AIRCRAFT EMISSIONS:
IMPACT ON AIR QUALITY
AND
FEASIBILITY OF CONTROL
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
AIRCRAFT EMISSIONS:
IMPACT ON AIR QUALITY
AND FEASIBILITY OF CONTROL
ENVIRONMENTAL PROTECTION AGENCY
Office of Air Programs
Research Triangle Park, North Carolina
April 1972
-------
Office of Air Programs Publication No. APTD-0757
n
-------
PREFACE
This report presents the available information on the present and predicted
nature and extent of air pollutant emissions from aircraft operations at major air-
ports in selected air quality control regions. In addition, an investigation on the
present and anticipated future technological feasibility of controlling such emis-
sions is presented. This report is published in accordance with Section 231 (a) of
the Clean Air Act as amended, which states:
"(1) Within 90 days after the date of enactment of the Clean Air Amendments
of 1970, the Administrator shall commence a study and investigation
of emissions of air pollutants from aircraft in order to determine--
"A. the extent to which such emissions affect air quality in air quality
control regions throughout the United States, and
"B. the technological feasibility of controlling such emissions.
"(2) Within 180 days after commencing such study and investigation, the
Administrator shall publish a report of such study and investigation..."
The data base for this report was developed largely by Northern Research and
Engineering Corporation, Cambridge, MassachusettsJ>2 under contract with the U.S.
Environmental Protection Agency (EPA). More detail on certain aspects of the study
is available in the contract reports.^ Further information on baseline emissions
from aircraft was obtained under several other contracts.
111
-------
CONTENTS
LIST OF FIGURES vij
LIST OF TABLES vii
INTRODUCTION 1
TECHNOLOGICAL FEASIBILITY OF CONTROLLING AIRCRAFT EMISSIONS 2
EFFECTS OF AIRCRAFT ON AIR QUALITY 2
CONCLUSIONS 5
CONTRIBUTION OF AIRCRAFT TO POLLUTANT CONCENTRATIONS 5
Air Quality Impact of Aircraft in Airport Areas 6
Air Quality Impact of Aircraft in Metropolitan Areas 8
EMISSION CONTROL OF AIRCRAFT TURBINE ENGINES 8
EMISSION CONTROL OF AIRCRAFT PISTON ENGINES 9
METHODOLOGY FOR IMPACT EVALUATION 11
SELECTION OF AIRPORTS 11
PROCEDURE OF AIR QUALITY ANALYSES AT STUDY AIRPORTS 11
Emission Factors 12
Activity Level 12
Methods of Impact Evaluation 14
RESULTS OF IMPACT EVALUATIONS 19
NATIONAL AMBIENT AIR QUALITY STANDARDS 19
RESULTS OF DISPERSION MODEL AIR QUALITY ANALYSIS 19
Predicted Concentrations Compared with Primary Air Quality
Standards at Air Carrier Airports 19
Predicted Concentrations Compared with Secondary Ambient
Air Quality Standards for Particulates and Sulfur Dioxide
at Air Carrier Airports 25
Results for General Aviation 27
Comparison of the Model's Predictions with Actual Air Quality Data ... 28
RESULTS OF OTHER ANALYSES OF AIR QUALITY IMPACT IN
AIRPORT AREAS 30
Emission Density Comparison 30
Analysis of Measured Carbon Monoxide Air Quality Data 30
Area Source Dispersion Model 33
Emission from Unburned-Fuel Dumping 36
FUTURE PROJECTION OF AIRCRAFT AND TOTAL EMISSIONS IN AIRPORT AREAS 37
-------
CURRENT AND PROJECTED CONTRIBUTION OF AIRCRAFT TO EMISSIONS
IN METROPOLITAN AREAS 38
TECHNOLOGICAL FEASIBILITY OF CONTROLLING AIRCRAFT EMISSIONS 41
EMISSION CONTROL BY ENGINE MODIFICATION 42
Engine Classification 42
Emission Control Methods and Effectiveness 43
Cost and Time Requirements for Control-Method Development
and Implementation 49
EMISSION CONTROL BY MODIFICATION OF GROUND OPERATIONS 53
Definition of Ground Operations 53
Emission Control Methods 54
Implementation Cost and Time Requirements 55
EMISSION MEASUREMENT TECHNOLOGY 56
Sampling and Test Procedures 57
Emission Measurement Instrumentation 57
REFERENCES 59
v1
-------
LIST OF FIGURES
Figure Page
1. Calculated Total Hydrocarbon Concentrations at Los Angeles Airport
and Downwind 35
LIST OF TABLES
1. Aircraft Classification System 13
2. LTO Cycles for 1970, 1975, and 1980 15
3. Frequency of Occurrence of "Worst Meteorological Conditions" for
High Carbon Monoxide and Hydrocarbons Concentrations at Study
Air Carrier Airports in 1970 17
4. National Ambient Air Quality Standards 20
5. Predicted Ambient Air Pollutant Concentrations from Aircraft Alone
and from Airport Vicinity Sources at Sites where Pollutant Concen-
trations Exceed Primary Air Quality Standards and where Public could
be Exposed for Time of Standard 22
6. Predicted Aircraft Contribution to Ambient Air Quality at Commercial
Air Carrier Airports, Compared with Primary Standards 24
7. Predicted Ambient Particulate and S0£ Concentrations from Aircraft
and from Airport Vicinity Sources at Sites where Secondary
Standards are Exceeded 26
8. Predicted Aircraft Contribution to Ambient Air Particulate and S02
Concentrations Compared with Secondary Sources 27
9. Comparison of Emission Densities for Airports Versus Urban Areas, 1970 . 31
10. Current and Projected Emissions from Aircraft and Airports 32
11. Comparison of Emission Densities for Airports Versus Urban Areas
for 1970, 1975, and 1980 39
12. Contribution to Total Emissions by Aircraft in Los Angeles Basin Area. . 40
13. Contribution to Total Metropolitan Emissions Using Major Airports
in Three Cities ; 40
14. Aircraft Engine Classification 42
15. Engine Modifications for Emission Control for Existing and Future
Turbine Engines 44
16. Effectiveness of tl - Minor Combustion Chamber Redesign - on Reduction
of Emissions from Turbine Engines 45
17. Effectiveness of Engine Modification in Control of Emissions from
Turbine Engines, by Operating Mode 46
18. Bases for Control Method Effectiveness Estimates for Turbine Engines . . 47
19. Engine Modifications for Emission Control for Existing and Future
Piston Engines 48
vii
-------
20. Current Uncontrolled Emission Rates for Piston Engines 49
21. Effectiveness of Engine Modifications in Control of Emissions
from Piston Engines, by Pollutant 50
22. Time and Costs for Modification of Current Civil Aviation Engines. ... 51
23. Cost Results for Turbine Engine Population by Separate Use Categories. . 52
24. Comparative Reductions Resulting from Control Methods Applied
at Los Angeles International Airport 55
25. Costs and Time for Operations Changes at Los Angeles International
Airport 56
26. Instrumentation for Measurement of Turbine Engine Emissions 58
viii
-------
AIRCRAFT EMISSIONS:
IMPACT ON AIR QUALITY
AND FEASIBILITY OF CONTROL
INTRODUCTION
Public awareness that aircraft were a source of air pollution developed in the
late 1950's with the introduction of turbine-engine aircraft. Visible exhaust
plumes from the engines and increased levels of exhaust odors at airports caused
complaints about the aircraft to be lodged. The complaints, in turn, stimulated
investigations into the nature and extent of aircraft emissions. The Air Quality
Act of 1967 specifically identified aircraft emissions as a subject of concern and
required an investigation by the Department of Health, Education, and Welfare. The
study,3 submitted to Congress on January 17, 1969, concluded, among other things,
that reduction of particulate emissions from jet aircraft was both desirable and
feasible.
In March 1970, at a meeting held by the Secretaries of Health, Education, and
Welfare and of Transportation, representatives of 31 airlines agreed to a schedule
for retrofitting JT8-D engines with reduced-smoke combustors, to be substantially
completed by the end of 1972. This agreement sought to significantly abate visible
(smoke) emissions from aircraft powered by this widely used engine.
i
The problem of defining the nature and extent of air pollution from aircraft
sources has received continuing attention as reflected by recently published studies
prepared by the Bay Area Air Pollution Control District4 and by the Los Angeles
County Air Pollution Control District under EPA contract5 and by the published pro-
ceedings*5 of the joint DOT/SAE Conference on Aircraft and the Environment.
To develop additional information helpful in determining the aircraft contribu-
tion to air pollutant concentrations and the feasibility of controlling aircraft
emissions, the Environmental Protection .Agency conducted studies in the following
areas.
-------
TECHNOLOGICAL FEASIBILITY OF CONTROLLING AIRCRAFT EMISSIONS
Information on emission control methods was necessary to determine the levels to
which aircraft emissions can feasibly be reduced. The specific objectives of the
investigation of aircraft emission control technology were:
1. To identify methods of controlling aircraft emissions through modification
of engines, fuels, and ground operation.
2. To estimate the effectiveness of these control methods in reducing aircraft
emission rates.
3. To estimate the time and cost of implementing these control methods.
4. To assess the technology of measuring emissions from aircraft engines, and
to identify areas where advancements in instrumentation or test procedures
are required.
It is felt that this analysis and the results obtained represent the best
available information on the present and anticipated future technological feasibility
of controlling aircraft engine emissions.
EFFECTS OF AIRCRAFT ON AIR QUALITY
While previous investigations had been made concerning the impact of aircraft on
air quality, additional study was necessary to assess the severity of this impact.
To provide additional insight on the localized impact of aircraft emissions, an
airport dispersion modeling study was undertaken.
The dispersion modeling provided estimates of air pollutant concentrations at
four major commercial airports - Los Angeles International, J. F. Kennedy, O'Hare,
and Washington National and permitted estimates to be made of that portion of
total pollutant concentrations attributable to aircraft alone. Pollutant concentra-
tions were estimated by the model at a total of 193 sites within a 5-kilometer
radius of the airports; concentrations at over 100 of these sites that are accessi-
ble to the public were used in the analysis of the results.
Four additional methods of evaluating localized and metropolitan impact of air-
craft on air quality were also used:
1. A limited analysis of the air quality data collected at Los Angeles
International Airport under EPA contract in 1970 was performed to further
evaluate the impact of aircraft CO emissions.
2. A simplified area-source modeling technique was applied to aircraft hydro-
carbon emissions at Los Angeles International Airport to estimate possible
aircraft contributions to oxidant concentrations downwind of the airport.
-------
Emission densities (tons of emissions per square mile per year) from the
four commercial airports were compared with emission densities from their
neighboring metropolitan areas to further investigate the potential of
localized impact of aircraft emissions.
Estimates were made of the present and future contributions of aircraft
emissions to the total emissions of the four metropolitan areas studied
(Los Angeles, New York City, Chicago, and Washington, D. C.).
-------
CONCLUSIONS
CONTRIBUTION OF AIRCRAFT TO POLLUTANT CONCENTRATIONS
The results of the several approaches used in the study to assess the contribu-
tion of aircraft to air quality provide consistent evidence that aircraft are impor-
tant contributors to air pollutant concentrations in excess of the primary (health-
related) and secondary (welfare-related) Federal ambient air quality standards in
localized areas of at least four major U.S. airports. These airports, Los Angeles
International, O'Hare, John F. Kennedy, and Washington National, are used by over
30 million passengers per year.
In addition to this evidence of an important aircraft air quality impact on a
localized basis around airports, the contribution by aircraft to the total emissions
from the metropolitan areas of New York City, Los Angeles, Chicago, and Washington,
D.C., may become increasingly important in achieving and maintaining air quality
standards, especially near airports in these regions, as other emission sources are
controlled.
Of the methods used to evaluate air pollutant concentrations, the dispersion
modeling provided the most useful information because of its ability to estimate air
pollutant concentrations caused by aircraft alone. Atmospheric dispersion models
similar to the one used in this study are widely accepted as a means of approximating
air pollutant concentrations, and have been specified by EPA as one method of demon-
strating that air quality implementation plans for metropolitan areas are consistent
with ambient air quality standards.
In estimating ambient air pollutant concentrations, the primary dispersion
modeling approach used in this study necessarily takes into account a great deal of
measured and estimated input data, including emissions, activity-use patterns, and
locations of the various aircraft types and other emission sources at the airport.
Meteorological conditions, including wind speed, wind direction, atmospheric sta-
bility, and mixing height are also input to the model, as are emissions from areas
surrounding the airport. The relationships between these input data and estimated
ambient air pollutant concentrations are based on empirical models derived from
actual measurements of pollutant dispersion patterns.
It is felt that the results of the dispersion modeling work used in this study
represent the best presently available estimates of aircraft contributions to air
-------
pollutant concentrations in the airport areas studied. Even so, the individual
concentrations predicted in these modeling results should be considered approximate.
Limited comparisons have been made between the model's predictions and air quality
data, but it is not possible at this time to state, in general, the extent to which
the model may underpredict or overpredict actual pollutant concentrations.
Although it is not possible at this time to assess with confidence the precise
impact of aircraft on air quality in airport environs, the total evidence from the
dispersion modeling and other methods of evaluation strongly support the general
conclusion that aircraft make important contributions to air pollutant concentra-
tions at airports and neighboring areas.
The following specific conclusions indicate the impact of aircraft emissions
on air quality at the four commercial airports considered in this study and in their
immediate vicinities.
Air Quality Impact of Aircraft in Airport Areas
Hydrocarbons and Oxidants Nonmethane hydrocarbon concentrations caused by aircraft
alone, as estimated by dispersion modeling, are estimated to be far in excess of the
hydrocarbon standard in a large number of airport areas. High hydrocarbon emission
densities from aircraft at these airports support these estimates. Despite sub-
stantial projected decreases in total aircraft hydrocarbon emissions at these air-
ports during the 1970's, ambient air hydrocarbon concentrations attributable to
aircraft alone are expected to continue to exceed the hydrocarbon standard in many
airport areas.
Although hydrocarbon concentrations observed in the atmosphere have not been
directly associated with health effects, the nonmethane hydrocarbon standard is
based on the role of hydrocarbons as a precursor of photochemical oxidants. Simple
area-source dispersion modeling analysis suggests that, under conditions particularly
conducive to high oxidant concentrations, aircraft-generated hydrocarbon concentra-
tions may alone result in oxidant concentrations in excess of the photochemical
oxidant air quality standard downwind of Los Angeles International Airport.
Dispersion modeling analysis indicates that fuel dumping accounts for maximum
ground-level, 1-hour-total hydrocarbon concentrations on the order of 18 pg/m3.
Hydrocarbon emissions caused by fuel dumping are estimated to range from 3.5 percent
of total aircraft hydrocarbon emissions at Los Angeles International Airport to 21
percent of those at Washington National Airport. Widespread complaints about oily
films on automobiles and other surfaces in airport vicinities suggest that conden-
sable exhaust hydrocarbon emissions and fuel dumping are at least partially respon-
sible for these films.
