lable at ScienceDirect

Atmospheric Environment 100 (2015) 218e229

Contents lists avai

Atmospheric Environment

journal homepage: www.elsevier .com/locate/atmosenv

Emissions of NOx, particle mass and particle numbers from aircraftmain engines, APU's and handling equipment at Copenhagen Airport

Morten Winther a, *, Uffe Kousgaard b, Thomas Ellermann a, Andreas Massling a,Jacob Klenø Nøjgaard a, Matthias Ketzel a

a Aarhus University, Department of Environmental Science, Frederiksborgvej 399, 4000 Roskilde, Denmarkb Routeware, Trægården 9, 4000 Roskilde, Denmark

h i g h l i g h t s

� 5 � 5 m high resolution NOx, PM and PN emission inventory for Copenhagen Airport.� Detailed aircraft and handling equipment results; entire airport and inner apron area.� Handling is a large NOx and PM source, and a small PN source at the inner apron.� ICAO FOA3.0 PM results suggest that over half of aircraft PM stems from fuel sulphur.� High/low sets of emission factors change the aircraft PN emissions by a factor of 14.

a r t i c l e i n f o

Article history:Received 11 July 2014Received in revised form23 October 2014Accepted 25 October 2014Available online 30 October 2014

Keywords:NOx

PMParticle numbersAircraft main enginesAPUHandling equipment

* Corresponding author.E-mail address: mwi@dmu.dk (M. Winther).

http://dx.doi.org/10.1016/j.atmosenv.2014.10.0451352-2310/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

This paper presents a detailed emission inventory for NOx, particle mass (PM) and particle numbers (PN)for aircraft main engines, APU's and handling equipment at Copenhagen Airport (CPH) based on timespecific activity data and representative emission factors for the airport. The inventory has a high spatialresolution of 5 m � 5 m in order to be suited for further air quality dispersion calculations. Results areshown for the entire airport and for a section of the airport apron area (“inner apron”) in focus. Themethodology presented in this paper can be used to quantify the emissions from aircraft main engines,APU and handling equipment in other airports. For the entire airport, aircraft main engines is the largestsource of fuel consumption (93%), NOx, (87%), PM (61%) and PN (95%). The calculated fuel consumption[NOx, PM, PN] shares for APU's and handling equipment are 5% [4%, 8%, 5%] and 2% [9%, 31%, 0%],respectively. At the inner apron area for handling equipment the share of fuel consumption [NOx, PM, PN]are 24% [63%, 75%, 2%], whereas APU and main engines shares are 43% [25%, 19%, 54%], and 33% [11%, 6%,43%], respectively. The inner apron NOx and PM emission levels are high for handling equipment due tohigh emission factors for the diesel fuelled handling equipment and small for aircraft main engines dueto small idle-power emission factors. Handling equipment is however a small PN source due to the lownumber based emission factors. Jet fuel sulphur-PM sensitivity calculations made in this study with theICAO FOA3.0 method suggest that more than half of the PM emissions from aircraft main engines at CPHoriginate from the sulphur content of the fuel used at the airport. Aircraft main engine PN emissions arevery sensitive to the underlying assumptions. Replacing this study's literature based average emissionfactors with “high” and “low” emission factors from the literature, the aircraft main engine PN emissionswere estimated to change with a factor of 14.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

In the recent years an increased concern about the health effectsin humans due to airport emissions is observed. Aircraft main

engines, APU's (Auxiliary Power Units) and handling equipment arethe most important sources of air pollutant emissions in an airport,and among the vast number of emitted pollutants a special focus isgiven to NOx, particle mass (PM) and particle number (PN). NOx isemitted in high quantities and traditionally considered as a goodtracer for all kind of engine emissions while PM and PN areconsidered to cause various adverse health effects (e.g. Pope andDockery, 2006; Schwarze et al., 2006). Particle size is a crucial

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 219

parameter and often are smaller particles e being emitted in largenumbers during engine combustion e considered to cause largereffects on human health, due to their ability to penetrate deeperinto the lungs and translocate inside the human body to other or-gans (e.g. Oberd€orster et al., 2004; Diapouli et al., 2007). A linkbetween elevated air pollutant concentrations of NOx and particles(especially mass based) and a large number of diseases such ascardiovascular diseases, pulmonary diseases, various cancer types,asthma, and diabetics could be established (Raaschou-Nielsen et al.,2012; Hertel et al., 2013).

