at the sites on the airport. The highest fraction of NO was measured at
the terminal area. This pattern agrees in general with the idea that
most of the NO emitted from aircraft at the airport remains NO for the
time it takes it to disperse away from the airport. Note that in a
couple of cases the N0~ level is high enough to raise concern about
violations of a short-term NO standard.
-------
Note: Peak values of concentration
do not appear on the figure.
(numbers in/jg/nr)
Figure 14. N02 Isopleths At Los Angeles International: Aircraft Sources
-------
Note: Peak values of concentration
do not appear on the figure.
(numbers in fig/m3)
Figure 15 - NOo ISOPLETHS AT LOS ANGELES.-1NTERNATIONAL: TOTAL EMISSION SOURCES
-------
-34-
j #> -.--v - « i -Ji «
£
MANHATTAN ^
BEACH
Note: Peak values of con- K.T
centration do not
appear on the figure.
BEACH
(numbers in/jg/m3)
FIGURE 16 - NO? ISQPLETMS IN THE VICINITY OF
LOS ANGELES INTE~RNATI ONAL: AIRCRAFT SOURCES
-------
-35-
>
, .. "" V^'^^--'--'^ X-'/ 3^'C Ba''"""'!! "-"-----. «,-.n,,,t .'
W'' X?{*&* '"'"'> 'i;o.,-,-.J i .>-; >
A \ MarinA >,..*'/.". 'i " \ .Xv^'i ;-'1 < ,«'v' i
Note; Peak values of con- w.r/?wos>i\
""~~"~ ».
central ion do not ap-
pear on the figure.
(numbers in/jg/m')
F1GURF 17 - N0? ISOPLF.THS IN THE VICINITY OF LOS
ANGELES INTERNATIOWAL: TOTAL EMISSION SOURCES
-------
WiavT^H^1:vK
^i;^j^;f.,^..;.e.v;:.:A: ..::'.;-!'; j ,'
CHr^v i><£: Vl '^.'^^r~-r-~
^?vi/im>^TO'SfeH
f u-.u* ;;iivV\vV^Y ^-'^':; //₯tf^--"-
/ LH...V.-L...'-''- '.\'A\\ \- ur-'-'x ' //;'/- v" >
^J-S'orif
-------
v>7V
:, V \
cr:^-v-'t:: (nunhers i n //g/rn )
i.ri-iii:.-;:::--" FIGURE ]_g NO . iSOPLETHS IN THE VICiMTY OF LOS ANGELES INTERNATIONAL: AIRCRAFT SOURCES
Annua 1 Average for 1'380
I
CO
-------
-38-
Table 5 Summary of Six Months of Air Quality Observations
at Washington National Airport -
Nitrogen Dioxide
Pollutant
Number of
Observations
Median
Value
Mean
Value
98th Percentile
Value
Maximum
Value
Nitrogen Dioxide - 1-hour values,
Station
2 hangars 3489 145 180 488 , 631
3 runway 3214 104 149 546 639
6 runway 2040 89.6 127 427 623
Total Obs. 8,743
-------
-39-
Table 6. Results of NO and NOx Monitoring
at O'Hare Airport
(Concentrations in
Time
8
9
10
11
12
13
14
15
16
17-8
18
19
NOx
Receptor No. 1,
13A
113
93
99
144
134
185
261
274
227
Receptor No. 4,
453
453
NO
access road in front of
33
13
13
27
24
34
56
75
66
90
near end of Runway 14L,
74
121
NO *
z
terminal ,
101
100
80
72
120
100
129
186
208
137
12/4/71
379
332
NO/NOx
12/2/71
25%
12
14
27
24
25
30
29
24
40
16%
27
Receptor No. 5, near end of Runway 9L, 12/7/71
16 2194 13 2181 <1%
20 1988 27 1961 1
Receptor No. 6, near end of Runway 27R, 12/7/71
21 185 27 158 15%
-------
-40-
Table 6. (continued)
Time NOx NO NO * NO/NOx
8
9
10
11
13
14-15
11-12
12-13
16-17
8-9
9
10
11
12
12
13
14
14-15
15-16
Receptor No.
337
200
143
350
181
100
Receptor No.
132
150
Receptor No.
113
Receptors No.
49
12
76
2
16
144
126
95
144
144
7, TWA gate G6, 1/11/72
148
95
81
162
108
40
8, Cargo Area, 1/15/72
1
3
21, Eastern Perimeter of
14
£.
189
105
62
188
73
60
131
147
Airport, 1/7/72
99
44%
47
57
46
60
40
4%
2
12%
33, 34, 35, Suburbs Around Airport
<1
<1
<1
<1
<1
<1
55
1
14
14
49
12
76
2
16
144
71
94
130
130
<1%
<1
<1
<1
<1
<1
44
1
10
10
* By subtraction
-------
-41-
It would have been interesting to compare the fraction of the NOx
attributable to aircraft and the fraction of the NOx that was in the
form of NO at each site. Unfortunately, Argonne did not do the modeling
calculations that would be necessary for this comparison.