-------
Carbon Monoxide Carbon monoxide concentrations, as estimated by the dispersion
model, are predicted to exceed the carbon monoxide air quality standards in some
airport areas of public access. In these areas, contributions by aircraft to these
concentrations are estimated to be major, and in some cases estimated concentrations
caused by aircraft alone are in excess of the standards. High carbon monoxide
emission densities from aircraft at these airports, plus consideration of air
quality data collected at some sampling sites at the Los Angeles Airport,5 provide
additional evidence of important contributions by aircraft to high CO concentrations
in airport areas. It is expected that during the 1970's, without aircraft emissions
standards, total carbon monoxide emissions from aircraft at the four airports gen-
erally will remain about the same. Because of decreasing automobile emissions,
however, the relative contributions of aircraft to carbon monoxide emissions in
airport areas can be expected to increase.
Nitrogen Dioxide Dispersion modeling estimates indicate that N02 concentrations in
excess of the Federal annual N02 air quality standard exist in areas of long-term
public access such as residential areas. Contributions by aircraft to these concen-
trations are estimated to be significant but well within the standard. In areas of
short-term public access on airport grounds, however, concentrations caused by
emissions from aircraft are predicted to be larger. In this analysis, NOx was con-
sidered as N02.
It is expected that without NOX aircraft emission standards, NOX emissions from
aircraft during the 1970's are expected to increase by 100 percent at O'Hare Airport,
30 percent at Washington National Airport, 270 percent at Los Angeles International
Airport, and 150 percent at John F. Kennedy Airport. In general, these increases in
NOX emissions indicate substantial increases in the relative contributions by air-
craft to total NOx emissions in the airport areas studied, and point up the proba-
bility of increasing total N02 concentrations in them.
Smoke and Particulates •• Visible smoke emissions from aircraft have resulted in
widespread public complaints, and increased soiling effects have been noted in air-
port areas.5 The dispersion modeling analysis indicates that estimated particulate
concentrations caused by aircraft alone are in excess of the secondary (welfare-
related) air quality standard in some airport areas and can be substantial contribu-
tors to particulate concentrations in excess of the secondary standard in many other
airport areas.
Contributions by General Aviation Aircraft Because the emission levels from piston-
powered general aviation aircraft are high for hydrocarbons and particularly high
for carbon monoxide in relation to engine size, these aircraft can be important
contributors to total emissions from those air carrier airports where general
-------
aviation activity is great. For example, it is estimated that while accounting for
32 percent of total aircraft activity at Washington National Airport, small piston-
powered general aviation aircraft account for an estimated 3 percent of the total
hydrocarbons and 14 percent of the carbon monoxide emitted from all aircraft using
this airport.
Additional dispersion modeling at two general aviation airports indicated that
pollutant concentrations attributed to general aviation aircraft were well within
existing Federal ambient air quality standards. This modeling indicated, however,
that ambient air lead concentrations caused by general aviation aircraft might
present a potential problem. Large increases expected in general aviation activity
during the 1970's are expected to increase the significance of aircraft emissions
at these airports.
Air Quality Impact of Aircraft in Metropolitan Areas
The contributions by aircraft to total emissions in the metropolitan areas of
Los Angeles, New York City, Washington, D. C., and Chicago are of concern because
of the requirements that these areas meet and continue to meet the ambient air
quality standards. The oxidant and hydrocarbon, carbon monoxide, and oxides of
nitrogen air quality standards are not met in these cities. For hydrocarbon, carbon
monoxide, and oxides of nitrogen emissions, aircraft contributions to the totals in
these four areas are estimated to range from 0.5 to 4.6 percent in 1970, and from
1.5 to 7.6 percent in 1980 without aircraft emission controls. These values indicate
the present and increasing future significance of aircraft emissions in metropolitan
areas, and point up the increasing importance aircraft emission controls could be
expected to have in allowing early and continued compliance with ambient air quality
standards, particularly in airport vicinities. Additional estimates for the Los
Angeles area indicate that in 1980, if aircraft emission standards are not imple-
mented, emissions from light piston-engine aircraft can be expected to account for
more than half of the total aircraft emissions of carbon monoxide and about 20
percent of the total aircraft emissions of hydrocarbons.
EMISSION CONTROL OF AIRCRAFT TURBINE ENGINES
1. Carbon monoxide and hydrocarbon emissions can be significantly reduced (50
to 75 percent) at major air carrier airports by the following methods:*
a. Modifying Aircraft Ground Operational Procedures.
If these reductions are achieved by increasing engine operating power
levels and at the same time reducing the number of engines used in
*A reduction in odor levels at airports may accompany the hydrocarbon emission
reduction.
-------
taxi/idle operations, such controls may result in fuel savings yielding
a net benefit to the carrier.
b. Engine Modifications.
There are various methods available for achieving such reductions. The
cost of these methods would vary depending on the particular control
method selected and the number of engines and engine models affected.
The cost of implementing any of these methods in the total air carrier
fleet would be on the order of $100 million. The time required to
implement any of the engine modification control methods in the air
carrier fleet is estimated to be from 5 to 10 years.
2. Nitrogen oxide emissions can be reduced significantly (50 to 75 percent)
through the use of water injection during takeoff and climb-out modes. The
cost of applying this control method to the total air carrier fleet is
estimated to be approximately $100 million, and 5 years or more would be
required for its implementation. The degree of control can be reduced or
increased by reducing or increasing the rate or duration of water injection.
The costs will not, however, vary proportionately.
3. Visible smoke emissions from turbine engines can be substantially reduced
by minor combustor modifications, and such modifications are already being
implemented for certain engines. The additional costs of eliminating smoke
emissions from all air carrier aircraft is estimated to be on the order of
$100 million.
4. A 50 percent reduction in particulate emissions can be achieved by major
modifications of combustors, but at high cost ($600 million) and with long
implementation times (7 to 10 years).
5. Large (75 to 90 percent) reductions in carbon monoxide and hydrocarbon
emissions and significant (50 percent) reductions in nitrogen oxide and
particulate emissions will accompany the introduction of advanced combustor
design concepts in future engines. Complete elimination of visible smoke
will also be possible with the advanced designs. Associated costs will be
on the order of 3 percent of the total engine cost. Engines with these
features will not appear in service before the late 1970's.
EMISSION CONTROL OF AIRCRAFT PISTON ENGINES
1. Substantial reduction (50 to 75 percent) in carbon monoxide and hydrocarbon
emission rates can be achieved by applying exhaust emission control devices
to aircraft piston engines. Retrofitting costs would be high, on the order
of $100 million or more, with an estimated implementation time of greater
than 5 years.
-------
Minor engine design changes can be made to reduce carbon monoxide and
hydrocarbon emission rates (50 percent) without the addition of auxiliary
control devices. These changes include modifications in combustor chamber
geometry, valve and spark timing, and fuel-air ratios. Such changes would
probably not result in any significant increase in new engine cost and
could probably be incorporated in new engines in 3 to 4 years. Larger
reductions can be obtained by adding auxiliary control devices.
Lead emissions could be reduced directly by the use of low-lead or lead-
free gasoline.
10
-------
METHODOLOGY FOR IMPACT EVALUATION
On a nationwide mass basis, aircraft have been estimated to emit between 1 and
2 percent of the total hydrocarbons, carbon monoxide, and oxides of nitrogen, and
about 0.3 percent of the particulate matter and sulfur oxides. On a more localized
basis, however, aircraft emissions can be expected to have much greater significance
at airports and in their neighboring communities. As a first step in evaluating the
contribution by aircraft to air pollutant concentrations in airport areas, specific
airports were chosen for detailed study.
SELECTION OF AIRPORTS
Airports were selected to represent, as nearly as possible, those airports at
which the impact of emissions from aircraft and related activities would be greatest.
The factors considered in evaluating the potential impact of individual airports
included: (1) aircraft activity levels, (2) airport area, (3) mean wind speed, and
(4) relative activity of different types of aircraft (commercial air carrier, general
aviation, and military). On the basis of these considerations and the availability
of airport and aircraft activity data, the airports selected for study were as
follows:
1. Commercial Air Carrier
a. Los Angeles International
b. Washington National
c. J. F. Kennedy International
d. O'Hare International
2. General Aviation
a. Van Nuys, California
b. Tamiami, Florida
PROCEDURE OF AIR QUALITY ANALYSIS AT STUDY AIRPORTS
Basic to the air quality analysis was the development of emission factors and
aircraft activity data for use in dispersion modeling and in other estimates of
total aircraft and airport emissions.
11
-------
Emission Factors
Pollutants emitted by aircraft engines include gaseous hydrocarbons, carbon
monoxide, oxides of nitrogen, particulate matter, and sulfur oxides. Because
different types of aircraft emit different concentrations and compositions of pol-
lutants and have different patterns of usage, it was necessary to classify aircraft
by type and to define the typical operational modes they go through in their landing
and takeoff (LTO) cycles. This classification made possible categorization of
emission factors by aircraft type and mode of operation.
The aircraft classification system that was used divided aircraft into 12
separate types that include the currently used commercial air carrier, general
aviation, and military planes. Provision was also made in the classification sys-
tem for the supersonic commercial aircraft of the future. The complete classifica-
tion system is presented in Table 1.
The aircraft modes of operation that were used, in order of their occurrence,
are:
1. Start-up and idle.
2. Taxi.
3. Idle at runway.
4. Takeoff.
5. Climb-out to 3,000-foot elevation.
6. Fuel dumping.
7. Approach from 3,000-foot elevation.
8. Landing.
9. Idle and shutdown.
10. Maintenance.
Emission data were obtained for all operational modes for engines typical of those
in each aircraft class.
Emissions on the grounds and in the vicinity of the six airports - from sources
other than aircraft - were also taken into account in the air quality analyses. The
other sources of emissions included airport heating plants, fuel storage losses,
automobiles and service vehicles, and areas neighboring the airports.
Activity Level
To estimate the impact of aircraft emissions on air quality near the ground, it
is necessary to take into account aircraft activity from the time an aircraft enters
the atmospheric mixing layer during approach until it leaves this layer again during
climb-out. In defining an LTO cycle, a height of 3000 feet above the runway was
used as a reasonable approximation to the atmospheric mixing depth over major U.S.
12
-------
Table 1. AIRCRAFT CLASSIFICATION SYSTEM
Aircraft
Category
Air
carrier
General
aviation
Military
Class
1
2
3
4
5
6
7
8
9
10
11
12
Ref 12
classi-
fication
-
-
1
2
4
3
6
-
-
-
-
7
Type
Supersonic
transport
Jumbo jet
transport
Long-range
jet transport
Medium- range
jet transport
Turboprop
transport
Business jet
Piston-engine
utility
Over 400,000 Ib
gross weight
100,000 - 400,000
Ib gross weight
10,000 - 100,000
Ib gross weight
Under 10,000 Ib
gross weight
Hel i copters
and V/STOL
Examples
Concorde
Tupolev TU-144
Boeing 747
Douglas DC-10
Boeing 707
Douglas DC-8
Boeing 727
Douglas DC-9
Lockheed Electra
Fairchild Hiller
FH-227
Lockheed Jetstar
North American
Sabreliner
Cessna 210
Centurion
Piper 32-300
Cherokee Six
Boeing
Stratofortress
Lockheed
Starlifter
LTV Crusader
Cessna 172
Sikorsky S-61
Vertol 107
Representative engine
Engine model
R-K/Snecma
Olympus 593
P&WA JT9D
P&WA JT3D
P&WA JT8D
Al lison
501-D13
P&WA JT12
Continental
10-520-A
P&WA TF33-P-3
P&WA TF33-P-7
P&WA J57-P-20
Continental
10-360
General
Electric CT58
Type
Turbojet
Turbofan
Turbofan
Turbofan
Turbo-
prop
Turbojet
Opposed
piston
Turbofan
Turbofan
Turbojet
Opposed
piston
Turbo-
shaft
Thrust
or power3
39,000 Ib
43,000 Ib
18,000 Ib
13,900 Ib
3,750 hp
2,900 Ib
292 hp
17,100 Ib
20,900 Ib
18,000 Ib
211 hp
1 ,390 hp
Engines
per
aircraft
4
4
4
2.6
2.5
2.1
lb
-
-
2
(A)
Equivalent shaft power.
^Representative of Van Nuys and Tamiami.
-------
3
metropolitan areas. The number of LTO cycles performed, and the relative lengths
of time spent, in each operational mode of an LTO cycle, combined with the appro-
priate emission factors, determine the quantities of pollutants emitted by aircraft.
Records of aircraft activity data were obtained for the selected airports for
1970 and were classified by time of day, day of week, and month of year. Prospective
growth in activity at the airports was estimated by projecting past and current
activity data to 1975 and 1980. The yearly activity data and projections are
summarized in Table 2.
The air carrier airports are so-called because of the preponderance of com-
mercial air carrier activity, which, in 1970, ranged from 66 percent of total
activity at Washington National to 92 percent at Chicago O'Hare. The activity at
Tamiami and Van Nuys Airports is approximately 99 percent general aviation aircraft.
Additionally, data were obtained on the use and locations of taxiways, runways,
terminals, hangars, heating plants, fuel storage areas, and roadways at each airport
in order to locate and quantify the various sources of emissions during the operation
of aircraft.
Methods of Impact Evaluation
The primary and most direct method of impact evaluation involved the application
of frequently used dispersion modeling procedures to estimate air pollutant concen-
trations caused by aircraft alone and by all sources located in the airport vicinity
(within a 10-kilometer radius of the airport center). The dispersion modeling was
particularly useful in this study because it allowed estimation of pollutant con-
centrations caused by aircraft alone. Dispersion models similar to the one used in
this study have been specified by EPA as one means of showing that implementation
plans for certain regions will be adequate to meet the ambient air quality standards.
Much of the analysis of aircraft impact presented in this report is based on model-
ing work performed, under EPA contract, by Northern Research and Engineering Cor-
poration. A general description of the modeling procedure is presented here; a more
detailed account of the modeling work and results is available in the contract
report.
The general procedure in the modeling study involved: (1) approximating
emission sources as continuous, stationary point sources of constant strength over
the time period being considered, (2) modeling the dispersion of pollutants from
these sources using an empirical mathematical model, and (3) estimating concentra-
tions at specified receptor points by summing the pollutant contributions from each
point source.
14
-------
Table 2. LTD CYCLES FOR 1970, 1975, and 1980
Airport type
Air carrier airports
Los Angeles
International
Washington
National
J. F. Kennedy
International
Chicago
O'Hare
General aviation airports
Van Nuys,
California
Tamiami,b
Florida
Type of aircraft
Air
carrier
203,900
109,800
188,800
314,300
20
-
General
aviation
59,900
55,500
27,800
21 ,200
279,400
200,800
Military
4,200
1,500
-
-
2,700
-
Helicopters
4,050
-
-
-
-
-
Total LTO cycles
1970a
272,000
166,700
219,200
339,900
281 ,600
200,800
1975
305,200
169,800
208,500
357,000
-
-
1980
358,100
173,500
241,300
410,800
700,000
-
aWhere parts do not add up to total, LTO cycles not classified by type were included in total.
b!971 estimated activity.