During the recent years several air quality measurement cam-paigns have been conducted in order to assess the impact of airportemissions on the air quality in airports and in the vicinity of air-ports. Some important examples of such measurement studies arelisted in the following.

Carslaw et al. (2006) measured the NOx concentration at sevengeographical sites around London Heathrow Airport in order todetermine the emission impact from aircraft activities. Schürmannet al. (2007) measured NOx and other emissions from airportsources on the air quality in Zurich Airport. For Los Angeles Inter-national Airport Westerdahl et al. (2008) measured NOx and ultrafine particle (UFP) concentrations in the vicinity of the airport todetermine the impact of airport emissions, and Zhu et al. (2011)made detailed PM2.5 and UFP measurements in the airport and ata background reference site in order to determine the emissionimpacts from aircraft take offs. Hsu et al. (2012) made UFP mea-surements at four monitoring sites surrounding T.F. Green Airportin Warwick in order to determine impact from aircraft landing andtake off activities (LTO's). At Copenhagen Airport Ellermann et al.(2011) conducted long term air quality measurements of PM2.5and NOx at the apron and two background measurement sites andmade screening measurements of UFP in order to assess the airquality in the airport.

The air quality measurements performed at Copenhagen Airportwere a part of a larger air quality assessment project “Investigationof the air pollution on the inner apron at Copenhagen AirportKastrup in relation to working environment” initiated due to thehealth concern of the airport staff working at the airport apron(Ellermann et al., 2011). In order to carry out the most reliable airquality assessment for the airport and to support future emissionabatement measures, it was a project goal from the beginning tocreate a precise source specific emission inventory in a high tem-poral (minutes) and spatial (5 m � 5 m) resolution for the entireairport and for a part of the airport apron area in focus referred to as“inner apron area”. This area is located between the terminal fin-gers and is characterized by aircraft movements to/from the gates,and large activities made with APU and handling equipment.

The inventory covered the source categories aircraft main en-gines, APU, handling equipment, and road transportation vehiclesas a minor source, and included the emissions of CO2, SO2, CO, HC,NOx, NO2 and PM. In a follow-up study “Inventory of emissionsincluding particle numbers at the inner apron area in CopenhagenAirport” documented by Winther et al. (2014), estimates of par-ticulate numbers (PN) were added to the inventory.

This paper aims to explain the emission inventory for Copen-hagen Airport. Source specific results are presented as totals for theentire airport and for the inner apron area. Emissions are alsopresented in a 5 m � 5 m spatial resolution suited for the subse-quent modelling of air quality in the airport. In this paper, the in-ventory results will focus on fuel consumption and the emissions ofNOx, PM and PN due to their important health effects as outlined inthe beginning of this paper.

The temporally and spatially very detailed inventory forCopenhagen Airport and the inner apron area, and the derivedsource contributions for main engines, APU's and handling

equipment, serve as a valuable new contribution to the researchfield. The precise emission estimates for handling equipment in theairport based on equipment type and technology specific activitydata also bring important new knowledge to the researchcommunity.

Section 2 presents the inventory's activity data, emission factorsand the emission calculation method. Section 3 explains theemission results for the inner apron area and for the total airportand spatial distributions of the emissions are shown as well.Moreover a sensitivity analysis is presented that examines the PMemission changes due to the use of zero sulphur jet fuel. Thesensitivity of PN emissions on the use of “low” and “high” PNemission factors from the literature is shown as well.

2. Method

The emission inventory uses detailed activity data for aircraftmain engines, APU's and handling equipment together with thetime-in-modes and location of their use at Copenhagen Airport(CPH). The flight activity data for CPH represent four days in 2009with preferable use of each of the four runways, chosen to produceemission results suited as input for subsequent air quality disper-sion modelling depending on wind direction. Handling activitiesare specified by equipment types and grouped according to aircraftsize. The time resolved airport activities are digitalized in a5 m � 5 m resolution on a GIS map of the airport as shown in Fig. 1.The vertical limit of the considered emissions is 100 m. Emissionsabove 100m are not regarded as relevant for local airport air qualityimpacts due to the high dilution of the emissions during the tur-bulent transport from these height levels down to earth surfacelevel.

