The objective of this study is to improve understanding of common emissions
(HC, CO and NOx) resulting from aircraft, based on the ICAO databank
(CAA, 2011). This databank includes exhaust emissions
and fuel flow rates for currently used turbofan engines during landing and takeoff
(LTO) phases. The investigation is carried out in such a way that the actual
flight data are considered. In order to perform this task a novel method is
developed, in which interpolation and extrapolation of the relationship between
the fuel flow and the emission indices data of specific type of engine is provided.
Data are obtained from flights of ten randomly selected B737-800 (hereafter
B738) commercial aircraft. Two types of turbofan engines are used in these aircraft:
CFM56-7B26 and CFM56-7B26/3.
The most frequently used domestic routes, aircraft types and the engine types
are considered for the data selection. For this purpose, ten randomly selected
B738 commercial aircraft are used. Five of them are for flights between the
Antalya International (AYT) and Sabiha Gokcen International (SAW) airports and
five are for flights between the Izmir Adnan Menderes International (ADB) and
the SAW airports in Turkey. The SAW airport is the arrival airport for each
The route selections relate to the frequency of flights to SAW. In 2009, for
domestic flights, the two most frequent arrivals to SAW are from ADB and AYT
with 2956 and 2684 total arrivals, respectively. For B738 aircraft only, the
numbers of arrivals reduce to 1540 for AYT and 1134 for ADB. Of the total arrivals
to SAW with B738 aircraft, therefore, flights from AYT and ADB account for 15
and 11%, respectively.
With respect to engines, six of the assessed flights were powered by CFM56-7B26
(hereafter 7B26) engines, while the remaining four were powered by CFM56-7B26/3
(hereafter 7B26/3) turbofan engines. Moreover, for the flights of each route,
three 7B26 and two 7B26/3 engines are utilized.
In the ICAO emission databank there are 17 different models of the CFM56-7B
series engine which can be defined according to by-pass ratio, overall pressure
ratio and thrust parameters. The engine family can be classified into three
groups: CFM56-7BX (single annular combustion), CFM56-7BX/2 (double annular combustion)
and CFM56-7BX/3 (improved emissions single annular combustion), where X denotes
model numbers such as 18, 20, 22, 24, 26 and 27 (U.S. Department
of Transportation, 2008). The classification is mainly based on the combustion
The ICAO emission database, has a disadvantage in that its emission indices
are obtained only for limited fuel flow rates. For instance, a single fuel flow
rate is accepted for the entire flight phase. This prevents a precise identification
of the emissions generated from commercial aircraft since the fuel flow rate
is not constant. To address this problem here, the approach used in this study
is developed based on linear and polynomial extrapolation and interpolation
for three types of emissions: HC, CO and NOx.
The variation of common emission indices with fuel flow rate indicates that
linear relationships are present for certain fuel flow fragments, where fuel
flow fragments are the fuel flow rate regions for a given flight phase for all
of the engines. For instance, the given fuel flow rates for the climb phases
of all the engine models constitute a fuel flow region while there are other
specific fuel flow ranges for the other three flight phases: Takeoff, approach
and landing. In other words, there is specific fuel flow region for each flight
phase for the engines of each group. For these regions the relationship between
the emission indices and the fuel flow rates can be straightforwardly identified.
However, for fuel flow rates outside the specified regions, particularly between
the two regions, the approach based on the aforementioned relationship can lead
to incorrect results. Therefore one needs additional methods which can be obtained
utilizing the sequential regions. For instance, if the overall fuel flow ranges
are divided into three parts (highest, medium and lowest regions indicating
the amount of fuel flow) then the region between highest and medium fuel flow
rates and also between the medium and the lowest fuel flow rates can be described
by additional relationships. The required model descriptions are presented in
the next section.
RESULTS AND DISCUSSION
The relationships are developed for two sections, representing both engine
series and given in Table 1 and 2. The fuel
flow rate ff range is also given in Table 1 and 2.
The coefficients of determination (R2) for the regression models
are given in the last columns of Table 1 and 2.
All of the regression models exhibit high coefficients of determination, mostly
over 0.960 (with the exceptions of 0.904 and 0.910 which relate to the fuel
flow range of CO emissions). This means that these models explain more than
96.0% of the variation in emission index when compared to the total variation.
The duration of the flight phases are depicted in Fig. 1.
In the ICAO database, emission indices are given for only four flight phases.
Utilizing the models developed in this study, one can obtain information related
to other flight phases, such as cruise and taxi.
|| Models emission indices for CFM56-7BX (flights 1, 3, 4, 6-8)
|*ff denotes fuel flow rate. **Coefficients in the models are
rounded to two decimals in most cases, *** EI of HC for this range is given
in ICAO database as a constant value of 0.1 gr HC kg-1 of fuel
|| Models emission indices for CFM56-7BX/3 (flights 2, 5, 9,
|*Coefficients in the models are rounded to two decimals in
||Durations of flight phases. The average total flight times
are 67.9 and 56.8 min for AYT and ADB, respectively
The breakdowns of the HC, C O and NOx emissions for the AYT-SAW
route are shown in Fig. 2. It is seen that certain emission
types produced at certain flight phases can vary greatly from those for the
other phases. On the other hand, the same flight phase can exhibit a relatively
lower amount of some kinds of emission types and a relatively higher amount
of other types of emissions.
||Emission breakdown by flight phase for (a-b) HC, (c-d) CO
AYT and (e-f) NOX and (g-h) Fuel ADB departures. The black line
in each graphic indicates the duration of the related phase. Phases P1 to
P14 are described in the legend of Fig. 1
Here it is noted that the engine power setting and phase duration significantly
affect emissions quantities.