-------
The point sources used in the modeling approximated the location and strength
of emission sources at each of the four airports studied. Lines along which auto-
mobile or aircraft movement occurred were represented by series of point sources.
Area sources, representing airport surroundings out to a 10-kilometer radius from
the airport center, were represented by circular arrangements of point sources
around the airports. Altogether, 149 to 276 point sources were used for each air
carrier airport depending on the size and complexity of the airport. The number of
sources was chosen to provide a reasonable approximation to airport and vicinity
emissions without excessive computer time and program complexity requirements.
The basis of the atmospheric dispersion modeling is an empirical, mathematical
approximation to pollutant dispersion after emission from a point source. This
approximation yields a plume whose concentration distribution is Gaussian in the
vertical and crosswind direction. The distribution is dependent upon downwind dis-
tance from the source and atmospheric stability. Eventually the upper boundary of
the atmospheric mixing layer restricts vertical plume spread and modifies the dis-
tribution of concentration in the vertical direction. This dispersion model should
be considered as a general approximation to airport dispersion patterns, as consid-
erable model development would be required to include more detailed small-scale
dispersion patterns, such as those around large buildings or near jet blasts.
In the calculation of long-term concentrations, the fact that -there is a dis-
tribution of meteorological conditions is used to simplify the basic dispersion
model. The result, known as the Martin-Tikvart model, approximates plume spread in
the crosswind direction and sums the contributions of all combinations of wind
speeds and atmospheric stabilities.
The concentration at any receptor point is obtained as the sum of the contribu-
tions from each point source of emissions. The accuracy of the concentration value
for this type of model is dependent upon the proximity of the receptor point and
the emission sources. Because the sources of emission are actually a collection of
points, lines, areas, and volumes, rather than merely a collection of points, as
assumed in the model, greater accuracy will generally result when the receptor point
is not in close proximity to any sources. In order to limit the portion of a pre-
dicted concentration attributable to the point source assumption, receptor locations
within 100 meters of a point source were not considered valid in this study.
The model estimated air pollutant concentrations both from aircraft alone and
from all sources at a number of sites located in and around the selected airports.
Receptors considered in this study were located according to the following overall
scheme: (1) one receptor at the center of each major terminal, (2) one receptor
100 meters from the head of each runway, (3) sixteen receptors on the airport
boundary, spaced equally on a compass rose located at the chosen center of the
16
-------
airport, (4) sixteen more receptors located in the airport surroundings, 5 kilo-
meters from the center of the airport and spaced equally on the compass rose, and
(5) mobile receptors 100 meters downwind of the two busiest runways for short-term
calculations. A maximum of 50 receptors was used, with the actual number depending
on the number of terminals and runways at each airport.
Long- and short-term ambient air pollutant concentrations were calculated for
1970 at all the receptor sites for each airport with the exception of Van Nuys, for
which no short-term concentrations were calculated because of a lack of local
meteorological data. The averaging times used for the air pollutant concentrations
corresponded to those for which the National Primary and Secondary Ambient Air
Quality Standards apply. The averaging times were: annual, 24-hour, 8-hour,
3-hour, and 1-hour concentrations.
The meteorological data used in calculating the annual average concentrations
were based on yearly averages for the specific airports. The short-term concentra-
tions (24-hour and shorter) were calculated using meteorological conditions that
occurred during 1970 at each airport, and that would be expected to yield high air
pollutant concentrations.
These meteorological conditions were selected on the basis of minimum wind
speed and associated maximum atmospheric stability. Examination of 1970 meteorolog-
ical data for the four air carrier airports indicated that these conditions occurred
with the frequency shown in Table 3 for the 1-hour carbon monoxide and 6 to 9 a.m.
hydrocarbons averaging times specified by the national ambient air quality standards.
This table shows that meteorological conditions conducive to high carbon monoxide
and hydrocarbon concentrations occurred at least twice at each airport and over 50
times at Los Angeles International Airport. The frequency of occurrence of these
worst meteorological conditions at these airports could be expected to be of the
same magnitude for 1971 and future years. This consideration is important because
Table 3. FREQUENCY OF OCCURRENCE OF "WORST METEOROLOGICAL CONDITIONS11
FOR HIGH CARBON MONOXIDE AND HYDROCARBONS CONCENTRATIONS
AT STUDY AIR CARRIER AIRPORTS IN 1970
Airport
Washington National
J. F. Kennedy International
Los Angeles International
O'Hare International3
1970 frequency of
occurrence for CO,
1-hour period
67
10
81
29
1970 frequency of
occurrence for hydrocarbons,
6- to 9-a.m. period
6
2
52
6
aData for O'Hare are incomplete and are based on 5 months: January, February, March,
October, and December.
17
-------
the short-term pollutant concentrations predicted by the model using these conditions
are compared with national ambient air quality standards that are not to be exceeded
more than once per year.
In addition to the detailed dispersion modeling, other methods of impact
evaluation that were used are listed below:
1. A comparison of hydrocarbon, carbon monoxide, and nitrogen oxides emission
densities was made between the four air carrier airports and their respec-
tive metropolitan areas.
2. An analysis was made of ambient air carbon monoxide concentrations
5
measured in and around Los Angeles Airport under a previous EPA contract.
3. An area source dispersion model was used to estimate hydrocarbon and possi-
ble resulting oxidant concentrations downwind from the Los Angeles Airport.
4. An estimate was made of the air quality effects of fuel dumping from
aircraft.
5. An estimate of contributions by aircraft to total emissions from the
metropolitan areas of Los Angeles, New York City, Chicago, and Washington,
D.C., was made for 1970 and 1980.
18
-------
RESULTS OF IMPACT EVALUATIONS
NATIONAL AMBIENT AIR QUALITY STANDARDS
To assess the significance of the aircraft contribution to pollutant concentra-
tions, the concentrations estimated by the dispersion model were compared with the
Q
national ambient air quality standards. These national standards have been set in
accordance with the Clean Air Act, as amended, and are applicable to areas acces-
sible to the general public external to buildings.
The standards consist of primary ambient air quality standards, designed to
protect against adverse health effects, and secondary standards, designed to protect
against adverse welfare effects such as plant and material damage or reduction in
visibility. The standards apply to hydrocarbons, carbon monoxide, nitrogen dioxide,
photochemical oxidants, sulfur dioxide, and particulate matter. The primary and
secondary standards are the same for hydrocarbons, carbon monoxide, nitrogen dioxide,
and photochemical oxidants, whereas the secondary standards are more stringent than
the primary standards for sulfur dioxide and particulates. The short-term standards
are not to be exceeded more than once per year. The standards are summarized in
Table 4.
It should be noted that nonmethane hydrocarbons at concentrations observed in
the atmosphere have not been associated with health effects. The relationship
between nonmethane hydrocarbons and photochemical oxidants indicates, however, that
peak photochemical oxidant concentrations are associated with hydrocarbon concentra-
Q
tions averaged over the 6 to 9 a.m. period. The nonmethane hydrocarbon standard
is based on this relationship. Thus, the average nonmethane hydrocarbon concentra-
tion of 160 vig/m for this period could result in a photochemical oxidant concen-
3
tration of 160 pg/m several hours later.
RESULTS OF DISPERSION MODEL AIR QUALITY ANALYSIS
Predicted Concentrations Compared with Primary
Air Quality Standards at Air Carrier Airports.
Predicted pollutant concentrations in the airport and its vicinity were compared
with primary ambient air quality standards only for those sites to which the general
public would reasonably have access for the exposure time specified for each air
quality standard. For example, a primary standard for a pollutant concentration
averaging time of 1 year was assumed not to apply to sites around terminal buildings
19
-------
Table 4. NATIONAL AMBIENT AIR QUALITY STANDARDS
Pollutant
Carbon monoxide
(Primary and secondary
standards are the same)
Nitrogen dioxide
(Primary and secondary
standards are the same)
Hydrocarbons (non-methane)
(Primary and secondary
standards are the same)
Parti cul ate 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 not 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 per year. For use as a guide in
devising implementation plans to meet the oxidant
standards.
75 micrograms per cubic meter, 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.
150 micrograms per cubic meter, maximum 24-hour
concentration not to be exceeded more than once per
year.
- 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 micrograms per cubic meter, annual arithmetic
mean.
260 micrograms per 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 be exceeded more than once per
year.
20
-------
because the general public would not be expected to remain in that area for a year's
time. In general, the duration of the public's exposure to pollutants (in agree-
ment with averaging times prescribed by the air quality standards) was estimated for
various airport area sites as follows:
1 hou» or 3 hours: roads or areas with no official parking facilities or where
parkint is prohibited, but where people might stay for a short time. (The 3-
hour du.ation is applicable to waterways in the vicinity of airports.)
1 hour, 3 hours, and 8 hours: air terminal areas and parking areas.
1 hour, 3 hours, 8 hours, and 24 hours: areas where a person could reasonably
stay overnight.
1 hour, 3 hours, 8 hours, 24 hours, and 1 year: residential areas in proximity
to airports.
It is important to note that particularly high concentrations of pollutants
from aircraft were predicted in some airport areas, such as ends of runways, where
public access is not allowed. Although the questions of occupational exposure are
beyond the scope of this report, the likelihood of exposure of airport workers to
high air pollutant concentrations should be recognized.
At each of the four commercial air carrier airports, ambient air pollutant con-
centrations, both from aircraft alone and from all sources (including aircraft) in
the airport vicinity, were predicted at a maximum of 50 sites, with the total number
being 193 sites for the four airports. These sites were located on or within a 5-
kilometer radius of the center of each airport. In general, the highest aircraft-
generated pollutant concentrations occurred on the airport grounds or at the airport
boundaries, whereas concentrations from aircraft at 5 kilometers from the airport
center were low.
Predicted pollutant concentrations, both from aircraft alone and from all
sources (including aircraft), are presented in Table 5 for sites that are accessible
to the general public for the averaging times specified in the primary air quality
standards and at which the total pollutant concentrations were predicted to exceed
the primary standards.
The type of area in which each site is located is also indicated in Table 5.
The type of area is indicated in the following manner:
T - Terminal area. Includes terminal buildings, observation decks, passen-
ger-unloading areas, and parking areas in the terminal vicinity.
P = Peripheral area. Applies to the vicinity of the airport boundary to
which the public has access, or along roadways that are used by the
21
-------
Table 5. PREDICTED AMBIENT AIR POLLUTANT CONCENTRATIONS FROM AIRCRAFT ALONE AND FROM AIRPORT VICINITY SOURCES AT SITES
WHERE POLLUTANT CONCENTRATIONS EXCEED PRIMARY AIR QUALITY STANDARDS AND WHERE PUBLIC COULD BE EXPOSED
FOR TIME OF STANDARD
6 to 9 a.m. maximum nonmethane hydrocarbon
concentration, yg/nr
Standard = 160 ug/m3
Site
location3
DCA (T)
DCA (T)
DCA
DCA
[P)
[P)
DCA (P)
DCA <
DCA
DCA
DCA
[P
P
P
P
DCA (P
DCA
DCA
DCA
DCA
DCA
rP
P
p
P
s
DCA (S)
DCA
DCA
DCA
S
S
s
DCA (S)
LAX
LAX
LAX
LAX
LAX
LAX
JFK
;T)
T)
T)
T)
T)
rT)
S)
Aircraft/
total
O/ 900
72/ 950
O/ 700
2/ 190
190/ 600
650/1350
80/ 700
170/ 700
140/ 700
190/ 450
0/1050
O/ 850
O/ 800
O/ 650
O/ 350
45/3000
0/1 1 50
O/ 450
O/ 240
O/ 800
250/ 480
V 280
160/ 500
O/ 370
O/ 500
O/ 550
120/ 300
Site
location
LAX
LAX
i?!
LAX (P)
LAX (S)
LAX
LAX
LAX
LAX
LAX
ORD
P)
P)
P)
s)
s}
T)
ORD (T)
ORD
ORD
ORD (P)
ORD
ORD
ORD
ORD
(P)
(P)
P)
P)
ORD (S)
ORD
ORD
S)
S)
ORD (S)
ORD (S)
JFK
JFK
;T)
JFK (P)
Aircraft/
total
O/ 280
280/ 470
O/ 170
O/ 770
45/ 160
45/ 220
SO/ 230
O/ 750
80/ 260
600/ 600
320/ 320
1700/3350
2400/2900
3/1700
0/3300
0/2300
O/ 420
O/ 400
150/ 800
860/ 940
700/1950
460/1730
0/4050
1440/2250
650/1620
450/ 750
Site
location
JFK
JFK
JFK
JFK
JFK
JFK
JFK
JFK
JFK
JFK
{I
T
T
T
T)
T)
T)
P)
P)
JFK (P)
JFK
JFK
JFK
JFK
P
P
P)
P)
JFK (P
JFK (P
JFK (P
JFK |
JFK
!P1
/
Pi
* i
JFK (P)
JFK
JFK (
S)
S)
JFK (S)
JFK
JFK <
JFK (
JFK (
S)
s)
s)
s)
Aircraft/
total
1000/2240
550/1750
800/1700
760/1400
860/1250
540/1090
3650/4200
70/ 330
490/ 800
480/ 950
420/ 740
370/ 510
220/ 380
190/ 420
O/ 260
O/ 380
55/ 490
150/ 750
90/ 450
70/1380
80/ 900
90/ 280
120/ 290
O/ 760
0/2750
O/ 340
11 650
70/ 300
1-hour maximum
CO concentration,
mg/m3
Standard = 40 mg/m
Site
location
JFK (T)
JFK (T)
JFK (T)
LAX (P)
LAX (P)
ORD (T)
ORD (S)
DCA (P)
DCA (P)
Aircraft/
total
85/100
4/ 45
3/ 44
55/ 62
32/ 45
2V 41
9/ 41
110/120
45/ 59
8-hour maximum
CO concentration,
mg/m3
Standard = 10 mg/m
Site
location
JFK (T)
JFK (T)
JFK (T)
JFK (T)
JFK (T)
ORD (T)
Aircraft/
total
4/28
3/26
4/17
11/18
9/19
3/12
Annual NOg
concentration,
ug/m3
Standard = 100 pg/m
Site
location
ORD
ORD
l\
ORD (S)
ORD (S)
ORD (S)
ORD
DCA
DCA
ORDb
ORDb
o
s
s
(p)
(P)
Aircraft/
total
23/160
5/140
5/130
5/130
6/130
5/130
2/120
1/100
17/130
19/140
aDCA = Washington National Airport, LAX = Los Angeles International Airport, JFK = John F. Kennedy International
Airport, and ORD = O'Hare Airport, Chicago. Also T = terminal area, P = peripheral area, and S = surrounding area.
Additional sites in residential areas expected to have high NOg concentrations.
-------
general public inside or near the boundary of the airport but that are
not near the terminal area.
S = Surrounding area. Applies to areas at a 5-kilometer radius from the
center of the airport.
There were a few sites that met the site-selection criteria for S02 and particu-
late matter, but the predicted contribution from aircraft at these sites was neg-
ligible. Accordingly, data for these sites are not included in Table 5.