Engine/power mode specific fuel flows (g/s) and NOx and PMemission indexes (EI's, g/kg fuel) are prepared in this study, and forparticulate numbers (PN) more general number emission indexes(EIn's, #/kg fuel) have been established. Emission rates for NOx andPM in g/s and number based emission rates for PN in #/s arederived from the fuel flows, emission indexes and the time-in-mode for each engine/power mode obtained from the real timespecific activity data for CPH. The emissions per grid cell arecalculated as the product of the emission rate and the time intervalcalculated by the digital activity map.

The digital activity map at CPH uses a number of assumptionsregarding priority taxiways, taxi speed, runway acceleration/deceleration and climb/landing gradients for aircraft and the spatiallocation of handling activities and APU usage close to the gates. Forfurther details please see Winther et al. (2006, 2014). The digitalactivity map is a further development of the one used to investigatethe odour nuisances from aircraft main engines and APU's (Fengeret al., 2006; Winther et al., 2006).

Fig. 1 shows a map of CPH. Terminal gates are numbered (e.g. B7or C28) andmarked with a small (black) dot. Themain engine start-upmarks are designated with larger (brown) dots (e.g. P or Q1). Theaircraft taxi ways close to the gates are visible as (green and red)lines and connect to the shared taxiways that lead from/to therunways (blue lines).

A map of the inner apron (area in pink frame) at CPH also infocus in this study is shown in Fig. 2.

2.1. Activity data

2.1.1. Flight operationsThe airport has provided flight activity data for four days in 2009

with preferable use of each of the runways 12, 30, 04Lþ 04R and22Lþ 22R. The data consist of aircraft type, registration number,airline operator, gate, off/on block time, specification of start/

Fig. 1. Detailed map of CPH showing terminal gates (black dots marked with letters AeF and a number), main engine start-up marks (brown dots), aircraft taxiways (red lines orgreen lines near the gates) and parts of the runways (blue lines). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of thisarticle.)

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229220

landing runway and time. The number of flight operations for thefour days split into aircraft arrivals and departures are displayed inSupporting information 1.

2.1.2. APU usageNo specific information is available for the use of APU's at CPH.

Instead the International Civil Aviation Organization (ICAO) pro-vides specific times-in-modes for the use of APU's (ICAO, 2011).These times-in-modes are shown in Table 1. The total APU timeprior to off-block is split into APU start up and Boarding corre-sponding to different APU engine loads.

The airport authorities at CPH generally accepts the APU's to runfor 5 min following on-block and prior to off-block, which issomewhat shorter than ICAO's time periods for departures given inTable 1. However, at CPH the duration of APU usage prior to off-block is allowed to exceed 5 min for outside temperatures lowerthan �10 �C or above 25 �C, and under these conditions the APU isallowed to run for 45 min on large aircrafts and 15 min on smallaircrafts (CPH Local Regulations, Chapter 9). In summary it wasconcluded to apply ICAO's recommended APU times-in-modes forCPH.

2.1.3. Usage of handling equipmentFor size classification the handling companies at CPH divide the

aircraft into four groups B-E, from the smallest aircraft in “B” to the

largest jets in “E” (Supporting information 1). The medium sized“C” aircraft (B737/MD80/A319/A320/A321/CRJ/ATR aircraft series)accounts for more than 80% of total arrivals and departures at CPH(Table 2). The larger “D” size aircraft (B757/A310/B767 series) and“E” size aircraft (A330/A340/B777/B747 series) in each case accountfor a little less than 3% of total arrivals and departures at CPH. The“B” aircraft less frequently use the gates at the inner apron. Theseaircraft are used primarily for short trips with a low demand forseating capacity, and use the domestic terminal gates and remotegates at CPH (A18-A33, F and E gates shown on Fig. 1).

The information of handling operations per equipment typeprovided by the handling companies is shown in Table 3. Thehandling activities are assumed to take place on the right side of theaircraft in the area stretched by the aircraft length and the winglength. For the subsequent emission calculations, a complete list ofthe handling equipment is available from the handling companiescomprising equipment ID number, fuel type, engine size, and en-gine age/or EU emission stage (for reasons of confidentially thesedetails are not shown).

2.2. Emission factors

2.2.1. Main enginesThe engine type and number of engines for each aircraft in CPH

flight statistics is accessible from “JP Airline-Fleets 2009/2010”

Table 1APU times-in-modes used at CPH. The APU time during push-back is calculated.