Argonne did use its AVAP model of Washington National to determine the
NOx contribution from aircraft. Argonne predicted the NOx concentration
from aircraft and the total NOx concentration for a grid of points at
National for seven different cases. On two "average-case" days 24-hour
NOx levels from aircraft were predicted to exceed the 100 y g/m level at
the ends of runways. NOx levels from all sources exceeded the level of
the standard by factors of 2 to 4 over the entire 6.5 square mile grid.
On a worst case day, a few more grid points show high 24-hour levels due
to aircraft, including a few points over the Potomac River. Argonne's
cases included some hourly averages. The grid point values for these
can be found in the Argonne report. Generally, with moderate to peak
aircraft activity and poor ventilation, NOx levels due to aircraft
emissions can exceed the 100 ug/m over the runways and terminal areas
and over a mile downwind of the airport boundary. It appears that
pockets with high levels of NOx, four or more times the 100 yg/m
level, can occur on the airport due to aircraft. These are also periods
of high NOx levels due to other sources. Argonnefs predictions for
total NOx concentrations include values up to twice as high as the
maximum N0_ levels measured by Geomet and summarized in Table 5.
Figures 20a and 20b,show the first of several NOx predictions made by
the Atlanta study. NOx emissions from aircraft at Atlanta in the year
of the study were roughly one-half to two-thirds those of O'Hare, JFK,
or L.A. International. The changes in the isopleths from Figure 20a to
20b were the result of a hypothesized 45% decrease in aircraft NOx
emissions. The reduction affects only the isopleths near the starting
ends of runways. Isopleths at the airport boundary are not affected at
all, indicating that aircraft NOx emissions make only a small contribution
at the boundary. As can be seen in Figure 20a, under normal conditions
hourly NOx levels are not very high. Figures 21a and 21b show that
downwind of the airport, total NOx emissions from all airport sources
(80% due to aircraft) do not raise ambient levels above 100 yg/m .
This level only occurs at the downwind edge of the airport. NOx levels
due to airport sources«drop rapidly with distance; at 8 kilometers, the
level is about 20 y g/m .
Under worst case conditions consisting of adverse weather and heavy
activity, total NOx concentrations on the airport are high enough to
raise concern about violations of a short-term N0» standard, Figure 22.
However, the model prediction is that with the exception of sites at the
end of the runways virtually none of these high concentrations can be
attributed to aircraft, even less than could be attributed to aircraft
under normal conditions. The low wind speed and shallow mixing depth
may explain this. Unfortunately, Reference 14 does not give a graph
like that of Figures 21a, b for the worst case conditions. The report
-------
AU. CONCENTRATIONS IN i.G/M3, 1-HOUR AVERAGE
Fig- 20a. Airport NO Cor.ccntrations for Baseline Conditions - Atlanta
ALL CONCENTRATIONS IN uG/M3, 1-HOUR AVERAGE
Fig- 20b. Airport N0x Concentrations for Engine Emission Standards- Atlanta
-------
z
o
-14 -12
228°(SW)
-10 -8 -6
UPWIND
-4
-43-
10
DOWNWIND
12 14
48°!NE)
DISTANCE FROM AIRPORT LOCATION POINT (ALP),km
Fig. 21a. Wind Line
Atlanta
NOx
Profiles Under Baseline Conditions, Summer -
300
0) . 2
GO i-*j
? 8
H CC
0) t
3 §
-------
.U/l
Fig. 22 Airport NO Concentrations for Worst Case Situation, All Sources
3
All Concentrations in u g/m , 1-Hour Average
*Concentration Attributable to Aircraft
- Atlanta
-------
-45-
does indicate that at 6 kilometers downwind of the airport the NOx
concentration from aircraft is about 100 yg/m and at 14 kilometers it
is 88 yg/m . These levels seem inconsistent with the near zero levels
predicted on the airport itself, but do seem consistent with the normal
case of Figures 21a, b.
Annual average NOx concentrations from all sources also were predicted
for the Atlanta airport. The full results can be found in the report
from the study or in Reference 1. Briefly, annual average NOx levels
were above the standard for N0~ at two locations on the airport, both
near runways. Several other points were within 20% of the standard and
the entire airport was above 50 y g/m . The aircraft contribution to
annual NOx levels outside the airport was only a few percent of the N09
standard.
The Salt Lake City airport has only 15% of the aircraft NOx emissions
inventory at Atlanta, but different weather conditions. Using its own
model, Geomet predicted annual average NOx concentrations for the airport
and surrounding area. Figure 23 is taken from the Geomet report. The
figure distinguishes between airport and non-airport contributions to
the total concentrations, but does not separate the airport contribution
into aircraft and non-aircraft. Assuming an aircraft/non-aircraft split
similar to other airports, the aircraft contribution at the Salt Lake
City is greater than at Atlanta. The high levels on an annual basis
suggest that there is reason for concern about possible violations of a
short-term N0« standard on and near the airport, depending on the
relative speeds of the oxidation and dispersion processes.
In summary: NOx levels due to aircraft have been predicted to reach
several times the 100 yg/m level at several commercial airports, on an
annual average basis. The highest levels occur near the ends of runways,
a pattern consistent with the fact that aircraft NOx emissions occur
mostly during runway modes. The predictions indicate that under adverse
conditions aircraft can cause short-term NOx levels above 100 yg/m a
mile or more downwind of the airport boundaries. Under these conditions
other NOx sources also contribute substantially to very high NOx levels.