It is important to note that the predicted hydrocarbon concentrations are
expressed in terms of nonmethane hydrocarbons. Conversion from predicted total
hydrocarbon concentrations was accomplished by assuming 50 percent nonmethane hydro-
carbons in the total hydrocarbon concentrations. This factor was based on ratios of
nonmethane to total hydrocarbons observed in actual air quality data collected in
Los Angeles.^ Thus, the predicted total hydrocarbon concentrations were divided by
two to make them comparable with the nonmethane hydrocarbon standard.
It should also be noted that the NOX emission rates used in this analysis,
particularly for newer types of jet engines, were found to yield total aircraft
emissions that were underestimated by a factor of approximately three when compared
with total aircraft emissions calculated from more accurate emission factors used to
establish baseline aircraft engine emission rates. The baseline emission factors
were not available in time for use in the dispersion model, but they have been in-
corporated into estimates of aircraft emissions presented later in this report. By
itself, the use of low NOX emission factors in the dispersion model would be expected
to yield underestimated ambient air concentrations. In this analysis, however, it
was assumed in predicting ambient N02 concentrations that all NOX could be considered
as N02- This is a conservative assumption, made because there exists no well-defined
relationship for the conversion of NO to N02- In the presence of high concentra-
tions of hydrocarbons, the NO to N02 conversion is accelerated, with best estimates
indicating that 90 percent of the NO is converted to N02 within a 2-hour period in
the presence of sunlight. The reaction is essentially negligible at night. Con-
sidering all NOX as N02 could by itself result in an appreciable overestimation of
annual average N02 concentrations. This overestimation of N02 concentrations and
the underestimation of NOx emissions resulting from the low NOx emission factors
would tend to be counteracting, however.
In Table 5, N02 concentrations, estimated by dispersion modeling, are shown for
two special sites that were selected because of their proximity to residential
areas where high N02 concentrations from aircraft were expected to occur. These
23
-------
sites are footnoted in Table 5, and were included in calculations for the summary
in Table 6.
Table 6. PREDICTED AIRCRAFT CONTRIBUTION TO AMBIENT AIR QUALITY AT COMMERCIAL AIR
CARRIER AIRPORTS, COMPARED WITH PRIMARY STANDARDS3
Average total concentration,
% of primary standard*3
Average contribution by air-
craft to total concentra-
tion, %
Average ambient air concen-
tration from aircraft
a Tone, % of primary
standard
Maximum ambient air concen-
tration from aircraft
alone, % of primary
standard"
Carbon monoxide
exposure
1-hour
(9 sites)
155 ,
(62 mg/ir T)
65
100 ,
(40 mg/m )
275 ,
(110 mg/nr)
8-hour
(6 sites)
200 ,
(20 mg/nH
25
60 3
(6 mg/nr)
110 -
(11 mg/nT)
Nonmethane
hydrocarbons
6 to 9 a.m.
exposure
(81 sites)
610,
(980 pg/rrT)
31
190 ,
(300 yg/nT)
2,300 ,
(3,650 pg/rrO
N02
annual
exposure
(10 sites)
131
(131 yg/m3)
7
9 T
(9 yg/nH
23
(23 yg/m3)
Based on sites listed in Table 5.
Actual average or maximum concentration given in parentheses.
Table 6, which summarizes the values listed in Table 5 in terms of the ambient
air quality standards, indicates the following:
1. At the selected sites where the 3-hour hydrocarbon air quality standard is
predicted to be exceeded, the average ambient air hydrocarbon concentration
from aircraft alone is predicted to be 190 percent of the standard, with a
maximum of 2300 percent of the standard.
The 81 public access points at which this standard is predicted to be
exceeded are located in areas both on and off airport grounds. The points
of highest aircraft contribution are located within airport areas that
include terminal buildings and their proximity, parking areas, and areas
along roadways near the ends of runways. Of the 81 sites, 20 were located
in airport terminal areas, 36 were located near the airport periphery, and
the remainder were located at a 5-kilometer radius from the centers of the
airports.
24
-------
2. At the selected sites where the carbon monoxide ambient air quality stand-
ards are exceeded, the average concentration from aircraft alone is pre-
dicted to be 100 percent of the 1-hour standard and 60 percent of the 8-
hour standard, with a maximum of 275 percent of the 1-hour standard.
The 1-hour carbon monoxide standard is predicted to be exceeded at two
sites on the periphery of Washington National Airport, at three terminal
sites at JFK Airport, at one terminal and one surrounding site at Chicago-
O'Hare Airport, and at two sites near the western periphery of the Los
Angeles Airport along Pershing Drive. The site of the highest predicted
1-hour CO concentration of 110 mg/m3 caused by aircraft alone is located
at the periphery of Washington National Airport. Carbon monoxide concen-
trations are predicted to exceed the 8-hour standard in five terminal
areas: four at JFK Airport and one at O'Hare Airport. The highest con-
centration produced by aircraft alone 11 mg/m3 is predicted to occur at
a terminal area of JFK Airport.
3. At the selected sites where the nitrogen dioxide concentrations are pre-
dicted to exceed the air quality standard, the predicted average ambient
air N02 concentration from aircraft alone is about 9 percent of the stan-
dard, with a predicted maximum of 23 percent of the standard.
Predicted nitrogen dioxide concentrations from 10 sites were compared with
the standard for N02- All the sites were near residential areas, with
three of the sites near the periphery of the airport. Among these three
sites, the site of highest predicted N02 concentration was located within
300 yards of a residential area approximately 2/3 mile east of the O'Hare
East-West runway.
Predicted Concentration Compared with Secondary Ambient Air Quality
Standards for Particulates and Sulfur Dioxide at Air Carrier Airports
The ambient air concentrations of particulates and S02 at sites where the
secondary standards are predicted to be exceeded are listed in Table 7. The public
has short-term access to many of the sites listed, and such sites are noted. Based
on these sites, the aircraft contributions to air quality are summarized in Table 8.
This table indicates that the estimated average ambient air concentrations from
aircraft alone for particulates are 168 percent of the annual and 33 percent of the
24-hour secondary particulate air quality standards at the sites considered. The
maximum particulate concentration from aircraft alone at these sites was estimated
to be 200 percent of the annual standard. The concentrations attributed to aircraft
are based on dry particulate emissions.
25
-------
Table 7. PREDICTED AMBIENT PARTICULATE AND S02 CONCENTRATIONS FROM AIRCRAFT AND
FROM AIRPORT VICINITY SOURCES AT SITES WHERE SECONDARY STANDARDS ARE EXCEEDED
(pg/tn3)
Particulate concentration
S02 concentration
Annual
Site
location3
DCAh
LAXb
LAX?
LAXb
ORD
ORD
ORD
Stands
60 uc
Aircraft/
total
67/ 84
86/ 93
120/130
97/110
65/ 86
72/ 91
53/ 91
rd =
/m3
Maximum 24-hour
Site
location
DCAb
DCAb
DCAb
DCA
DCA.
DCAb
LAX
LAX
ORD
ORD
ORD.
JFK?
JFKb
JFKb
JFKb
JFKb
JFKb
JFK5
JFKb
JFKb
JFKb
JFKb
JFKb
JFKb
JFKb
JFK?
JFKb
JFKb
JFKb
JFK?
JFKb
JFKb
JFKb
JFKb
Aircraft/
total
140/220
0/170
0/220
170/240
150/210
94/170
280/300
610/630
150/180
160/170
80/200
37/250
36/290
80/260
52/200
48/150
190/470
110/280
250/270
89/190
95/250
120/310
12/160
1/360
1/300
120/260
14/170
17/180
18/200
6/180
0/210
0/200
0/780
0/160
Standard =
150 yg/m3-
Annual
Site
location
DCAb
DCA
DCA
DCAb
DCAb
DCAb
DCA?
DCAb
DCAb
DCAb
LAXb
LAX°
LAXb
ORD
ORD
ORDb
JFKb
Aircraft/
total
5/ 66
25/ 69
46/ 79
3/ 87
2/ 83
3/220
5/ 65
I/ 70
O/ 61
O/ 65
55/ 65
81/ 91
64/ 74
62/ 80
54/ 75
36/ 62
5/ 78
Standard =
60 M
g/m3
Maximum 24-hour
Site
location
DCA.
DCAb
DCA
DCA.
DCA5
DCA
ORD
JFK
JFK
JFK
JFK
JFK
JFK
Aircraft/
total
3/ 570
7/ 320
21 430
2/1500
O/ 330
46/ 310
TOO/ 380
15/ 520
15/ 760
90/ 340
21 / 300
25/ 350
O/ 290
Standard =
260 ug/m3
Maximum 3-hour
Site
location
DCAb
DCA
ORDb
JFKD
JFK
Stand
1300
Aircraft/
total
25/1700
0/3700
2000/2000
14/4900
1800/2000
ard =
vg/m3
aDCA = Washington National Airport; LAX = Los Angeles International Airport; JFK = John F. Kennedy
International Airport; and ORD = O'Hare Airport, Chicago.
A site to which the public has access.
The S0£ ambient air concentrations attributable to aircraft alone are estimated
to be 37 percent of the annual, 9 percent of the 24-hour, and 2 percent of the 3-
hour secondary S02 air quality standards at the sites considered. The maximum SC>2
concentration from aircraft alone at these sites was estimated to be 135 percent of
the annual standard.
These percentages of the secondary air quality standards are very significant
because the secondary particulate and S02 standards are set at levels necessary to
prevent adverse welfare effects such as material damage, plant damage, and restric-
tion of visibility. Smoke generated by aircraft should be further considered in
26
-------
Table 8. PREDICTED AIRCRAFT CONTRIBUTION TO AMBIENT AIR PARTICIPATE AND S02
CONCENTRATIONS COMPARED WITH SECONDARY STANDARDS'
a
Average total concentration,
% of secondary standard*3
Average contribution by aircraft
to total concentration, %
Average ambient air concentra-
tion due to aircraft, % of
secondary standard13
Maximum ambient air concentra-
tion due to aircraft, % of
secondary standard^5
Particulate matter
exposure
Annual
(3 sites)
185
(111)
91
168
(101)
200
(120)
24-hour
(24 sites)
167
(250)
20
33
(50)
167
(250)
Sulfur dioxide
exposure
Annual
(10 sites)
145
(87)
25
37
(22)
135
(81)
24-hour
(5 sites)
122
(316)
8
9
(24)
35
(90)
3-hour
(2 sites)
253
(3300)
1
2
(20)
2
(25)
aBased only on sites accessible to public; footnoted in Table 7 with "b".
K *?
Actual average or maximum concentration given in parentheses in vg/m .
view of the fact that it causes significant reductions in visibility5 and results in
widespread public complaint.
Results for General Aviation
The two general aviation airports that were studied were Tampa-Miami (Tamiami)
Airport in Florida and Van Nuys Airport in California. At Van Nuys Airport, which
had higher activity than Tamiami, total emissions from aircraft were less than those
at the commercial aviation airports for hydrocarbons, carbon monoxide, sulfur
dioxide, nitrogen dioxide, and particulates, although CO emissions from Van Nuys
Airport approached those of Washington National Airport. This comparison is evident
in a table presented later in this report (Table 10). Lead emissions from aircraft
at general aviation airports, however, are roughly a factor of 10 greater than lead
emissions from those at the primarily commercial air carrier airports.
At Tamiami Airport, predicted ambient air pollutant concentrations at sites in
and around the airport indicated that the contribution from general aviation aircraft
to total concentrations of hydrocarbons, carbon monoxide, oxides of nitrogen, sulfur
dioxide, and non-lead particulates was well within the ambient air quality standards.
At Van Nuys Airport, however, annual average air-lead concentrations from aircraft
alone were predicted to be as high as 1.9 yg/m3 near the airport boundary. Increased
blood-lead levels have been associatedlO with ambient air-lead concentrations above
2 to 3 pg/m3. Future increases in total lead emissions may result in concentrations
further in excess of these values from general aviation aircraft alone.
27
-------
Piston-engine general aviation aircraft account for a significant fraction of
the total activity at the commercial air carrier airports considered in this study:
32 percent at Washington National Airport, 21 percent at Los Angeles Airport, 13
percent at 0. F. Kennedy Airport, and 6 percent at Chicago's O'Hare Airport. At
Washington National Airport, piston-powered general aviation aircraft accounted for
an estimated 3 percent of the total hydrocarbon, 14 percent of the carbon monoxide,
and 1 percent of the NOx emissions from all aircraft. These values, although maxi-
mums among the four commercial air carrier airports, indicate the very significant
contributions possible, especially to CO emissions, from general aviation aircraft
at some commercial airports. It should be noted that at 7 of the 23 largest hub
airports in the United States, general aviation activity comprises larger percentages
of total activity than it does at Washington National Airport.
Comparison of the Model's Predictions with Actual Air Quality Data
All atmospheric dispersion models have some inherent uncertainties because of
the complexity of simulating weather and dispersion conditions with a mathematical
model and because of inaccuracies in emission and activity data. The scarcity of
major airport air quality data that were directly comparable with the model's pre-
dictions limited any thorough comparisons. Because of the importance that the
model's predicted pollutant concentrations would play in estimating the impact of
aircraft emissions, however, it was necessary to investigate, within limitations,
the uncertainties in the model's predictions. Consequently, data from two studies
of air quality at airports were used to compare with the model's predictions. One
was conducted at Washington National Airport by EPA personnel during July 20 through
23, 1971; the other was a study done at Los Angeles Airport during May through
November 1970 by the Los Angeles Air Pollution Control District under an EPA con-
tract. 5
The Washington National Airport study was very limited in duration and scope
and, consequently, can only be considered to yield approximate results. Areas near
the main runway at National were sampled for carbon monoxide and hydrocarbons for
1 hour each day using comparatively crude sampling techniques. Particulate-matter
samples were collected with four high-volume samplers around the main runway for
three consecutive 4-hour periods from 7 a.m. until 7 p.m. and for one 12-hour period
from 7 p.m. to 7 a.m. One sampler at the end of the main runway operated 24 hours
from 7 a.m. to 7 a.m., and one located in front of the main terminal operated for
three consecutive 4-hour periods and one 12-hour period each day. Meteorological
data for the sampling period were obtained from the National Weather Service at the
airport, and activity data were obtained by actual observations and from the Federal
Aviation Agency for the same days. These data and the actual sampler locations were
28
-------
used as input to the model in order to reduce possible differences between the
model's predictions and the observed concentrations that could be caused by using
assumed conditions. The values predicted by the model for the same sampling times
were then compared with the observed data. For the carbon monoxide concentrations,
13 comparisons were possible. The model's predictions were lower than the observed
concentrations in eight cases and higher in the remaining five cases.
In regard to total hydrocarbons, 13 comparisons between observed and predicted
concentrations were possible also. Comparisons of total hydrocarbons showed a much
greater variation between the observed and predicted concentrations than existed for
carbon monoxide, but for all the comparisons the model's predicted concentrations
were much lower than the observed values.
A total of 85 comparisons was possible between the model's predicted concentra-
tions and the observed values for particulate matter. In 82 of the comparisons, the
model's predicted concentrations were lower than the measured concentrations. In
the remaining three comparisons, the predicted concentrations exceeded those actually
observed.