Arrival APUstart-up

Boarding Push-back Main enginestart-up

APU load Normal Start-up Normal Normal High2 engines 300 s 180 s 216 s Calculated in modela 35 s4 engines 300 s 180 s 318 s Calculated in model 140 s

a The aircraft is assumed to be pulled by the push-back tractor with a speed of5 km/h (1.5 m/s) along the green lines from the gates toward the nearest point ofmain engine start-up (brown dots in Fig. 1).

Fig. 2. Map of the Inner apron (area in pink frame) situated between the terminal fingers. (For interpretation of the references to colour in this figure legend, the reader is referred tothe web version of this article.)

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 221

(www.flightglobal.com). Jet engine fuel flows (kg/s), NOx emissionindexes (EI's, g/kg fuel) and soot numbers comes from ICAO's En-gine Exhaust Emission Database (www.caa.co.uk) and PM EI's arecalculated by using the ICAO FOA3.0 method (ICAO, 2011). Turboprop engine fuel flows and NOx EI's comes from the Swedish FOIturbo prop engine exhaust emission database (Totalf€orsvaretsForskningsinstitut, www.foi.se) and PM EI's are derived fromRindlisbacher (2009).

Table 2Number of arrivals and departures by aircraft size category at CPH during four days, as t

Operation Total airport

B C D E To

Arrival 202 1207 43 44 14Departure 210 1203 40 40 14% of total 13.8 80.6 2.8 2.8 1

For aircraft main engines, idle engine power fuel and emissiondata are used during landing and taxiing activities at CPH; the idleengine power assumption for landing corresponds with the lowengine power setting often being used by aircraft engines duringthe final descent from 100 m vertical height to the point of touchdown at CPH (pers. comm. Martin Porsgaard, Dir. Environment &CSR, SAS). During take off and climb out to 100 m vertical height,take off power fuel and emission data are used.

The aircraft specific emission rates are derived from the enginespecific fuel rate, the EI's and the number of engines fitted to theaircraft. The EI's and fuel and emission rates for all engines areshown in Supporting information 2. The aggregated fuel andemission rates for aircraft main engines are shown in the Table 6(total CPH and inner apron area), derived from the main enginefuel and emission results presented in Section 3.

During the recent years several measurement studies have beenmade of particulate numbers in the exhaust plume from aircraftmain engines related to kg fuel burned (number emissionindex ¼ ‘EIn’, unit: #/kg fuel). Table 4 shows the EIn minimum andmaximum values measured at idle and take off engine loads,

otal and related to the gates at the inner apron.

Inner apron

tal B C D E Total

96 16 417 20 16 46993 16 406 20 13 45500.0 3.5 89.1 4.3 3.1 100.0

Table 3List of handling equipment, duration of handling period, working time, and engine load factor for handling machinery used at Copenhagen Airport.

Arrival Departure

Aircraft category / B C D E B C D EHandling period (min) / 15 20 30 40 15 20 30 40Handling equipment type Working time (min) Working time (min) Load factorBaggage truck 9 10 15 25 9 10 15 25 0.15Conveyor belt 10 20 20 20 10 20 20 20 0.15Push-back at gate 0 0 0 0 10 10 10 10 0.15Push-back movinga 0 0 0 0 Calc. Calc. Calc. Calc. 0.75Container loader 0 15 27.5 35 0 15 27.5 35 0.45Container transporter 0 15 27.5 35 0 15 27.5 35 0.35Fuel (dispenser truck) 10 15 30 50 0 0 0 0 0.1Fuel (refuelling truck)b 10 15 30 50 0 0 0 0 0.1Cleaning high loader 0 0 10 15 0 0 10 15 0.45Cargo/Post tractor 0 5 5 5 0 5 5 5 0.15Toilet truck 0 0 0 0 0 10 20 20 0.25Catering B/C/D/E 1 3 5 5 0 0 0 0 0.1/0.2/0.22/0.22Water truck 0 0 0 0 0 7.5 15 15 0.25

a The time depends on the distance from the gate to the point for main engine start-up and is calculated in the model.b The fuel truck is used for refuelling of aircraft at the gates not equipped with fuel pipe lines (Supporting information 1).

Table 4Minimum and maximum EIn values (#/kg fuel) for main engines at idle and take off power settings for the different measurement studies reported in the literature.