Measurements confirm that ambient N0~ levels from all sources can reach
above 600 yg/m during short periods on the airport itself. The fraction
of NOx that is in the form of N0~ has been shown by measurements to be
lower on the airport than in surrounding areas.
-------
Figure 23.
Mean Annunl Predicted Values of NOX showing contributions for airport sources (SLCIA) and non-
airport sources (SLC), and totals, in micro grams/cu. m. Contoiir lines are drawn for totals.
-------
-47-
DISCUSSION
Conclusions Regarding Air Quality Violations On and Near Airports
Having reviewed the results of the individual studies it is possible to
draw together conclusions about air quality violations in the vicinity
of the large commercial airports. In doing so the areas where past
studies have been inconclusive will also be apparent.
Carbon Monoxide
Violations of the CO standards have been monitored at the L.A. and
Washington airports in areas where aircraft and ground service equipment
would seem to be the only significant nearby sources. Violations due to
aircraft alone have been predicted at these two airports and at O'Hare
and JFK. Both the predicted and monitored sites with high CO include
terminal areas where persons have access. Thus, there are at least a
few sites at airports where meeting the CO standards requires aircraft
emission reductions.
However, there may be more than these few. CO levels have been monitored
at only a few sites at a couple airports. The model predictions do not
account for obstructions to dispersion which would tend to increase CO
concentrations. The model predictions for the patterns of CO levels do
not have much spatial resolution except for Atlanta, which apparently
does not have any CO problems. Thus it is difficult to tell how much of
a terminal area experiences violations. The studies to date do not give
much indication of how often each airport has CO problems due to aircraft.
And there are many airports which have not been studied at all. Given
the known tendency for CO levels to depend on very local emission and
dispersion conditions it would not be surprising if smaller airports
also had CO problems.
The past studies have also not given a consistent projection of future
CO problems. CO emissions have been dropping at some of the airports as
a result of fleet mix changes, but projections beyond 1985 are fragmentary.
The general indication from the projections that do exist is that CO
emissions will either remain fairly constant or increase if no standards
are imposed. This suggests that CO problems are not going to correct
themselves naturally. It would be desirable to know more accurately how
CO emissions will change in the future for each of the top commercial
airports. This would give a better indication of whether whatever CO
problems do exist now will or will not be alleviated without new controls.
-------
-48-
Hydrocarbons
It is clear that aircraft emissions result in hydrocarbon concentrations
on the airports that are very high compared to the 160 yg/m standard.
Although the highest of these concentrations occur only over small areas
of the airports, the region of concentrations greater than 160 yg/m
can extend several miles past the airport boundaries. When emissions
from all sources are considered, the entire area around a major commercial
airport can experience very high HC levels.
However, the standard for HC is intended only as a guide in meeting the
standard for oxidants. In addition to HC concentrations, time, sunlight,
and the presence of NC- are required for oxidants to form. Most commercial
aircraft operations take place during daylight. Scheduled operations
pick up about 6 am at major airports and continue fairly steadily until
8 or 9 pm. Thus sunlight is not in short supply. Neither is NQo*
since aircraft emit NOx as well. The Systems Applications study
found that the HC/NOx mix in aircraft emissions was in the ozone-producing
range. Ozone production is enhanced when aircraft emissions mix with
automobile emissions. The time required for the photochemical reactions
does affect the conclusions that can be drawn about aircraft-caused ..
oxidant problems. The time required is on the order of a few hours.
In this time a given mass of HC moves downwind and disperses to a lower
concentration than indicated in the model prediction. There are, however,
predicted cases in which high concentrations exist sufficiently far
downwind that the HC is concentrated above 160 yg/m for over an hour.
Given the possible errors in the predictions, this suggests that there
may be areas downwind of the airports that experience oxidant violations
due to aircraft alone. These areas are probably fairly restricted. Too
near the airport the pollutants have not had enough time to react; too
far away the oxidant has dispersed to below the level of the standard.
Nitrogen Dioxide
Concentrations of NOx due to aircraft are high over the airports and
significant portions of surrounding areas. High N02 levels have been
monitored at airports, but these may have been due to NOx sources
upwind of the airport rather than to emissions on the airport itself.
Most of the aircraft NOx is emitted as NO and must oxidize to NO,,. As
with HC, whether NO- concentrations reach high levels depends on the
relative swiftness of the oxidation reactions and the dispersion process.
A portion of the NO emitted by aircraft each year escapes conversion
while at the airport because it is emitted at night. As Table 7 shows,
35-50% of scheduled operations occur in darkness on days in autumn and
15-30% in days in spring. NO emitted during winter months tends more to
disperse before oxidizing because lower temperatures and sunlight
intensity result in lower reaction rates. And some NO disperses before
it oxidizes even under the most favorable conditions. The EPA Office of
Air Quality Planning and Standards has judged that these factors likely
combine to result in no violations of the annual N0» standard due to
aircraft emission alone.