Although for all three sets of comparisons (carbon monoxide, total hydrocarbons,
and particulate matter), the scatter in the data and the variations in the compari-
sons were substantial, the majority of the values predicted by the model were lower
than the actual pollutant concentrations.
The study performed at Los Angeles Airport in 1970 obtained continuous carbon
monoxide hourly concentrations at fixed receptors near the terminal areas and at the
north-eastern and western boundaries of the airport. Additionally, a mobile sampling
unit collected hourly carbon monoxide data at points around the perimeter of the
airport on a random schedule. Data for the meteorological conditions occurring
during the sampling period were obtained and used as input to the model in order to
minimize the uncertainties that would have occurred if assumed conditions had been
used. Because of the large amount of concentration data collected during the 6-month
study, predicted concentrations were not calculated for all of the sampling times.
Calculations of concentrations by the model were made only for time periods during
the study selected on the basis of meteorological conditions conducive to high con-
centrations of air pollutants, or on the basis of comparatively high observed con-
centrations at the sampling sites. A total of 32 comparisons was possible between
predicted and the actually observed carbon monoxide concentrations. Of the 32 com-
parisons made, the predicted concentrations were lower than the observed in 26 cases,
and the predicted concentrations were higher than those observed in the remaining 6.
29
-------
In both the Los Angeles Airport and the Washington National Airport studies,
the model's predictions were lower than the corresponding observed concentrations
for the majority of the comparisons made. These comparisons suggest that for the
majority of hydrocarbon, carbon monoxide, and particulate predictions, ambient air
pollutant concentrations may be greater than those estimated by the model. Because
of the limited number of total comparisons possible, however, and the large scatter
in the available comparisons, it would be premature to form general conclusions
regarding the tendency of the atmospheric dispersion model to over- or underestimate
ambient air pollutant concentrations.
RESULTS OF OTHER ANALYSES OF AIR QUALITY IMPACT IN AIRPORT AREAS
Emission Density Comparison
Emission densities in airport areas in tons per square mile per year are, in
most cases, of greater magnitude than emission densities from their neighboring
metropolitan areas for hydrocarbon, carbon monoxide, and nitrogen oxides emissions.
Estimated emission densities for airports and for their neighboring metropolitan
areas are presented in Table 9. Of the four airports compared in Table 9, Los
Angeles Airport has the highest emission densities. The emission densities from
aircraft alone at this airport are 1.5 times the CO, 4.4 times the hydrocarbon,
and 1.1 times the NOX emission densities of the Los Angeles area. At the other
three airports listed in Table 9, the emission densities from all airport sources
are equal to or greater than those in the neighboring metropolitan areas, except for
nitrogen oxides in New York and carbon monoxide in Washington, D. C.
This emission density comparison suggests that, in these four airport areas,
the contribution by aircraft to hydrocarbon, CO, and NOx air pollutant concentrations
is substantial. Such contributions are particularly important in these instances
where major airports lie in or near metropolitan areas in which national ambient air
quality standards are currently exceeded.
Analysis of Measured Carbon Monoxide Air Quality Data
The likelihood of significant airport and aircraft contributions to local air
pollution levels is suggested by carbon monoxide air quality data gathered from May
to November 1970, during a Los Angeles Airport study5 under EPA contract. It is
important to realize that not all of the CO impact discussed here is attributable
to aircraft. It is possible that automobile activity in some of the airport and
"downwind" sampling areas discussed below accounts for very large proportions of
the measured concentrations. An estimate presented in Table 10, however, indicates
that aircraft were responsible for 55 percent of the total CO emitted at the Los
Angeles Airport in 1970.
30
-------
Table 9. COMPARISON OF EMISSION DENSITIES FOR AIRPORTS VERSUS URBAN AREAS, 1970
Los Angeles metropolitan area
Los Angeles Airport - all emission sources
Los Angeles Airport - aircraft alone
New York metropolitan area
JFK Airport - all emission sources
JFK Airport - aircraft alone
Washington, D.C., metropolitan area
National Airport - all emission sources
National Airport - aircraft alone
Chicago metropolitan area
O'Hare Airport - all emission sources
O'Hare Airport - aircraft alone
Area,3
mi 2
1250.0
3.9
3.9
320.0
4.5
4.5
61.0
1.0
1.0
227.0
6.7
6.7
Emission densities, b tons/mi 2-day
Carbon
monoxide
7.2
20.6
11.2
14.5
19.6
7.7
12.5
10.2
6.6
8.1
14.1
6.0
Hydro-
carbons
2.0
10.3
8.8
3.4
7.7
5.8
1.7
2.4
1.7
2.5
5.4
3.9
Nitrogen
oxides
1.0
2.0
1.1
3.6
2.1
0.8
1.7
1.7
1.0
1.4
1.9
0.8
Includes those areas used in the operation of the airport, but not necessarily the total area
owned by the airport.
Emissions used to calculate airport emission densities are based on all aircraft emissions within
each airport area.
-------
CO
ro
Table 10. CURRENT AND PROJECTED EMISSIONS3 FROM AIRCRAFT AND AIRPORTS
(tons/year)
Airport
and year
Los Angeles
1970
1975
1980
Washington
National
1970
1975
1980
John F. Kennedy
1970
1975
1980
O'Hare
1970
1975
1980
Van Nuys
1970
1975
1980
Participates
Aircraft
570
610
680
231
242
286
570
550
550
900
970
1,100
3.2
5.4
7.7
Airport
total
616
627
693
253
253
297
660
605
583
1,001
1,023
1,100
3.7
5.7
7.8
NOX
Aircraft
3,060
6,790
11,490
820
980
1,090
2,580
4,660
6,370
3,760
5,760
7,440
12.1
19.8
28.6
Airport
total
4,369
8,110
12,480
1,074
1,211
1,277
4,846
6,640
7,580
6,290
7,520
8,540
27.5
34.1
36.3
S02
Aircraft
431
490
623
105
121
143
418
415
442
562
600
718
0.033
0.066
0.099
Airport
total
434
561
726
319
330
352
902
913
957
605
660
803
0.33
0.55
0.88
Lead
Aircraft
0.3
0.9
1.0
0.5
0.5
0.5
0.3
0.4
0.9
0.2
0.4
0.6
3.2
5.3
7.6
Airport
total
35.2
22.0
7.8
4.8
2.1
0.9
53.9
27.5
7.8
63.8
28.6
7.7
3.6
5.5
7.6
Carbon monoxide
Ai rcraf t
16,030
16,630
18,480
2,410
2,700
3,030
12,590
11,280
10,680
14,740
13,840
13,530
1,650
2,750
3,960
Ai rport
total
29,230
28,730
27,280
3,731
3,691
3,470 .
32,390
26,680
18,380
34,540
31 ,440
22,330
1,870
2,860
4,070
Total
hydrocarbons
Aircraft
12,570
8,660
4,770
610
680
720
9,490
5,700
2,830
9,580
6,300
3,710
100
165
242
Airport
total
14,660
10,530
5,760
864
823
775
12,680
8,010
3,930
13,210
8,830
4,920
132
198
264
aBased on aircraft emissions below 3000 feet altitude.
Includes lead.
-------
One-hour CO concentrations at Los Angeles Airport, measured at sampling stations
just outside two satellite terminal areas, were 2.9 and 2.8 parts per million greater,
on the average, than CO concentrations measured near the western (most frequently
upwind) edge of the airport where the average concentration over a 6-month period
was 3 ppm. Additionally, the concentration measured near the eastern (most frequently
downwind) edge of the airport was an average of 1.1 ppm greater than the CO concen-
tration measured near the western edge of the airport.
Examination of data from a mobile sampling site gives further indication of the
potential increases in carbon monoxide concentrations in air passing over the airport.
Mobile sampling site 24, located near the east end of the southern runway complex of
Los Angeles Airport, was operated for a total of 30 hours between July and October
1970. For 19 of the 30 hours, the wind was blowing across the airport from the west
or west-southwest direction, and for 18 of these hours, it was possible to compare
CO concentrations at this site with concentrations measured near the western upwind
edge of the airport. The average of the 1-hour CO concentrations for the 18 hours
was 5.5 ppm at the site downwind of the airport, and 1.9 ppm at the site at the
western upwind edge of the airport, an increase of 3.6 ppm. Increases in concentra-
tions ranged from zero to a maximum of 7 ppm. These increases occurred under typical
summer and fall Los Angeles meteorological conditions. Under conditions particularly
conducive to high CO concentrations, expected to occur more often in the wintertime,
these increases could be substantially greater. Such increases in the carbon mon-
oxide concentration in air passing over the airport are significant in comparison
with the national ambient air quality standards for CO of 9 ppm for an 8-hour period
and 35 ppm for a 1-hour period.
Although the 1-hour CO ambient air quality standard was exceeded infrequently
at most outside samplers in the airport area during the study, in October and Sep-
tember alone the 8-hour national ambient air quality standard was exceeded 12 times
at site 204 near the western (most frequently upwind) edge of the airport, 31 times
at site 208 in the airplane loading area outside satellite terminal 7, and 22 times
at site 209 near the eastern (most frequently downwind) periphery of the airport.
Monthly carbon monoxide concentrations at downtown Los Angeles (4 to 6 ppm) and
Lennox (6 to 7 ppm) were within the range measured at the airport stations (2 to
18 ppm).
Area Source Dispersion Model
The previously discussed modeling analysis estimated that hydrocarbon concen-
trations from aircraft alone would be well in excess of the standard in airport areas
(Tables 5 and 6). The nonmethane hydrocarbon standard of 160 yg/m3, however, is to
be used as a "guide in devising implementation plans to meet the oxidant standards."
33
-------
In Air Quality Criteria for Hydrocarbons,9 the point is made that on "...days...
when meteorological conditions were most conducive to the formation of photochemical
oxidant, nonmethane hydrocarbon concentrations of 200 pg/m3 (0.3 ppm C) for the 3-
hour period from 6:00 a.m. to 9:00 a.m. might produce an average 1-hour photochemical
oxidant concentration of up to 200 yg/m3 (0.10 ppm) 2 to 4 hours later." Therefore,
one way to relate the hydrocarbon emissions to oxidant formation would be to examine
the hydrocarbons and their concentrations caused by aircraft an average of 3 hours
downwind of the airport area. To this end an additional modeling effort was under-
taken for Los Angeles Airport to examine the hydrocarbon concentrations 3 hours
downwind of the airport. It must be understood, however, that results should be
considered very approximate.
The modeling method used in this analysis involved approximating both airport
and surrounding emissions by area sources, and relating these emissions to downwind
pollutant concentrations by assuming Gaussian pollutant distribution in the vertical
and crosswind directions. For each receptor point, the concentration caused by
small-area elements was determined by integrating in the crosswind and upwind
directions over each source region. The airport and surroundings were considered
as separate source regions, and concentrations from each of these two sources were
calculated separately and added together to obtain the total concentration at each
receptor. Near the airport source, concentrations are the same as from an area
source of infinite extent.^ At greater distances, edge effects caused by the
finite width of the airport are considered by including the integration in the cross-
wind direction. Also included is the limit to vertical mixing imposed by a more
stable layer aloft.
For the purpose of this modeling, the airport was assumed to cover an area of
3.2 by 3.2 kilometers. The time period for the analysis, 8 a.m. to 11 a.m., was
chosen on the basis of recurring meteorological conditions conducive to high air
pollutant concentrations. A diurnal correction factor was applied to the resulting
concentrations to correct for the disproportionately greater amount of activity that
occurred during this 3-hour period than occurred during other 3-hour periods during
the day. The specific meteorological conditions used for the time period considered
were representative of severe conditions, from an air pollution standpoint, that are
expected to occur at least once a year in the Los Angeles area.
The results of the analysis are shown in Figure 1. The three curves show
total hydrocarbon concentrations downwind of Los Angeles Airport resulting from the
surroundings plus total airport emissions, total airport emissions alone, and air-
craft emissions alone. The initial concentration at the western airport boundary
(0 kilometer on the graph) is shown to be zero, which is a result of the proximity
34
-------
AIRPORT EAST
BOUNDARY (3.2 km)
TOTAL ALL
SOURCES
TOTAL
AIRPORT
AIRCRAFT ALONE
METEOROLOGICAL CONDITIONS USED:
WIND FROM WEST
STABILITY CLASS =3
WIND SPEED = 1.5 ra/sec
MIXING HEIGHT = 200 m
10 12 14 16 18 20 22 24
DOWNWIND DISTANCE, kilometers
Figure 1. Calculated total hydrocarbon concentrations at Los Angeles Airport
and downwind.
35
-------
of the airport western boundary to the ocean, wind direction from the west, and the
assumption of negligible hydrocarbons in air off the ocean. At a point 3 hours
downwind (16 kilometers from the eastern boundary), the total hydrocarbon concen-
tration caused by aircraft alone is expected to be approximately 330 yg/m3. Assuming
that 50 percent of the total hydrocarbons are methane,9 the nonmethane hydrocarbon
concentration 3 hours downwind becomes 165 yg/m3. which is 103 percent of the non-
methane hydrocarbon standard of 160 yg/m3. Thus, this estimate suggests that non-
methane hydrocarbon concentrations caused by aircraft alone may remain in excess of
the standard for the 3 hours necessary for possible formation of oxidant concentra-
tions in excess of the oxidant standard.
Emissions from Unburned-Fuel Dumping
Because of design of turbojet engines currently used by air-carrier airlines,
residual unburned fuel is collected in drain cans at shutdown and during start-up.
On takeoff, ram air pressure generated as the aircraft accelerates causes these
drain cans to vent automatically to the atmosphere. This source of hydrocarbon
emissions is seen to be significant when the large number of takeoffs of air carrier
aircraft at the commercial airports is considered. The total hydrocarbon emissions
resulting from fuel dumping at commercial airports in 1970 were estimated as follows:
(1) Los Angeles International, 440 tons; (2) Washington National, 132 tons; (3) J.F.
Kennedy International, 360 tons; and (4) Chicago-O'Hare International, 506 tons.
A dispersion model was used to predict total hydrocarbon concentrations, for a
1-hour period, associated with fuel dumping at Washington National Airport. Implicit
in this analysis was the assumption that the dumped fuel evaporates fully before
reaching the ground. Meteorological conditions assumed in the model were selected
to be conducive to high ground-level concentrations of the unburned hydrocarbons.
These conditions are of the type most likely to occur around midday in late spring
and summer. The results of the analysis indicated that the peak ground-level total
hydrocarbon concentration was 18 yg/m3, with a concentration of 10 yg/m3 occurring
over an area of approximately 0.2 square mile.
Widespread complaints about oily films on automobiles and other surfaces in
airport vicinities have been voiced. Preliminary investigations at Heathrow Airport
in London^ confirmed the deposition of oily droplets under low-flying incoming
flights at the end of a runway under colder, wet conditions; the studies also
showed that the localities of complaints about odors or oily deposits were, in the
majority of cases, on the flight paths of aircraft. Thus, it is likely that the
films are at least partially due to fuel dumping and condensable exhaust hydrocarbon
emissions from aircraft.