Study Airport Aircraft types/engine types Engine mode EIn min EIn max

Mazaheri et al. (2009) Brisbane B737, B767, B777, A320, A330 Idle 3.29Eþ 16 3.78Eþ 16Take off 2.09Eþ 16 5.42Eþ 16

Mazaheri et al. (2011) Brisbane B737, B767, A320, A330 Idle 1.63Eþ 15 8.36Eþ 15Take off 4.65Eþ 16 3.15Eþ 17

Herndon et al. (2005) Boston, Logan Average Idle 2.10Eþ 16Take off 8.80Eþ 15

Herndon et al. (2008) Atlanta,Hartsfield

CF34, JT8D, CFM56, PW2037, CF6 Idle 4.00Eþ 15 8.20Eþ 15

Take off 1.80Eþ 15 5.60Eþ 15Lobo et al. (2012a) Oakland CFM56-3B/7B,V2500-A5,

JT8D, CF6-50/80, CF34-3BIdle 2.54Eþ 16 2.09Eþ 17Take off 2.04Eþ 16 7.08Eþ 16

Johnson et al. (2008) Brisbane B737 Idle 2.40Eþ 16 3.70Eþ 16Take off 9.00Eþ 15

Zhu et al. (2011) Los Angeles, LAX Average Take off 3.40Eþ 16Hu et al. (2009) Santa Monica Average Take off 5.00Eþ 16Kinsey et al. (2010) CFM56-7B24/3B1, AE3007A1/1,

PW4158, RB211-535E4BIdle 4.59Eþ 15 9.43Eþ 16Take off 1.05Eþ 16 3.58Eþ 16

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229222

respectively, for each study. Due to the large variation in specificEIn's reported it has been decided to derive the average values foridle (3.91Eþ 16/kg fuel) and take off (4.62Eþ 16/kg fuel), and usethese average numbers as the basis for the particulate numbercalculations made in the present study.1

For APU, the basis fuel consumption, NOx and PM emissionfactors (kg/h) come from ICAO (2011). Fuel consumption and NOx

factors distinguish between different aircraft size (seating capac-ity), old/new aircraft types (c.f. Supporting information 2) and APUengine loads (start-up/normal/high; c.f. Table 1). For PM theemission factors distinguish between short and long haul aircrafttypes.2

Particle number basedemission factorsmeasured forAPUare veryscarce. This study uses EIn data from Lobo et al. (2012b) measured atidle (4.97Eþ 16/kg fuel) and high (2.70Eþ 16/kg) APU loads. For APUnormal running conditions an average value of 3.84Eþ 16/kg fuel isused. Since EIn data for APU is based on measurement data for one

1 Measurement data from Kinsey et al. (2010) was provided directly by JohnKinsey, USEPA. Unfortunately the data was received after deadline of the calcula-tions in our study.

2 For PM, ICAO (2011) provided fuel related emission factors for short and longhaul aircraft types of 0.3125 and 0.1333 g/kg fuel, respectively. The A330, A340,B767, B777 and B747 aircraft types are regarded as long haul in the calculations.

engine only, the number based emission factors are likely tochange when new measurement results become available.

The full set of basis emission factors for APU is very compre-hensive and is consequently not shown here. For more detailsplease refer to Supporting information 2. For different powermodes and types of operations, aggregated fuel and emission ratesand fuel related emission factors are shown in Table 6 (total CPHand inner apron area), derived from the APU fuel and emissionresults presented in Section 3.

2.2.2. Handling equipmentFuel consumption and NOx and PM emission factors (g/kWh) for

diesel fuelled handling equipment come from EMEP/EEA (2013),grouped according to the EU emission directive stages Euro IeV forroad transport vehicles (e.g. cleaning and catering trucks) and theemission stages IeIV (and older) for non road engines. The stock ofhandling equipment contains almost 650 items. For reasons ofconfidentiality the stock list is not publically available. Moreemission factor descriptions and size and technology stratifiedemission factors are given in Supporting information 2.

For each item of handling equipment, fuel and emission ratesare derived by combining the g/kWh specific fuel consumption/emission factor with the technology level and engine size specificstock data provided by the handling companies at CPH, and theaverage information of engine load factor and equipment working

Table 5Table of data source references for activity data and emission factors used in this project.