-------
-49-
Table 7. Daylight Scheduled Operations As
Percentage of Total Scheduled
Operations At Five Airports
Airport
Washington National
J.F.K.
L.A. International
O'Hare
Atlanta
Scheduled Aircraft Operations During
Indicated Daylight Hours* as Percentage
of Scheduled Operations On Given Date
November 2, 1973
65 (6:30am-5:00pm)
47 (6:30am-5:00pm)
60 (6:00am-5:00pm)
53 (6:15am-4:45pm)
55 (6:45am-5:45pm)
Source: Reference 22
*Daylight hours are approximate local time.
May 3, 1974
85 (6:00am-8:00pm)
70 (5:45am-7:45pm)
74 (6:15am-7:30pm)
81 (5:45am-7:45pm)
76 (6:45am-8:15pm)
-------
-SO-
On a short-term basis high N0_ levels due to aircraft are more likely to
occur since conditions favorable to rapid oxidation and slow dispersion
will occur occasionally for short periods. The review of the monitoring
and modeling results noted several cases in which short-term NOx concen-
trations were very high. If these occur under conditions favorable to
rapid oxidation, short-term NO- levels could exceed those that may be
specified by EPA if it issues a new ambient standard. The issue of NO
to N0~ conversion is the one least resolved by the past air quality
studies and most open to speculation at the present time.
The Regional Perspective
The air quality impact of aircraft emissions alone is only one of the
important considerations. There are and will be for some time other
emission sources in the region around the major commercial airports.
Reference 23 indicates that even with the controls which are now evisioned
most cities will still experience violations of the oxidant standard in
the year 2000 and the several largest cities will experience NO-
violations in addition. Thus, even when aircraft would not cause violations
they contribute to the severity of already existing violations. Since
CO problems are usually quite local and due to motor vehicle traffic,
aircraft CO emissions are not of much regional concern. On the airports
there may be sites where aircraft CO emissions plus access traffic CO
emissions combine to cause localized air quality problems. Oxidants and
N0_ levels, on the other hand, have been found to be more uniform in
urban areas, indicating better mixing of emissions. Thus the HC and NOx
emissions from aircraft are likely to be contributing to air quality
problems over much of their urban areas. This is particularly true for
regions like Los Angeles, where more than one-half of the peak NO
levels on high NO- days are carried over from the previous day. The
combination of the different night and day mixing patterns would tend to
spread the emissions from any source over a larger area than indicated
in any of the airport model predictions.
From a regional perspective, the size of aircraft emissions, the amount
they can be reduced, and the cost-effectiveness of controlling them
relative to controlling other sources are the three items of interest.
Figures 1-4 indicated the range of disagreement about the size of the
inventories during the 1970s. References 14, 19, 25, and 26 project
emissions past 1980 for some airports; only the latter two consider the
case of no aircraft emission standards. The trend shown in Figures 1-4
is that NOx has been increasing during the 1970s and HC and CO decreasing.
Part of the HC and CO reduction was due to the almost completed smoke
retrofit programs. Changes in fleet mix contributed to the trend also.
Figures 24 and 25 show what will happen to the past trend after 1980 at
two airports. The figures agree that without controls the decline in HC
will end by about 1990 when increases in fleet size overwhelm the trend
to larger, cleaner engines.
-------
-51-
12
10
0
1'jr-,
o 2
19110
I'J.-J
19"0
!995 2000
2UCO
Recommended
Current
1-7S.)
Figure 24. Projected Aircraft Emissions to the
Year 2000 at J.F.K.
-------
-52-
CARBON MONOXIDE
HYDROCARBONS
o
§ 40,000
I
cc
£ 20,000
o
5:
UI
GROUND SOURCES
oo
o
o
I
Of.
<
Ul
>
a:
LJ
a.
10
z
o
s
LJ
20,000
10,000
TOTAL
1970 1980 1990
YEAR
2000
1970 1980 1990 2000
YEAR
NITROGEN OXIDES
ca
o 20,000
o
o
i
o;
LLf
o: 10.000
LU
O-
uo .
O
TOTAL-
1970 1980 1990 2000
YEAR
Figure 25. Projected Emissions to the Year
2000 at Los Angeles International
-------
-53-
EPA defines a major point source as one with emissions of any one pollutant
greater than 100 tons/year. By this definition, aircraft operating at
the major commercial airports must be considered very major sources:
the estimates shown in Figures 1-4 are in thousands and tens-of-thousands
of tons/year. Yet, commercial aircraft contribute only about a percentage
point or less of the total emissions in any air quality control region.
Tables 8-10 present some data that give a feel for the absolute and
relative importance of commercial aircraft emissions. The general
conclusion that can be drawn is that commercial aircraft are a significant
source of emissions even though their percentage contribution may seem
small. This is because a large part of any air quality control region's
emissions comes from many small sources.
While total emissions give a first indication of the importance of
aircraft emissions, it is the size of the achievable reductions that
determine whether controls are administratively worthwhile. Three
studies have addressed this question. The Atlanta study calculated
the effect the 1979 standards would eventually have on annual aircraft
emissions at Atlanta if traffic volume and fleet mix stayed constant.