36
-------
FUTURE PROJECTION OF AIRCRAFT AND TOTAL EMISSIONS IN AIRPORT AREAS
Based on projections from Reference 1, revised to incorporate more accurate
emission factors, total emissions of pollutants from aircraft and from all sources
at different airports were projected from the baseline year of 1970 through 1980.
These projections are presented in Table 10 for total hydrocarbons, carbon monoxide,
nitrogen oxides, sulfur dioxide, particulate matter (including lead), and lead. The
aircraft projections are based on present emission rates for each aircraft engine
class and do not incorporate potential future reductions in emissions as a result of
aircraft emission standards. The projected aircraft emissions reflect increased
activity and changes in the mix of existing engines.
At the four air carrier airports during the 1970's, as a result of continued
introduction of jet engines found in present-day new jet aircraft, total emissions
of carbon monoxide from aircraft are not projected to change greatly. Hydrocarbon
emissions, however, although predicted to increase by 18 percent at Washington
National Airport, are expected to decrease by about 60 to 70 percent at Los Angeles,
John F. Kennedy, and O'Hare Airports. The estimated average increase in aircraft
operations is 20 percent at these airports during the 1970's, indicating, in general,
lower hydrocarbon and carbon monoxide emissions from the newer and, in many cases,
larger engines. The substantial increases in aircraft NOX emissions of 275 percent
at Los Angeles, 146 percent at John F. Kennedy, 98 percent at O'Hare, and 33 percent
at Washington National Airports, however, reflect the greater amounts of NOX emitted
during an entire LTO cycle from the newer engines. Some increases in S02 and par-
ticulate emissions from aircraft are projected to occur, and such increases usually
follow increases in aircraft operations.
At Van Nuys Airport, the increases in all pollutants paralleled the large pro-
jected increases in activity at this airport. During the 1970's, emissions of
hydrocarbons, carbon monoxide, N0x» and lead from aircraft are projected to increase
by about 140 percent.
It is estimated, as Table 10 indicates, that in 1975 CO emissions from aircraft
at Van Nuys Airport will exceed CO emissions from aircraft at Washington National
Airport. This estimation indicates the increasing importance of general aviation
aircraft emissions, and emphasizes that during an LTO cycle, CO emissions from a
small general aviation piston engine can, in many cases, be expected to approach CO
emissions from a commercial air carrier turbine engine.
The comparison of emission densities (airport versus metropolitan area) for
1975 and 1980 demonstrates that emission densities from all airport sources will,
for every city except New York, exceed those of the metropolitan areas in which they
37
-------
are located. In 1980, the emission densities from aircraft alone at these airports
are predicted, except for New York, to exceed emission densities from the metropoli-
tan areas.
Thus the ratio of the airport emission densities to those of the metropolitan
areas will, in most cases, be increasing. In some instances they will be increasing
dramatically. The trends can be identified in Table 11, which indicates that air-
craft can be expected to become increasingly significant contributors to air pollu-
tant concentrations in airports and their vicinities.
CURRENT AND PROJECTED CONTRIBUTION OF AIRCRAFT TO EMISSIONS
IN METROPOLITAN AREAS
The contribution by aircraft to total hydrocarbon, carbon monoxide, and oxides
of nitrogen emissions from the four metropolitan areas of Los Angeles, New York City,
Washington, D. C., and Chicago is of concern because the ambient air quality stand-
ards for these pollutants are exceeded in all four cities. While the ambient air
quality standards are required to be met by the mid-1970's, preliminary estimates-,
based primarily on data from drafts of implementation plans, indicate that a great
deal of difficulty will be experienced in meeting some of the standards in these
and other cities by that time, without highly disruptive changes such as major
traffic and land-use controls.
The contributions by aircraft to emissions from these areas may have signifi-
cant effects on the ability of these areas to meet and maintain air quality stan-
dards, particularly in areas neighboring airports. In the Los Angeles basin area,
an area of particularly high pollutant concentrations, the estimated contribution by
aircraft to total emissions in 1970 and 1980 is detailed in Table 12.
It is evident from Table 12 that aircraft emissions will become increasingly
important as other emission sources, particularly automobiles, are controlled. It
is also evident that by 1980 the hydrocarbon and, especially, the carbon monoxide
emissions from aircraft using airports other than Los Angeles International Airport
(primarily single-engine, piston-powered general aviation aircraft) will become
particularly significant when compared with emissions from the primarily commercial
jet aircraft using Los Angeles International Airport.
In Table 13, the contribution to metropolitan emissions by aircraft from the
major airports of each of the three other cities is estimated. The aircraft con-
tributions, as shown in Table 13, are of significance presently and, in general,
will become much more significant in the future.
38
-------
Table 11. COMPARISON OF EMISSION DENSITIES FOR AIRPORTS VERSUS URBAN AREAS FOR 1970, 1975, and 1980
Los Angeles metropolitan area
Los Angeles Airport - all
emission sources
Los Angeles Airport - air-
craft alone
New York metropolitan area
Airport - all emission -
sources
Airport - aircraft alone
Washington D.C. metropolitan
area
National Airport - all
emission sources
National Airport - air-
craft alone
Chicago metropolitan area
O'Hare Airport - all
emission sources
O'Hare Airport - air-
craft alone
Area,9
mi 2
1250.0
3.9
3.9
320.0
4.5
4.5
61.0
1.0
1.0
227.0
6.7
6.7
Emission densities, D tons/mi2-day
1970
Carbon
monoxide
7.2
20.6
11.2
14.5
19.6
7.7
12.5
10.2
6.6
8.1
14.1
6.0
Hydro-
carbons
2.0
10.3
8.8
3.4
7.7
5.8
1.7
2.4
1.7
2.5
5.4
3.9
Nitrogen
oxides?
1.0
2.0
1.1
3.6
2.1
0.8
1.7
1.7
1.0
1.4
1.9
0.8
1975
Carbon
monoxide
4.8
20.2
11.7
11.4
16.2
6.9
7.9
10.1
7.4
6.3
12.9
5.7
Hydro-
carbons
1.1
7.4
6.1
2.4
4.9
3.5
1.1
2.3
1.9
1.7
3.6
2.6
Nitrogen
oxides?
0.9
3.5
2.6
3.6
2.7
1.5
1.5
1.8
1.2
1.4
2.0
1.3
1980
Carbon
monoxide
2.8
19.1
13.0
5.5
11.2
6.5
3.3
9.5
8.3
1.4
9.1
5.5
Hydro-
carbons
0.9
4.0
3.4
1.3
2.4
1.7
0.4
2.1
2.0
0.9
2.0
1.5
Nitrogen
oxides
0.8
5.6
4.9
3.2
3.0
2.3
1.3
1.9
1.4
1.2
2.3
1.8
Airport areas represent those areas devoted to the operation of the airport, but not necessarily the total area
owned by the airport.
Emissions used to calculate airport emission densities are based on all aircraft emissions within each airport area.
CO
10
-------
Table 12. CONTRIBUTION TO TOTAL EMISSIONS
BY AIRCRAFT IN LOS ANGELES BASIN AREA*
Pollutant
CO
HC
NOX
Estimated contribution, %
1970
0.8 (0.5)b
1.4 (1.4)
0.7 (0.7)
1980
3.6 (1.4)
1.5 (1.2)
3.3 (3.2)
Includes the densely populated 1250-square-mile
area of Los Angeles County in the Los Angeles
Basin.
"Values in parentheses indicate contributions
caused by aircraft from Los Angeles Airport
alone.
Table 13. CONTRIBUTION TO TOTAL METROPOLITAN EMISSIONS
BY AIRCRAFT USING MAJOR AIRPORTS IN THREE CITIES
Metropolitan area
and airport
City of Chicago:
O'Hare and Midway
Airports
New York City:
J.F.K. and
LaGuardia Airports
Washington, D. C.:
Washington National
Airport
Pollutant
CO
HC
NOX
CO
HC
NOX
CO
HC
NOX
Estimated contribution, %
1970
2.3
4.6
3.3
0.9
2.5
0.8
0.5
1
1.3
1980
5.9
5.1
7.6
2.3
2.4
2.1
2.4
4.8
2.3
Tables 12 and 13 are presented to indicate that control of aircraft emissions
could make a significant contribution to allowing both implementation and mainte-
nance of total metropolitan emission reductions consistent with the national ambient
air quality standards.
40
-------
TECHNOLOGICAL FEASIBILITY
OF CONTROLLING AIRCRAFT EMISSIONS
Information on emission control methods is necessary to determine the levels to
which aircraft emissions can feasibly be reduced. The results of an earlier Federal
study,3,12 indicated that practical control approaches include modification of air-
craft engines, fuel, and ground operational procedures. More recently, the Aero-
space Industries Association (AIA) has distributed a report!3 that summarizes the
results of extensive investigations conducted by industry on: (1) emission charac-
teristics of aircraft gas turbine engines; (2) causes of such emissions; and (3)
methods for their reduction. The AIA report also identifies the possibility of
reducing emissions through modifications in engines (especially combustor design)
and in ground operational procedures.
Thus, the" current assessment of control methods must consider each of these
approaches. In assessing the feasibility of a control method, four factors must be
explored: (1) the effect of the method on the functioning or capacity of the air-
craft system; (2) the effectiveness of the method in reducing emissions; (3) the
cost of utilizing the method; and (4) the time required for implementing the method.
Information on emission-measurement instrumentation is necessary to ensure that air-
craft emissions can be measured with the accuracy and sensitivity necessary for
enforcing the desired standards.
The specific objectives of this investigation of aircraft emission control
technology were:
1. To identify methods of controlling aircraft emissions through modification
of engines, fuels, and ground operations.
2. To estimate the effectiveness of these control methods in reducing aircraft
emission rates.
3. To estimate the time and cost of implementing these control methods.
4. To assess the technology of measuring emissions from aircraft engines, and
to identify areas where advancements in instrumentation or test procedures
are required.
The investigation of fuel modification was discontinued after preliminary
analysis indicated that no significant reductions in emissions could be achieved by
41
-------
modifying fuels, except for reductions in sulfur or lead that result in proportion-
ate reductions of S02 and lead emissions.
A list of specific emission control methods involving engine modifications was
formulated on the basis of preliminary analyses which indicated that each method
was feasible and offered a significant reduction in one or more emission classes.
Feasibility was based on the following factors:
1. No reduction in engine reliability (safety).
2. Little or no reduction in engine performance (power-weight ratio).
3. Reasonable cost of implementation.
The preliminary list of control methods was then subjected to more detailed
analysis of control effectiveness and implementation costs. Control methods involv-
ing changes in ground operations were evaluated in a similar manner.
In the evaluation of the emission control methods involving engine modifications,
the following emission classes were given primary consideration:
1. Carbon monoxide (CO).
2. Nitrogen oxides (NOX).
3. Total hydrocarbons (including drained fuel) (THC).
4. Dry particulates (DP).
5. Smoke.
EMISSION CONTROL BY ENGINE MODIFICATION
Engine Classification
To facilitate analyses of engine modifications, aircraft engines have been cate-
gorized according to their thrust or power level. The classification system that
has been adopted is indicated in Table 14.
Table 14. AIRCRAFT ENGINE CLASSIFICATION
Engine class
Tl
T2
T3
PI
Engine type
Turbine
Turbine
Turbine
Piston
Power range,
Ib thrust or eshp
Less than 6,000
6,000 to 29,000
Greater than 29,000
All piston engines
42
-------
This classification system, although it is based simply upon power level,
effectively groups together engines of similar emission potential (when the emission
rates are normalized according to an appropriate engine-size parameter). Also, the
effectiveness and cost of the control methods are similar for all engines within
each class. Thus, the classification system has been particularly useful for this
program and may also provide a rational basis for the formulation of emission con-
trol standards.
Three classes of turbine engines are defined, and all piston engines are in-
cluded in a single class. The system is effective in that it categorizes engines
according to their principal applications and according to certain design charac-
teristics that affect emission rates.
The small turbine engine class (Tl) includes most of the turboshaft and small
turbojet and turbofan engines used in business and small commercial aircraft. These
engines should be considered separately because the relatively small size of the
combustor components (or large surface-volume ratio) makes control of certain emis-
sions more difficult than with larger engines.
The next turbine engine class (T2) includes most of the turbojet and turbofan
engines used in medium-to-large commercial aircraft. The design characteristics of
most of these engines are basically similar.
The third turbine engine class (T3) includes large turbofan engines for "jumbo"
transport aircraft and the SST engines currently in use or under development.
Emission Control Methods and Effectiveness
An analysis has been conducted of the technology for controlling emissions from
aircraft engines by means of engine modifications. The purpose of this analysis was
to identify specific methods of reducing pollutant emissions from aircraft engines
and to indicate the reductions in rates of emission attainable by these methods.
Various engine modifications have been identified that appear to be feasible in the
sense that they can be applied to aircraft without degrading engine reliability or
seriously reducing aircraft performance. The costs of implementing these control
methods also appears to be within reasonable limits, at least from preliminary
analysis.
Turbine Engines - The engine modification control methods considered feasible for
turbine engines are listed and described briefly in Table 15. Six methods are, at
least in principle, applicable to existing engines by retrofitting of new or modi-
fied parts, and to engines currently in production. Two methods are considered to
be applicable only to future engines of new design, inasmuch as the modifications
43
-------
Table 15. ENGINE MODIFICATIONS FOR EMISSION CONTROL FOR EXISTING AND FUTURE
TURBINE ENGINES
Control method
Modification
Existing engines
tl - Minor combustion
chamber redesign
t2 Major combustion
chamber redesign
t3 Fuel drainage control
t4 - Divided fuel supply
system
t5 Water injection
t6 - Modify compressor air
bleed rate
Minor modification of combustion chamber and fuel
nozzle to achieve best state-of-art emission
performance.
Major modification of combustion chamber and fuel
nozzle incorporating advanced fuel injection concepts
(carburetion or prevaporization).
Modify fuel supply system or fuel drainage system to
eliminate release of drained fuel to environment.
Provide independent fuel supplies to subsets of fuel
nozzles to allow shutdown of one or more subsets dur-
ing low-power operation.
Install water injection system for short duration use
during maximum power (takeoff and climb-out) opera-
tion.
Increase air bleed rate from compressor at low-power
operation to increase combustor fuel-air ratio.
Future engines
t7 Variable-geometry
combustion chamber
t8 Staged injection
combustor
Use of variable airflow distribution to provide inde-
pendent control of combustion zone fuel-air ratio.
Use of advanced combustor design concept involving a
series of combustion zones with independently con-
trolled fuel injection in each zone.
required are too extensive to be applied to engines for which development has been
completed.
The first control method consists of simple modifications of the combustor and
fuel nozzle to reduce all emission rates to the best levels currently attainable
within each engine class. The degree of control attainable depends upon the per-
formance of specific engines compared with those engines in the same class that have
the best emission performances. Each of the other control methods is more specifi-
cally directed at one or two pollutant classes.