Due to incomplete combustion, the emissions of HC and CO for lower engine power
settings are observed to be much greater than those for higher power settings
(Sutkus et al., 2001; Schurmann
et al., 2007; Mazaheri et al., 2009;
Anderson et al., 2006). Therefore, such emissions
obtained for flight phases such as taxi and descent are found to be higher than
the phases such as takeoff and climb. This pattern is observed the graphs in
Fig. 2. Accordingly, HC and CO emissions resulting from aircraft
operation during taxi (P2 for taxi in and P14 for taxi out) and descent (P12),
where the engine power setting is relatively low, are observed to be at higher
levels. For instance, the HC emissions are calculated as 3.7 and 3.8 kg for
the sum of P2 and P14 and P9, respectively. Since the phase duration can have
a great effect on the emissions, the duration should be considered along with
the amount of the emissions.
As can be seen from the graphs in Fig. 2, for relatively
lower power settings as for the idle and taxi (P2 and P14) phases, higher quantities
of HC and CO emissions are observed, depending on the phase duration. For instance,
the highest levels of HC emissions are observed for the phases of descent, taxi-in
and taxi-out, for which the corresponding durations are 18, 12 and 4 min.
The flight phases P3 (take off) and P12 (landing) are assumed to occur in the
vicinity of the airport since the flight phases occur at a low height over the
runway. For instance, the average heights for AYT-SAW routes are determined
to be 19 m for P3 and 12 m for P12. As a result, the durations of the idle and
taxi phases have significant effects on the quantities of HC emissions. For
the above example, the longer ground operation duration (P1-P3) at departure
airport AYT yields a higher HC emission (2.9 kg) while the shorter ground operation
(P12-P14) in the arrival airport SAW yields a lower HC emission (0.8 kg). Thus
for the high air traffic in busy airports, where aircraft often wait in taxi
sequence or holding point for takeoff in long queues, or for airports with long
taxi ways, there may be significant HC emissions in the vicinity of the airport.
The descent flight phase (P9) is the longest phase. In this phase, depending
on the descent procedure, the power settings of the engine can be idle or at
a value slightly higher than the idle. For instance, the N1 RPM of the engine
for the 9th flight is calculated at around 30-35% of the actual RPM at full
speed, after 89% of the descent time has elapsed. During the same phase, the
EGT values are concentrated at 426-429°C. By comparison, the distributions
of the N1 RPM and the EGT values for the taxi phase are concentrated at around
18-24% and 470-530°C, respectively. Although the lower power settings may
lead to higher HC emissions, the net benefit can be positive due to reduced
fuel consumption and amounts of other emissions, such as NOx and
The emissions breakdown of CO by flight phase is similar to those for HC emissions.
However, the mass of CO emissions can be an order of magnitude higher than that
of HC emissions. As seen in Fig. 2, the highest CO emissions
are observed for the flight phases of descent (34.0 kg), taxi-in (25.3 kg) and
taxi-out (7.4 kg) which is similar to the case for HC emissions. CO emissions
are found to be quite low for the remaining flight phases, for which power settings
are relatively high.
The highest levels of NOx emissions are observed during the climb,
cruise and descent phases. The average NOx emissions for the five
flights is 31.3 kg, of which 16.0% is produced in ground operations, 34.6% in
climb, 20.6% in cruise and 15.9% is descent. Taxi-in exhibits higher NOx
emissions than taxi-out, due to the longer phase duration. Emissions and the
fuel consumptions for the second route (ADB-SAW) are also shown in Fig.
For short range domestic commercial flights, several key findings and conclusions
can be drawn from the results of the present emissions study:
||For the common emission species, CO, HC and NOx,
linear and nonlinear models are developed for all of the flight phases,
for two routes and two engine models. The models are based on the ICAO emission
measurements. Actual flight data are obtained from flight data records.
The models permit evaluation of emissions for all of flight phases for various
fuel flow rates
||The models can be used for the other aircraft types using the same types
of the engines. That is, for the same engine type and known fuel flow rates,
emissions quantities can be calculated using the developed models
||A breakdown of emissions by flight phase is obtained and the findings
agree well for flights on both routes. The highest CO and HC emissions are
found in the descent phase, followed by the taxi phases (in and out). The
highest NOx emissions are found in the climb phase, followed
by cruise and descent. There are less but not negligible NOx
emissions in the taxi phase due to the high taxi duration
||The mean total flight emissions are calculated as 8 kg of HC, 75 kg of
HC and 31 kg of NOx for the AYT route and 6 kg of CO, 60 kg of
HC and 28 kg of NOx for the ADB route
||The average ground time of the flights are calculated as 22-25% of the
total flight time for both routes. Since the HC and CO emissions are mostly
produced in the lower power settings of the engine, decreasing the taxiing
time provides significant abatement of those two emissions
The authors thank Pegasus Airlines for their co-operation and the Sabiha Gokcen
Airport Authority for its kind assistance in data acquisition. The authors also
thank Anadolu University for financial support.