HC, CO, and NOx reductions were 71%, 62% and 45% respectively. These
estimates do not hold exactly for the proposed amendments to the standards,
because the study accounted for reductions in emissions from non-
commercial aircraft and the standards it used for commercial aircraft
were somewhat more stringent than the proposed amendments. Figure 24
includes projections of aircraft emissions to the year 2000 for JFK for
various control scenarios, as taken from Reference 25. Note that the
timing of the emission reductions depends on the scenario. The HC and
CO retrofit ("Recommended" curves) advances the date of maximum reduction
for those pollutants. Reference 27 estimated the cumulative, national
reductions that would result from the proposed amendments. Tables 11
and 12 compare these estimated reductions from aircraft standards to
those from other control strategies. Again, the potential reductions
from the aircraft standards are small but not insignificant in comparison.
27
EPA has estimated in a separate report the cost effectiveness of the
engine modifications that will result from the proposed amendments and
compared its estimates to the cost effectiveness of other control strategies.
One of the conclusions reached was that the modifications that will be
needed to comply with the proposed 1981 HC and CO requirements for newly
manufactured engines and the 1985 HC and CO requirements for in-use
engines are cost-effective ways of reducing regional HC emissions. Not
enough information on the operating costs associated with the proposed
1984 HC, CO, and NOx requirements is available at present to reach a
conclusion on the cost-effectiveness of aircraft NOx controls.
-------
-54-
Table 8. Commercial Aircraft Emissions As
Percentage of Total Air Quality
Control Region Emissions
AQCR
Los Angeles
San Francisco
NY-NJ-Conn
Chicago
St. Louis
Cincinnati
Baltimore
Boston
Houston
S.E. Wisconsin
Washington, B.C.
Atlanta
Percent of AQCR Emissions
Attributable To Commercial Aircraft
CO
0.22
0.37
0.32
0.19
0.34
0.14
0.32
0.35
0.32
0.19
0.46
1.08
NOx
0.51
0.94
0.53
0.55
0.30
0.29
0.56
0.66
0.47
0.34
0.89
2.36
HC
0.41
0.84
0,65
0.67
0.63
0.47
0.41
0.64
0.39
0.18
1.11
2.50
Source: Reference 28
-------
-55-
Table 9. Comparison of Aircraft Emissions With Emissions
From Other Mobile Sources - Three Air Quality
Control Regions
AQCR
Chicago
New York
Detroit
(summer only)
Mobile Source
All Aircraft*
Locomotives
All Aircraft*
Merchant Vessels
All Aircraft*
Outboards
HC
T/yr
6370
4030
7850
20580
2325
8440
CO
T/yr
14280
4780
19230
21920
5360
21640
NOx
T/yr
4340
12860
5390
39240
1289
40
*Includes military, civil, and commercial aircraft. Commercial aircraft
dominate in all three AQCR's.
Source: Reference 29.
-------
-56-
Table 10. 1973 HC Emissions From Selected
Sources in Baltimore AQCR
Sources Tons of HC
Light-Duty Vehicles and Trucks 86,200
Heavy-Duty Trucks 16,000
Solvent Evaporative Loss 9,700
Gasoline Handling Loss 8,600
Solid Waste Disposal 3,950
Lawn and Garden Equipment 2,000
Vessels 1,300
Off Highway Vehicles 1,200
Locomotives 1,200
Commercial Aircraft 1,023
Residential Fuel 744
Sources: References 20 and 28
-------
-57-
Table 11. Comparison of National HC Reductions Achievable
With Various Control Strategies, Including
Proposed Commercial Aircraft Controls
Strategy: HC
Source and HC Reduction*
Degree of Control 1000 T/yr
Degreasing 0-40% 300
Degreasing 41-90% 360
Dry Cleaning 0-80% 480
Refining 0-67% 1530
Arch. Coatings 0-100% 1000
Open Burning 0-25% 300
Paint and Varnish 0-70% 90
Metal Decorating 0-90% 140
Letterpress and Lithography 0-90% 270
Gasoline Handling 16-50% 570
Gasoline Handling 51-91% 680
Gasoline Distribution 67-99% 430
Gas HDV Evap. 5.8-0.5 g/mi 580
LDV Exhaust 0.9-0.41 g/mi ' 930
LOT and Gas HDV Exhaust 2.9-0.8 g/mi 840
Commercial Aircraft 1981 NME 13.7
Commercial Aircraft 1985 retrofit 35
Commercial Aircraft 1984 NME 103
* For all but aircraft controls, this figure is the reduction in 2000
achieved by imposition of controls now. For 1981 NME controls, this
figure is the average annual reduction from 1984 to 1996. For 1985
retrofit controls, this figure is the reduction in 1985; the reduction
declines as these already worn engines are retired. For 1984 NME
controls, this figure is the reduction in 2000.
Source: References 23 and 27.