Source type Data type Reference

Activity data Flight operations Id, aircraft type, gate, runway CPH statisticsEngine type, no. of engines JP Airline-fleets 2009/2010

APU Times-in-modes ICAO doc 9889Handling equipment Stock, handling periods, working time, engine loads CPH handling companies

Emission factors Jet engines NOx EI, soot number ICAO engine exhaust emission databasePM EI FOA3.0 method

Turbo props NOx EI FOI turbo prop engine exhaust emission databasePM EI Rindlisbacher (2009)

Jet and turbo props Average PN EIn's Literature referencesAPU's NOx and PM EI ICAO doc 9889

PN EIn Lobo et al. (2012b)Handling equipment NOx and PM EI, PN EIn EMEP/EEA (2013)

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 223

time shown in Table 2. Aggregated fuel and emission rates and fuelrelated emission factors are shown in Table 6 (total CPH and innerapron area), derived from the results presented in Section 3. Theweighted factors are given for handling activities during aircraftarrival and departures, and separately for push-back tractors at gateand during motion.

Due to lack of specific data for handling equipment, particlenumber data are taken from the European road transport emissionmodel COPERT IV.(EMEP/EEA, 2013). The COPERT IV factors in #/km

Table 6Fuel consumption and emissions of NOx and particles (mass and numbers), fuel and emequipment for the four days at CPH represented by flight data. Totals for CPH and for th

Area Source Operation Times � 103

Fuelkg

NOx

kgPMkg

Total CPH Main engines Landing (descent) 72 14,288 61.7 1.7Main engines Landing (runway

deceleration)35 7741 33.8 0.9

Main engines Landing (runway taxi) 28 5280 22.6 0.6Main engines Taxi arrival (taxi way) 463 99,863 434.4 11.5Main engines ME during push back 18 1966 8.2 0.2Main engines ME during ME start 50 12,839 56.7 1.4Main engines Taxi departure (taxi way) 438 89,877 388.7 10.6Main engines Taxi departure (queing) 638 140,708 611.0 15.9Main engines Take off (runway) 55 113,683 2825.4 17.5Main engines Take off (climb out) 14 30,283 745.4 4.6APU APU arrival 386 11,230 90.5 3.4APU APU start-up 232 4650 26.4 1.4APU APU boarding 283 8309 67.5 2.4APU APU during pushback 69 1991 16.4 0.6APU APU during ME start 50 1961 17.4 0.5Handling Handling arrival 1809 5982 197.1 13.2Handling Handling departure 1794 7621 278.1 17.5Handling Push back at gate 620 929 33.0 1.1Handling Push back moving 69 515 18.4 0.6Main engines Total 1811 516,527 5188 65APU Total 1019 28,141 218 8Handling Total 4291 15,046 527 32All Total 7121 559,714 5933 106

Innerapron

Main engines Taxi arrival (taxi way) 13 3145 13.9 0.4

Main engines ME during ME start 12 3270 14.7 0.4Main engines Taxi departure (taxi way) 3 662 2.9 0.1APU APU arrival 126 3667 29.9 1.1APU APU start-up 74 1493 8.5 0.5APU APU boarding 90 2626 21.6 0.8APU APU during pushback 27 796 6.6 0.2APU APU during ME start 12 484 4.4 0.1Handling Handling arrival 551 1949 65.0 4.3Handling Handling departure 531 2465 90.5 5.5Handling Push back at gate 273 417 14.1 0.5Handling Push back moving 27 209 7.1 0.3Main engines Total 28 7077 32 1APU Total 329 9065 71 3Handling Total 1383 5041 177 11All Total 1739 21,182 279 14

are transformed into EIn values (#/kg fuel) by dividing with thecorresponding fuel consumption factors (kg/km) at urban driving.The derived EIn's are relatively constant for diesel cars, vans andtrucks technologies in COPERT IV without particulate filtersinstalled, which is also the case for the handling equipment at CPH.Hence, an EIn value of 3.1Eþ 15/kg fuel derived for a small sizedEuro II/III truck (3.5e7.5 tons total weight category) is selected inour case. A discussion on handling equipment EIn's is provided inthe conclusions of this paper.

ission rates and fuel related emission indices for main engines, APU and handlinge inner apron area at different power modes and types of operation.

PNNo.