The actual reduction in emission rate achievable through the use of a control
method varies with the pollutant considered, the engine class, and the engine
operating mode. Estimates of the effectiveness of each control method have been
made for all combinations of these factors and are presented in Tables 16 and 17.
The estimation of emission control effectiveness for turbine engines is based upon
reductions attainable from "best current emission rates." These rates are defined
as those attainable through control method tl, minor combustion chamber redesign.
44
-------
Table 16. EFFECTIVENESS OF tl MINOR COMBUSTION CHAMBER
REDESIGN3 ON REDUCTION OF EMISSIONS FROM TURBINE ENGINES
(Emission rates in lb/1000 Ib of fuel)
Engine
class
Tl
Tl
Tl
Tl
T2
T2
T2
T2
T3
T3
T3
T3
Pollutant
CO
THC
NOX
DP
CO
THC
NOX
DP
CO
THC
NOX
DP
Mode
Idle/taxi
25
10
3
0.2
45
10
2
0.2
50
10
3
0.1
Approach
5
1
7
0.5
6
1
6
0.5
3
1
10
0.1
Takeoff
2
0.2
11
0.5
1
0.1
12
0.5
0.5
0.1
40
0.1
Minor combustor redesign is assumed
invisible or "smokeless" levels for
to reduce the smoke to
all engine classes.
It is predicted that all engines in each class could be modified to achieve
these "best rates." The values of these rates are listed in Table 16. These "best
rates" are not the lowest rates indicated for each engine class, but are rates near
the low end of those emission rates that appear to be realistically attainable. The
use of the "best rate" basis is necessary to allow effectiveness estimates to be
made for each engine class. Because of the wide variations in actual emission rates
for turbine engines, the use of average rates as a basis for an effectiveness anal-
ysis would be less significant. Estimates are based upon demonstrated performance
in a few cases. In most instances, however, no direct experience has been obtained
with these control methods on aircraft'engines. Therefore, to a large extent,
estimates of effectiveness are based on theoretical analyses of engine performance
under the operating conditions associated with the control methods. The bases for
these estimates are summarized in Table 18.
Emission-control effectiveness is indicated in Tables 16, 17, and 18 for each
control method and for each pollutant for which a significant degree of control
would be expected. Pollutants for which little or no control would be expected are
not listed. Effectiveness is indicated separately for each engine class. No esti-
mates have been made for control of reactive hydrocarbons, odor, or aldehydes
45
-------
Table 17. EFFECTIVENESS OF ENGINE MODIFICATION IN CONTROL
OF EMISSIONS FROM TURBINE ENGINES, BY OPERATING MODE3
Control
method
t2b
t2
t3
t3
t3
t4
t4
t4
t4
t4
t4
t5
t5
t5
t6
t6
t6
t6
t6
t6
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
t7 or t8
Engine
class
Tl
T2
Tl
T2
T3
Tl
Tl
T2
T2
T3
T3
Tl
T2
T3
Tl
Tl
T2
T2
T3
T3
Tl
Tl
Tl
Tl
T2
T2
T2
T2
T3
T3
T3
T3
Pol lutant
DP
DP
THC
THC
THC
CO
THC
CO
THC
CO
THC
NOX
NOX
NOX
CO
THC
CO
THC
CO
THC
CO
THC
NOX
DP
CO
THC
NOX
DP
CO
THC
NOX
DP
Mode
Idle/ taxi
0.5
0.5
NCC
NC
NC
0.25
0.25
0.25
0.25
0.25
0.25
NC
NC
NC
0.5
0.5
0.5
0.5
0.5
0.5
0.1
0.1
NC
0.5
0.1
0.1
NC
0.5
0.1
0.1
NC
0.5
Approach
0.5
0.5
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
NC
0.5
NC
NC
NC
0.5
NC
NC
NC
0.5
Takeoff
0.5
0.5
°dd
°d
Od
NC
NC
NC
NC
NC
NC
0.25
0.25
0.25
NC
NC
NC
NC
NC
NC
NC
NC
0.5
0.5
NC
NC
0.5
0.5
NC
NC
0.5
0.5
Emission rate is fraction of best current rate assumed to be
attainable through minor combustion chamber redesign and with
control method cited.
t2 Major combustion chamber redesign.
t3 Fuel drainage control.
t4 = Divided fuel supply system.
t5 = Water injection.
t6 = Modify compressor air bleed rate.
t7 = Variable-geometry combustion chamber.
t8 = Staged injection combustor.
CNC indicates no change.
Refers to raw fuel drainage only.
46
-------
Table 18. BASES FOR CONTROL METHOD EFFECTIVENESS ESTIMATES FOR TURBINE ENGINES
Control method
Rationale
tl Minor combustion
chamber redesign
t2 Major combustion
chamber redesign
t3 - Fuel drainage control
t4 - Divided fuel supply
system
t5 Water injection
t6 Modify compressor air
bleed rate
t7 Variable-geometry
combustion chamber
t8 Staged injection
combustor
The assumption is made that emission rates for all
engines within a given class can be reduced to common,
optimum levels (on a lb/1000 Ib fuel basis) by minor
combustor modifications. These optimum emission rates
are based on the best performance reported for each
engine class, excluding extreme data points.
Estimates are based on reports of carbureting fuel
injector performance and reduction of smoke emission.
Concept is incorporated in some Class T3 engines.
Estimates are based on assumption that best emission
rate for Class Tl and T2 engines is at an exhaust
visibility threshold at maximum power. Carburetion
appears to reduce smoke level, and presumably particu-
late emissions, to approximately half that level.
Estimate is based on the assumption that fuel drainage
can be completely eliminated by collecting drained fuel
and returning to fuel tank.
Control method results in combustion zone fuel-air ratio
similar to that at approach condition. Reduction in CO
and THC from idle to approach is approximately 90 per-
cent in Class Tl and T2 engines and 90 percent in
Class T3 engines. Effectiveness is reduced by one order
because combustor is not operating at "well-designed"
condition.
Water injection is assumed only at takeoff at a rate
equal to twice the fuel rate. Water injection into
compressor or diffuser is assumed to be by system
similar to those in current use. Effectiveness based
upon published results with steam injection.14 Water
injection assumed to be of equal effectiveness when
injected upstream of combustor.
Assumptions are (1) fraction of air that can be bled is
small so that engine operating point is nearly unchanged,
(2) combustor f/a varies inversely with air bleed rate,
and (3) CO and THC emissions at idle vary as the (air
mass flow rate)3 and inversely as (f/ap. This )i
relationship is based upon data from Reference 12.
If maximum air bleed rate is 20 percent, CO and THC
emission rates are reduced by 50 percent.
Combustor primary zone is assumed to operate at a con-
stant f/a equal to normal f/a at approach power condition
(primary equivalence ratio = 0.6). CO and THC emissions
at idle are reduced to levels corresponding to approach
power, or by 90 percent for Classes Tl and T2 and 90 per-
cent for Class T3. Combustor incorporates design char-
acteristics that provide good mixture in combustion
zone. This feature and constant f/a operation combine
to reduce NO emissions at full power by 50 percent^ and
particulate emissions by 50 percent at all power levels
as in t2.
47
-------
because control methods applicable to these emissions have not been identified.
Some reductions in these emissions may occur along with reductions in THC emissions.
Any of the modifications defined for existing turbine engines (tl through t6) could
be combined to achieve increased emission control effectiveness with the exception
of modifications t4 and t6. These two are mutually exclusive.
Piston Engines The control methods considered feasible for aircraft piston engines
are listed with brief descriptions in Table 19. These methods include most of the
approaches used to control carbon monoxide (CO) and total hydrocarbon (THC) emissions
that have been developed for automotive engines. Methods for controlling nitrogen
oxide (NOX) emissions are not included inasmuch as the fuel-rich operating conditions
of aircraft piston engines result in low NOX emission rates. NOX emission control
may be required, however, in conjunction with any attempt to reduce CO and THC emis-
sions by changing engine operating conditions.
Table 19. ENGINE MODIFICATIONS FOR EMISSION CONTROL FOR EXISTING AND FUTURE
PISTON ENGINES
Control method
Modification
Existing engines
pi Simple air injection
p2 Thermal reactors
p3 Catalytic reactors
for HC and CO
control
p4 - Direct-flame
afterburner
p5 Water injection
p6 Positive crankcase
ventilation
p7 Evaporative emission
controls
Future engines
p8 Engine redesign
Air injected at controlled rate into each engine
exhaust port.
Air injection thermal reactor installed in place of,
or downstream of, exhaust manifold.
Air injection catalytic reactor installed in exhaust
system. Operation with lead-free or low-lead fuel
required.
Thermal reactor with injection of air and additional
fuel installed in exhaust system.
Water injected into intake manifold with simultaneous
reduction in fuel rate to provide for cooler engine
operation at leaner fuel-air ratios.
Current PCV system used with automotive engines applied
to aircraft engines. Effective only in combination
with one of preceding control methods.
A group of control methods used singly or in combina-
tion to reduce evaporative losses from the fuel system.
Control methods commonly include charcoal absorbers and
vapor traps in combination with relatively complex
valving and fuel flow systems.
Coordinated redesign of combustion chamber geometry,
compression ratio, fuel distribution system, spark and
valve timing, fuel-air ratio, and cylinder wall temper-
ature to minimize emissions while maintaining opera-
tional reliability.
48
-------
Eight piston-engine control methods are listed in Table 19 including the use of
direct-flame afterburners and water injection, methods that are not being considered
currently for automotive engines. Afterburners are included here since they might
be used to advantage because they can utilize the high velocity airflow around the
aircraft. Then too, the adaptation of other methods may be less feasible for air-
craft than for automobiles. The piston-engine emission-control methods were identi-
fied and evaluated through reviews of published evaluations of these methods. Of
the methods identified, all are considered applicable to existing engines except
those that would require redesign of the basic engine or its control systems.
Effectiveness estimates for piston engines are based on reductions from current
uncontrolled rates listed in Table 20. Emission rates from piston engines do not
vary widely, so that control effectiveness can be based on average rates for existing
engines. The effectiveness estimates shown in Table 21 (which are based on published
results of effectiveness of automotive emission controls) are based in most cases
on the application of individual control methods without other engine changes.
Method p6 (PCV) is an exception; it is considered to be useful only in combination
with method pi, p2, p3, p4, p5, or p8.
Table 20. CURRENT UNCONTROLLED EMISSION RATES
FOR PISTON ENGINES16
(lb/1000 Ib of fuel)
Pollutant
CO
THCa
NOX
(as N02)
Mode
Idle
896
48
7
Taxi
882
76
4
Approach
918
80
4
Takeoff
849
18
6
aTotal hydrocarbon (THC) emission rates have been
increased by 50 percent to account for crankcase blow-
by emissions. Evaporative emissions are not included
in these rates.
Piston-engine modifications pi through p5 are designed to serve the same function
and, thus, are mutually exclusive. All of the others could be combined with any of
the modifications pi through p5 to achieve increased emission-control effectiveness.
Cost and Time Requirements for Control-Method Development
and Implementation
Existing Engines - The cost and time requirements of applying each control method
applicable to existing engines have been estimated. These estimates are of a pre-
liminary nature and are intended to indicate the magnitude of the costs involved in
49
-------
Table 21. EFFECTIVENESS OF ENGINE MODIFICATIONS
IN CONTROL OF EMISSIONS FROM PISTON ENGINES,
BY POLLUTANT3
Control method
pi -
P2 -
P3 -
p4
p5
p6
p7 -
P8 -
Simple air injection
Thermal reactor
Catalytic reactor
Direct-flame
afterburner
Water injection
Positive crankcase
ventilation (PCV)
Evaporative emission
control
Engine redesign
Controlled
emission rate
CO
0.5
0.25
0.25
0.1
0.25
NC
NC
0.5
THCb
0.5
0.25
0.25
0.1
0.25
d
e
0.5
Lead
NCC
NC
0.1
NC
NC
NC
NC
NC
Emission rate is fraction of uncontrolled emission
rate after installation of control method and
applies to all operating modes.
DExhaust HC only.
"NC indicates no change.
PCV would eliminate blow-by emissions when used in
combination with pi, p2, p3, p4, p5, or p8. Blow-
by THC emission estimated to be equal to 30 per-
cent of uncontrolled exhaust emission.
Evaporative controls would reduce THC emissions
due to evaporation from fuel supply. Magnitude of
uncontrolled emissions is unknown.
controlling emissions from all civil and military aircraft. Cost and time require-
ments have been estimated separately for control-method development and implementa-
tion. Development includes all effort required from initial stages through certifi-
cation of the control method for a specific engine class and tooling for production.
Implementation includes initial installation of the control method on all engines of
a given class, and any costs associated with additional effort or materials required
for the control method throughout the remaining service life of the engines.
Because few of the control methods have actually been developed for or applied
to aircraft engines, and because many factors affect total implementation costs, many
uncertainties are involved in the estimates. The estimates of development cost
and time requirements are considered to be reliable. They are judged to be accurate
within a factor of about 2. The estimates of implementation costs are considered
50
-------
to be less reliable. The cost and service life of a modified engine component can-
not be predicted accurately. Yet these factors strongly affect the cumulative costs
of operating and maintaining the modified engine. This uncertainty is unfortunate
because implementation costs could be far greater than development costs for some
control methods. Thus, the estimates of implementation costs can only be regarded
as indicative of cost penalties that might be involved with control-method imple-
mentation.
Estimates are given in Table 22 of the development time, development costs, and
implementation costs for application of each control method to the current popula-
tion of all civil engines.
Table 22. TIME AND COSTS FOR MODIFICATION OF CURRENT
CIVIL AVIATION9 ENGINES
Control method
Turbine engines
Minor combustion
chamber redesign
Major combustion
chamber redesign
Fuel drainage control
Divided fuel supply
Water injection
Compressor air bleed
Piston engines
Simple air injection
Thermal reactor
Catalytic reactor
Direct-flame
afterburner
Water injection
Positive crankcase
ventilation
Evaporative emission
control
Development
time,
years
2.5 to 5
5 to 7.5
3 to 5.5
5 to 7.5
2.5 to 4
4 to 6.5
1 . 5 to 3
3 to 6
2.5 to 5
3 to 6
1.5 to 3
1 to 2
1.5 to 2.5
Development
cost,
I06 dollars
37
74
35
84
25
90
9
25
22
25
9
4
4
Implementation
cost,
10b dollars
343
589
44
102
151
58
165
424
535
424
81
94
269
a"Civil aviation" includes air carrier and general aviation engines.
The development time requirements listed in Table 22 are the periods required to
reach the point where installation of the control methods in existing engines could
begin. Installation of any control method in all existing engines would require an
51
-------
additional time period that is dependent primarily on the availability of engine
maintenance facilities.
The time for implementation is estimated to be 2-1/2 years for turbine engines
and 5 years for piston engines. Table 23 presents costs by category: air carrier,
general aviation, and military.