-------
-58-
Table 12. Comparison of National NOx Reductions Achievable
with Various Control Strategies, Including Proposed
Commercial Aircraft Controls
Strategy: NOx NOx Reduction
Source and in Year 2000
Degree of Control 1000 T/yr
New Utility Boilers 25-50% 3520
Industrial Boilers 25-65% 2460
Old Utility Boilers 25-50% 620
Stationary 1C Engines 25-75% 2870
Other Mobile 10-2.0 g/mi 2900
LDV 2.0-1.0 g/mi 1910
Utility Boilers 50-90% 6620
Stationary 1C Engines 75-90% 860
LDV 0.4 g/mi 1150
Commercial Aircraft 1984 NME 48.4
Sources: References 23 and 27
-------
-59-
When comparing emission reductions and cost effectiveness for mobile and
stationary sources it is important to note where the reductions will be
achieved. State control of stationary sources can be more selective
than EPA control of mobile sources. Some of any reduction in mobile
source emissions occurs in areas that do not have air quality problems.
But comparatively, commercial aircraft operations tend to take place in
major metropolitan areas which tend to be in "problem" air quality
control regions. According to Reference 28, 79% of commercial.,aircraft
HC emissions took place in the 79 air quality control regions which
experienced oxidant violations in 1973/1974. In comparison, only 64% of
light-duty vehicle HC emissions take place in these regions. Twenty-one
percent of commercial aircraft NOx emissions take place in the seven
regions which had N0? violations versus 18% for light-duty vehicles.
Compensating the disadvantage of uniform mobile source control compared
to selective stationary source control is the administrative advantage
of setting and enforcing national standards.
In summary; Even if emissions from commercial aircraft alone do not
cause violations of the national ambient air quality standards, they do
contribute significantly to present, and future, violations of the
oxidant and NO- standards in many air quality control regions. The HC
requirements ot the proposed amendments to the aircraft standards are a
cost-effective approach to oxidant control. Not enough information is
available to reach a conclusion on the cost-effectiveness of the NOx
requirements.
The Planned FAA Air Quality Study
FAA is planning to conduct a new airport air quality study, beginning
during the comment period for the proposed amendments. This study will
be most useful if it is aimed at complementing the past studies. The
findings and failings of the past studies should be considered in
setting the goals of the new study.
The FAA study is planned to take 18 months to complete. The study will
apply the AVAP model to several airports, the same model used in the
more recent past studies. FAA intends to improve on past studies by
accounting for non-airport emission sources better. This may allow more
convincing validation than with past studies. EPA will participate and
intends to give more consideration to issues of photochemical reactions
than has been given in past studies. New monitoring data will be collected.
Details beyond these are not yet available.
The new study will be the first since the 1971 study by NREC to consider
more than one airport. The model it will use is considered more reliable
than NREC's. The study may eventually be considered definitive, depending
on how many of the unanswered questions it addresses for each airport.
-------
-60-
The conclusions that can be reasonably drawn from the discussion of the
past studies are (1) aircraft emissions must be controlled to correct CO
violations now occuring at a few and possibly more terminal sites, (2)
there probably are no annual N0? violations attributable to aircraft
alone, (3) there may well be violations of a short-term N02 standard
attributable to aircraft, but there is still a large amount of uncertainty
regarding this, (4) there may be areas downwind of airports that experience
oxidant violations due to aircraft alone, (5) aircraft HC emission
control is a cost effective approach to regional ambient oxidant control,
and (6) aircraft NOx emission control has an uncertain cost effectiveness.
The new study will contribute to the development of the aircraft regulations
only if it affects these six conclusions.
The cost effectiveness conclusion for HC control follows from an analysis
of the control technology and not from findings on airport air quality.
Consequently, a new air quality study will not change it. The cost
effectiveness of NOx control can only be resolved by more complete cost
data. A new study could give consistent HC and NOx emission forecasts
for a number of airports and control scenarios. This would be an improvement
over the few forecasts available now and would give a more exact estimate
of the potential improvement in each air quality control region. While
better information of this sort is always desirable, the difference
between it and EPA's present estimates will probably not be great.
CO is an area where a new study could possibly have an effect. The
point of present uncertainty is on how extensive and frequent CO violations
from aircraft are and will be. A new set of model predictions may not
resolve this uncertainty. The AVAP model does not account for obstructions
to dispersion, and these could be important to CO levels. The AVAP
study of Washington National did not predict the CO violations monitored
by Geomet, for example. Thorough CO monitoring in the terminal areas of
several airports would indicate the severity of the CO problem, although
the importance of the aircraft contribution could only be inferred from
the geometry of each situation. This monitoring would be expensive but
the cost would be small compared to the cost of the HC/CO controls.
The possibility of localized oxidant violations attributable directly to
aircraft emissions is a minor consideration given the certainty that HC
control is a cost-effective approach to reducing regional oxidant
concentrations. Thus, there is not much reason for a new study to spend
much effort on the question of oxidant violations attributable to aircraft.
The area of present uncertainty with the greatest potential of being
brought closer to resolution by a new study is the one concerning short-
term N0_ levels. The uncertain cost-effectiveness of aircraft NOx
controls and the relatively few regions which experience annual N0«
-------
-61-
violations make the efficiency of the controls simply as a way of achieving
regional annual average NO- air quality uncertain. If aircraft NOx
controls are needed to prevent or reduce violations of a short-term NO-
standard, that will be the stronger argument for the controls. Thus the
need for NOx controls may hinge on short-term NCL. This is the area
most open to speculation at present. The AVAP model to be used in the
new study does not predict short-term N0« levels. But more relevant
analysis of the AVAP results could be made than has been in past studies.
Monitoring of NO and N0~, when combined with model predictions on the
fraction of NOx from aircraft, may shead more light on how much of the
aircraft NO oxidizes the NO- before dispersing. Better inventories for
non-airport NOx emissions should improve model predictions of the fraction
of NOx from aircraft. The actual NO/NO- composition of aircraft exhaust
should be determined also, either through measurement or kinetic modeling.
The NO/NO composition is known for some engines but not all. No
studies have been made of how much additional oxidation of NO to NO.
takes place in the high temperature, high concentration exhaust plume.
In summary; The new FAA study should devote its attention to resolving
present uncertainties concerning the effect of aircraft emissions on
localized, short-term CO and N0? concentrations. Monitoring will be just
as crucial to resolving these uncertainties as will dispersion modeling.
-------
-62-
CONCLUSIONS
4
The 1971 study by Northern Research and Engineering was the most
comprehensive of the airport air quality studies. Unfortunately it was
also probably the least accurate, certainly the least demonstrably
accurate. If the argument for emission controls on commercial aircraft
depended only on the 1971 NREC study and the 1972 EPA report it would
be weak. However, more recent studies by both EPA and FAA have confirmed
some of the NREC conclusions and added others. These have been summarized
in the Discussion section of this report.
Very briefly, CO violations attributable to aircraft are occurring at at
least a few terminal gate areas and at the ends of runways. The areas
of high CO attributable to aircraft are not extensive, but can include
areas where persons have regular access. HC emissions from commercial
aircraft create regions of high HC concentrations that include the major
commercial airports and the areas several miles downwind. NOx emissions
create smaller regions of high NOx concentrations, but ones that still
extend beyond the airport boundaries.
The past air quality studies do not provide an answer to the question of
whether N0« and oxidant violations can be attributed to aircraft. This
is because they have not been able to consider the photochemical processes
by which these criteria pollutants are formed from HC and NOx. It seems
reasonable that on an annual basis most of the aircraft NOx disperses
from the airport before it can oxidize to N0~. On a short-term basis,
only judgment is available at present and that suggests that the chances
of short-term violations are much better than for annual violations.
Oxidant violations on the airport due to aircraft could only occur under
stagnant conditions. With a wind, the area of violation would be well
downwind of the airport. However, it is reasonable to assume that even
in the relatively far future (at least to the year 2000) aircraft HC
emissions will be in addition to those from other sources and that the
total will cause widespread oxidant problems in most American cities.
Aircraft HC controls are a cost effective way of helping to reduce this
oxidant problem. In particular, the EC/CO retrofit program is the only
control strategy which reduces aircraft HC emissions in the near future.
The desirability of the modifications that would be forced into use
by the HC standards is reasonably well demonstrated in the results of
the past studies. Further study of HC will not contribute to the develop-
ment of the amendments to the existing standards. There is still some
uncertainty about the number of localized areas which are experiencing
CO problems because of commercial aircraft. Monitoring of CO levels at
a number of airport terminals would help resolve this uncertainty. There
is a great amount of uncertainty and speculation about the need for
aircraft NOx controls, as this largely hinges on the contribution of
-------
-63-
aircraft to short-term N02 levels. Past studies have not considered
this question in any depth.
The new air quality study should attempt to resolve the two areas of
present uncertainty by concentrating on short-term, localized CO and NCL
air quality. CO monitoring will be more definitive than CO modeling.
Monitoring should also include both NO and N0» in order to help resolve
the questions of how quickly aircraft NO oxidizes to NO-. A study which
is not designed to effectively address the CO and NO- questions will be
a minimal contribution to the development of the aircraft amendments.
-------
-64-
REFERENCES
1. Potential Impact of NOx Emissions from Commercial Aircraft on
N02 Air Quality, prepared by Bruce C. Jordan, EPA Office of Air
Quality Planning and Standards, Research Triangle Park, North
Carolina, November 15, 1977.
2. Nature and Control of Aircraft Engine Exhaust Emissions, Northern
Research and Engineering Corporation, Report No. 1134-1, Contract
PH 22-68-27, November 1968.
3. Study of Jet Aircraft Emissions and Air Quality in the Vicinity of
the Los Angeles International Airport, Air Pollution Control District,
County of Los Angeles, Contract CPA 22-69-137, April 1971.
4. The Potential Impact of Aircraft Emissions Upon Air Quality,
Northern Research and Engineering, Report No. 1167-1, Contract
No. EPA-68-02-0085, December 29, 1971.
5. Aircraft Emissions: Impact On Air Quality and Feasibilty of
Control, EPA, 1972.
6. Report No. 7, Argonne Airport Vicinity Air Pollution Study, Center
for Environmental Studies, Argonne National Laboratory, Interagency
Agreement No. DOT-FA71WI-223, December 4, 1972.
7. Model Verification - Aircraft Emissions Impact On Air Quality,
Geomet Incorporated, EPA-650/4-74-049, September 1974.