Table 23. COST RESULTS FOR TURBINE ENGINE POPULATION
BY SEPARATE USE CATEGORIES
Engine
class
Tl
Tl
Tl
Tl
Tl
Tl
T2
T2
T2
T2
T2
T2
T3
T3
T3
T3
T3
T3
Control
method
tl
t2
t3
t4
t5
t6
tl
t2
t3
t4
t5
t6
tl
t2
t3
t4
t5
t6
Cost
scaling
factor
0.35
0.35
0.35
0.35
0.35
0.35
1.00
1.00
1.00
1.00
1.00
1.00
1.64
1.64
1.64
1.64
1.64
1.64
Development
cost per
engine family,
106 dollars
0.90
1.80
0.99
1.80
0.62
2.20
0.90
1.80
0.99
1.80
0.62
2.20
0.90
1.80
0.99
1.80
0.62
2.20
Implementa-
tion cost
per engine,
103 dollars
12.4
21.3
1.6
3.7
5.5
2.1
35.5
69.9
4.5
10.5
15.6
6.0
58.3
100.0
7.4
17.2
25.6
9.9
Total cost, 106 dollars
Air
carrier
19.2
34.5
7.7
14.9
9.8
15.5
243.0
418.0
39.5
87.0
108.7
61.5
10.0
19.0
7.4
13.7
5.9
16.0
General
aviation
90.5
159.3
25.6
51.5
43.6
48.1
17.8
31.0
4.0
8.3
8.2
7.1
0.0
0.0
0.0
0.0
0.0
0.0
Civil
aviation3
109.7
193.8
33.5
66.4
53.4
63.6
259.8
449.6
43.5
95.3
116.9
68.6
10.0
19.0
7.4
13.7
5.9
16.0
Military
aviation
533.0
921.0
87.0
190.0
240.0
131.0
1,032.2
1,774.4
137.9
317.4
454.9
190.6
16.8
29.4
3.9
8.0
7.8
7.0
"Civil aviation" includes air carrier and general aviation engines.
To put the implementation costs in a different perspective, they may be expressed
as fractions of total engine costs. For a typical class T2 (turbine) engine, the
cost of installing and maintaining control methods ranges from $4,500 to $61,000,
assuming a 10-year engine life. Based on a total engine cost of $250,000, these
control-method implementation costs represent 2 to 25 percent of the total engine
cost. For a typical piston engine, estimated control-method implementation costs
range from $600 to $4,000, also based upon a 10-year engine life. For a total
52
-------
engine cost of $6,000, these implementation costs represent 10 to 65 percent of the
total engine cost.
The turbine engine cost and time estimates were developed by using the applica-
tion of low-smoke combustors to the JT8D engine class as a reference. Cost and time
requirements for this modification, which is considered to be a minor combustor re-
design for a class T2 engine, were estimated in detail in 1969.17 Requirements for
other control methods were determined essentially by proportioning the cost and time
expenditures according to the complexity of the method, with respect to the reference
case. Requirements for other engine classes were determined by using appropriate
scaling factors and by again using the JT8D modifications as reference.
The piston-engine time and cost estimates are based largely on experience to
date with emission controls for automobile engines.
Future Engines - Cost estimates also have been developed for incorporation of emis-
sion controls in future engines, that is, engines that have not yet been developed.
These estimates have been defined only as fractions of total engine cost inasmuch
as no reasonable basis is available for estimating the numbers of engines that would
be affected.
Emission control in turbine engines that is attained through the use of advanced
combustor-design concepts is estimated to represent an increase in total engine cost
of 3 to 4 percent. Emission control in piston engines that is achieved by engine-
design modifications would not necessarily result in any significant increase in
engine cost. If greater control of emissions is required than can be achieved by
engine design modifications, however, one or more of the control methods applicable
to existing engines will be necessary. The cost of these control methods, which
involve the addition of auxiliary devices such as thermal reactors, will be signifi-
cant, probably in the range of 5 to 10 percent of total engine cost.
These estimates represent the increased costs of new engines with emission con-
trols installed. It is possible that there may be additional continuing costs for
maintenance of the control methods. These maintenance costs would be similar to those
for modification of existing engines, which were estimated to represent 2 to 25 per-
cent of total turbine-engine costs and 10 to 65 percent of total piston-engine costs.
EMISSION CONTROL BY MODIFICATION OF GROUND OPERATIONS
Definition of Ground Operations
The cycle of operations performed by an aircraft during its arrival at and depar-
ture from an airport can be defined quite precisely because most of these operations
53
-------
are prescribed by airport or aircraft operating procedures. Characteristic operating
or LTD (landing-takeoff) cycles have been defined for various classes of aircraft
for purposes of estimating pollutant emissions.
The LTD cycle can be separated logically into flight and ground operations.
Flight operations include the approach and climb-out modes as well as the landing and
takeoff runs, even though the latter occur on the ground. Ground operations include
the taxi and idle modes of the cycle. This separation is logical for two reasons.
First, flight operations as defined here are those that cannot readily be modified
in order to attempt to reduce pollutant emissions. Second, flight operations are
conducted almost entirely with aircraft engines at full or part power, and, under
these conditions, pollutant emission rates are quite different from those at the low
power levels that are characteristic of ground operations. Aircraft ground opera-
tions contribute substantially to the concentrations of CO and THC that exist at air-
carrier airports because of the relatively high emission rates of these pollutants
at low engine power levels, and because ground operations are largely confined to
limited areas within the airport boundaries.
Emission Control Methods
Seven methods have been identified that offer some degree of control of CO and
THC emissions at air-carrier airports by modification of ground-operation procedures.
1. Increase engine speed during idle and taxi operations.
2. Increase engine speed and reduce number of engines operating during idle
and taxi.
3. Reduce idle operating time by controlling departure times from gates.
4. Reduce taxi operating time by transporting passengers to aircraft.
5. Reduce taxi operating time by towing aircraft between runway and gate.
6. Reduce operating time of aircraft auxiliary power supply by providing
ground-based power supply.
7. Manually remove fuel from fuel drainage reservoirs.
The first two methods reduce emissions by requiring that engines be operated at
more efficient power settings, and the next four methods reduce emissions by reduc-
ing operating time of either main or auxiliary engines. The effectiveness of these
methods in reducing emissions varies considerably. Table 24 summarizes the reduc-.
tions in CO and THC emissions that would result at Los Angeles International Airport
from the seven suggested ground-operation changes.
54
-------
Table 24. COMPARATIVE REDUCTIONS RESULTING FROM CONTROL
METHODS APPLIED AT LOS ANGELES INTERNATIONAL AIRPORT
Control method
Resultant emissions,
% of uncontrolled
emissions
CO
Hydrocarbons
1. Increase engine idle speed
2. Increase idle speed and use minimal
engines for taxi
Two engines
Single engine
3. Eliminate delays at gate and runway
4. Transport passengers between
terminal and aircraft
5. Tow aircraft to avoid taxi emissions
6. Avoid use of aircraft auxiliary
power units (APU)
7. Control emptying of fuel drainage
reservoirs
71
53
39
90
100
34
99.5
100
93
66
51
91
100
42
98.5
98.4
The control methods listed above are not, with the possible exception of number
3, applicable to small, piston-engine aircraft, and, therefore, do not seem to offer
means for controlling emissions at general aviation airports. Delay times at take-
off are significant at some general aviation airports; however, aircraft ground
traffic at general aviation airports may not be sufficiently controlled for a system
of controlled gate departures or engine start-up to be effective in reducing delays.
On the other hand, one operational technique that might be very effective in reducing
CO and THC emissions from light aircraft during idle and taxi modes is the required
use of leaner carburetor mixture settings during these modes.
Implementation Cost and Time Requirements
The cost and time requirements of the control methods involving ground-opera-
tion modifications have been estimated for one specific airport, Los Angeles Inter-
national . These estimates involve many uncertainties and, therefore, must be regard-
ed as preliminary. The estimates are considered to be accurate within a factor of 2;
that is, the true costs and time for implementation at the airport considered prob-
ably are within a range of 50 to 200 percent of the estimates. A summary of the
estimates is presented in Table 25. Implementation of these methods at other air-
ports would involve costs of the same magnitude. The actual costs, however, would
vary with activity level and the present availability of auxiliary equipment.
55
-------
Table 25. COSTS AND TIME FOR OPERATIONS CHANGES AT LOS ANGELES
INTERNATIONAL AIRPORT
Control method
Increase engine speed
Increase speed, reduce
number
Control gate departure
Transport passengers
Tow aircraft
Reduce APU operation
Manual drainage
Time,
years
0
0.3
5
2.5
1
0.5
0.5
Initial cost,
106 dollars
U
0
15
65
1.2
1.3
0.04
Annual operating
cost change,3
106 dol lars
8.5
-2.5
-1.5
5.0
0.4
1.5
3.0
Minus sign indicates an estimated savings.
EMISSION MEASUREMENT TECHNOLOGY
Reliable methods for measuring the rates at which pollutants are emitted from
aircraft engines are required for the support of an emission-control program. Emis-
sion measurements are required for evaluating the effectiveness of control methods,
and specific measurement methods must be incorporated in emission-control standards.
An assessment has been conducted of the state of emission-measurement technology
to determine whether measurement techniques are sufficiently well advanced to support
the development of emission-control methods and the implementation of emission
standards for aircraft engines. The conclusion drawn from this assessment is that
current measurement technology will meet the requirements of an emission-control
program. Measurement techniques for particulate emissions are inadequate at present
but development of improved techniques is being initiated through cooperative govern-
ment-industry action.
Measurement of emission rates from an aircraft engine involves three major
requirements:
1. A test procedure specifying engine operating conditions.
2. A sampling technique for obtaining a representative sample of exhaust gas.
3. Analytical instrumentation for determining pollutant concentrations in the
exhaust-gas sample.
Aircraft engine manufacturers and certain government agencies have devoted sub-
stantial effort toward providing for these requirements for measuring emissions from
turbine engines.
56
-------
Sampling and Test Procedures
Obtaining a representative sample of exhaust gas from an aircraft engine for
analysis of emission rates is not a trivial procedure. Sampling emissions from
turbine engines is difficult at the outset because of the jet-blast environment in
which the sampling equipment must be installed. Beyond that problem, the following
factors are all found to have significant effects on the composition of the exhaust
sample:
1. Engine power level.
2. Temporal and spatial variations in exhaust composition.
3. Sampling-line diameter, length, material, and temperature.
4. Ambient temperature and humidity.
5. Ambient pollutant levels.
Procedures for sampling and analyzing turbine-engine exhaust gases have been
under development for several years by engine manufacturers and various government
agencies. More recently, the Society of Automotive Engineers E-31 Committee has
been formed to standardize these procedures. Standardization of measurement tech-
niques will minimize the variations resulting from the factors listed above; how-
ever, the sources of error in collecting exhaust samples and the variability of
samples among different engines must be considered in the establishment of any
emission-control standards.
Sampling requirements for aircraft piston engines can be expected to be similar
to those for automobile engines. There are no apparent factors that would cause
variability in exhaust samples beyond those factors already recognized as affecting
automobile exhaust samples.
Emission Measurement Instrumentation
Measuring the concentrations of most gaseous pollutants in exhaust samples from
aircraft engines is generally within the capabilities of existing instruments, and
should remain so even when engines are modified to reduce emission rates.
The various types of instruments that are available and in current use for air-
craft emission measurement have been reviewed. Instruments that appear to be most
suitable for measuring turbine-engine emissions at the present time are presented
in Table 26.
57
-------
Table 26. INSTRUMENTATION FOR MEASUREMENT
OF TURBINE ENGINE EMISSIONS
Measurement method
Pollutant class
Non-dispersive infrared (NDIR)
Heated flame ionization
Chemi1umi nescence
Chemiluminescence3
SAE smokemeter (ARP1179)
None
Determined from fuel analysis
3-MBTH
Human odor panel
CO and 0)2
THC
NO
N02
Smoke
Parti culates
Aldehydes
Odor
The non-dispersive ultraviolet instrument (NDUV)
may also prove acceptable for N0£ measurement.
58
-------
REFERENCES
1. Northern Research and Engineering Corporation. The Potential Impact of Aircraft
Emissions Upon Air Quality. Final Report to the U.S. Environmental Protection
Agency. Research Triangle Park, North Carolina. Contract Number 68-02-0085.
December 1971.
2. Northern Research and Engineering Corporation. Assessment of Aircraft Emission
Control Technology. Final Report to the U.S. Environmental Protection Agency.
Research Triangle Park, N.C. Contract Number 68-04-0011. September 1971.
3. Nature and Control of Aircraft Engine Exhaust Emissions. Report of the Secre-
tary of Health, Education, and Welfare to the United States Congress.
December 1968.
4. Aviation Effects on Air Quality. Bay Area Pollution Control District, San
Francisco, California. February 1971.
5. Jet Aircraft Emissions and Air Quality in the Vicinity of the Los Angeles Inter-
national Airport. Air Pollution Control District, County of Los Angeles,
California. April 1971.
6. Aircraft and the Environment. Proceedings of SAE/DOT Conference. February
8-10, 1971. Washington, D.C.
7. Nationwide Emission Trends: 1940-1970. U.S. Environmental Protection Agency.
Research Triangle Park, N.C. (Scheduled to be published in June 1972.)
8. National Primary and Secondary Ambient Air Quality Standards. Environmental
Protection Agency. Federal Register 36_(84):8187, April 30, 1971.
9. Air Quality Criteria for Hydrocarbons. U.S. DHEW, PHS, EHS, National Air Pol-
lution Control Administration. Publication No. AP-64. Washington, D.C.
March 1970. p. 8-5.
10. Airborne Lead in Perspective. Committee on Biological Effects of Atmospheric
Pollutants. Division of Medical Sciences, National Research Council, National
Academy of Sciences. Washington, D.C. 1971.
59
-------
11. Hanna, Steven R. A Simple Method of Calculating Dispersion for Urban Areas.
J.A.P.C.A., 21(12):774-777. December 1971.
12. Northern Research and Engineering Corporation. Nature and Control of Aircraft
Engine Exhaust Emissions. Final Report to the National Air Pollution Control
Administration. Durham, N.C. Contract Number CPA 222-68-27. November 1968.
13. Aerospace Industries Association. A Study of Aircraft Gas Turbine Engine
Exhaust-Emissions. Washington, D.C. August 1971.
14. Dibelius, N.R., M.B. Hilt, and R.H. Johnson. Reduction of Nitrogen Oxides from
Gas Turbines by Steam Injection. ASME Paper 71-67-58, American Society of
Mechanical Engineers.
15. Fletcher, Ronald S., Richard D. Siege!, and E. Karl Bastress. Northern Research
and Engineering Corporation. Report Number 1162-1, Design Criteria for Control
of Nitrogen Oxide Emissions from Aircraft Turbine Engines.
16. Scott Research Laboratories, Inc. A Study of Exhaust Emissions from Recipro-
cating Aircraft Power Plants. Scott Project Number 1136. Final Report to the
U.S. Environmental Protection Agency. Research Triangle Park, N.C. Contract
Number CPA 22-69-129. December 1970.
17. Northern Research and Engineering Corporation. Time Requirements for Retro-
fitting Jet Aircraft with Improved Combustor Design. Final Report to National
Air Pollution Control Administration, Durham, N.C. Contract Number CPA 22-69-90.
July 1969.
60
------- |