8. Air Quality Assessment Model Applied to Washington National
Airport, Lawrence E. Wangen and Lester A. Conley, Energy and
Environmental Systems Division, Argonne National Laboratory,
Interagency Agreement Project Order AFWL 75-PO-T-071.
9. Impact On Air Quality of the Proposed Expansion of the Salt Lake
City International Airport - Environmental Impact Statement,
Scott D. Thayer, Geomet Inc., Report No. EF-306, March 29, 1974.
10. Air Quality Analysis of a Proposed Cleveland Airport Lake Site -
Final Report, Jonathan D. Cook and Robert C. Koch, Geomet Inc.,
Report No. EF-557, August 1976.
11. Evaluation of Air Quality Impact for Alternative Cleveland Jetport
Sites - Final Report, Eric R. Sandey and Scott D. Thayer, Geomet
Inc., Report No. EF-471, May 19, 1975.
12. Airport Vicinty Air Pollution Study, D. M. Rote et. al., Energy
and Environmental Systems Division, Argonne National Laboratory,
Report No. FAA-RD-73-113, December 1973.
-------
-65-
13. Airport Vicinity Air Pollution Study - Model Application and
Validation and Air Quality Impact Analysis at Washington National
Airport, I.I. Wang, et. al., Energy and Environmental Systems
Division, Argonne National Laboratory, July 1974.
14. An Evaluation of Strategies For Airport Air Pollution Control,
R.R. Cirillo, et. al., Energy and Environmental Systems Division,
Argonne National Laboratory, January 1975.
15. Air Quality Impact of Aircraft At 10 USAF Bases, Dennis F. Naugle,
et. al., Air Quality Research Division, Tyndall AFB, Florida,
Report No. CEEDO-TR-76-23, April 1977.
16. An Ambient Air Quality Model For Assessment of U.S. Naval Aviation
Emittants, G.R. Thompson and D.W. Netzer, Naval Postgraduate School,
Monterey, California, Report No. NPS-67Nt76091, September 1976.
17. Investigation of Soils and Vegetative Damage: Vicinity of Hartsville,
Atlanta's International Airport, K.W. Brown, et. al., Monitoring
and Support Laboratory, U.S. E.P.A., Las Vegas, Nevada.
18. Introductory Study of the Chemical Behavior of Jet Emissions in
Photochemical Smog - Final Report, G.Z. Whitten and H. Hogo,
Systems Applications, Inc., Contract No. NAS2-8821, May 1976.
19. Technical Support Report - Aircraft Emissions at Selected Airports
1972-1985, Standards Development and Support Branch, Emission
Control Technology Division, Office of Air and Waste Management,
U.S. E.P.A., Report No. AC 77-01, January 1977.
20. An Assessment of the Potential Air Quality Impact of General
Aviation Aircraft Emissions, Bruce C. Jordan, Office of Air Quality
Planning and Standards, U.S. E.P.A., June 17, 1977.
21. Analysis of Aircraft Exhaust Emission Measurements, Cornell
Aeronautical Laboratory, Contract No. EPA-68-04-0040, October 1971.
22. Profiles of Scheduled Air Carrier Airport Operations by Equipment
Type - Top 25 U.S. Airports November 1973 and May 1974, Office of
Aviation Policy, Policy Development Division, Federal Aviation
Administration, September 1974.
23. Air Quality Noise and Health - Report of a Panel of the Interagency
Task Force on Motor Vehicle Goals Beyond 1980, Interim Report,
March 1976.
24. Control Strategies For Oxidant and Nitrogen Dioxide - A Report on
the ARE Photochemical/Transport Workshop at UCLA, January 6-7, 1977
and an Evaluation of the Need for the Control of Light-Duty Vehicle
NOx Emissions to 0.4 Grams/Mile, Air Resources Board, State of
California, January 25, 1977.
-------
-66-
25. Major Issues To Be Resolved Regarding the Aircraft Engine Emission
Standards, Office of Mobile Source Air Pollution Control, U.S.
E.P.A., 1977.
26. Airport Emissions Through the Year 2000, Howard M. Segal, Boeing
Company, Pacific Northwest International Section - Air Pollution
Control Association, Paper No..73-AP-48, November 30. 1973.
^ j
27. Cost Effectiveness Analysis of the Proposed Revisions in the Exhaust
Emission Standards for New and In-Use Gas Turbine Aircraft Engines,
Technical Support Report for Regulatory Action, Richard S. Wilcox
and Richard Munt, Standards Development and Support Branch, Emission
Control Technology Division, Office of Air and Waste Management,
U.S. E.P.A., Reports No. AC 77-02 and AC 78-01, January 1978.
28. 1973 National Emissions Report, Office of Air Quality Planning
and Standards, U.S. E.P.A., Report No. EPA-450/2-76-007, May 1976.
29. Gaseous Emissions from Unregulated Mobile Sources - Final Report,
Donald E. Zinger and Lawrence H. Hecker, University of Michigan
School of Public Health, Department of Environmental and Industrial
Health, prepared for Emission Control Technology Division, EPA,
October 1976.
------- |