Fuelg/s

NOx

mg/sPMmg/s

PN#/s

NOx

g/kg fuelPMmg/kg fuel

PN#/kg fuel

5.59Eþ 20 199 858 24 7.77Eþ 15 4.32 119 3.91Eþ 163.03Eþ 20 218 953 25 8.53Eþ 15 4.37 116 3.91Eþ 16

2.07Eþ 20 187 801 23 7.32Eþ 15 4.28 122 3.91Eþ 163.91Eþ 21 216 938 25 8.43Eþ 15 4.35 115 3.91Eþ 167.69Eþ 19 112 467 11 4.38Eþ 15 4.18 96 3.91Eþ 165.02Eþ 20 258 1140 27 1.01Eþ 16 4.41 106 3.91Eþ 163.52Eþ 21 205 887 24 8.02Eþ 15 4.33 118 3.91Eþ 165.50Eþ 21 221 958 25 8.63Eþ 15 4.34 113 3.91Eþ 165.25Eþ 21 2058 51,147 316 9.50Eþ 16 24.85 154 4.62Eþ 161.40Eþ 21 2103 51,763 317 9.71Eþ 16 24.61 151 4.62Eþ 164.31Eþ 20 29 234 9 1.12Eþ 15 8.06 298 3.84Eþ 162.31Eþ 20 20 114 6 9.98Eþ 14 5.68 300 4.97Eþ 163.19Eþ 20 29 239 9 1.13Eþ 15 8.12 294 3.83Eþ 167.63Eþ 19 29 238 9 1.11Eþ 15 8.22 301 3.83Eþ 165.30Eþ 19 39 350 11 1.06Eþ 15 8.88 269 2.70Eþ 161.85Eþ 19 3 109 7 1.03Eþ 13 32.95 2206 3.10Eþ 152.36Eþ 19 4 155 10 1.32Eþ 13 36.49 2294 3.10Eþ 152.88Eþ 18 1 53 2 4.64Eþ 12 35.52 1234 3.10Eþ 151.59Eþ 18 7 267 9 2.32Eþ 13 35.69 1239 3.10Eþ 152.12Eþ 221.11Eþ 214.66Eþ 192.24Eþ 221.23Eþ 20 246 1087 28 9.61Eþ 18 4.43 114 3.91Eþ 19

1.28Eþ 20 275 1235 32 1.07Eþ 19 4.49 116 3.91Eþ 192.59Eþ 19 232 1017 26 9.06Eþ 18 4.39 113 3.91Eþ 191.41Eþ 20 29 237 9 1.12Eþ 18 8.16 298 3.84Eþ 197.42Eþ 19 20 116 6 1.01Eþ 18 5.70 302 4.97Eþ 191.01Eþ 20 29 241 9 1.12Eþ 18 8.22 296 3.84Eþ 193.05Eþ 19 29 240 9 1.11Eþ 18 8.29 302 3.84Eþ 191.31Eþ 19 41 373 11 1.10Eþ 18 9.18 263 2.70Eþ 196.04Eþ 18 4 118 8 1.10Eþ 16 33.33 2228 3.10Eþ 187.64Eþ 18 5 170 10 1.44Eþ 16 36.71 2248 3.10Eþ 181.29Eþ 18 2 52 2 4.74Eþ 15 33.85 1194 3.10Eþ 186.49Eþ 17 8 260 9 2.36Eþ 16 34.10 1200 3.10Eþ 182.77Eþ 203.59Eþ 201.56Eþ 196.52Eþ 20

Fig. 3. Percentage shares of fuel consumption, NOx and particle emissions (mass andnumbers) for aircraft main engines (landing, taxi, take off), APU's and handlingequipment at CPH (total airport).

Fig. 4. Percentage shares of fuel consumption, NOx and particle emissions (mass andnumbers) for aircraft main engines, APU's and handling equipment at the inner apronarea.

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229224

Table 5 presents an overview of the references for the datasources used to provide activity data and emission factors in thisproject.

2.3. Calculation method

Emission rates (g/s) for the main engines, APU, engine start-upand handling equipment are derived from the emission factors.Emissions and fuel consumption are calculated for each 5 m � 5 mgrid cell as the product of emission rate (g/s) and the calculatedactivity interval (s) estimated by the digital activity map of theairport:

DEj;k;l�t; i

�¼ ERj;k;l$Dtj;k;l

�t; i

�

Where DE ¼ Emission/fuel consumption (g) in the i'th grid cell attime t, ER ¼ Emission rate/fuel consumption (g/s), j ¼ source type(main engines, APU, handling equipment), k ¼ engine id (mainengines), aircraft size/age (APU), handling weighted average(weighted from equipment type, technology level, engine size, loadfactor, working time, c.f. Section 2.2), l ¼ power mode (main en-gines: idle, take off; APU: normal, start up, high; handling:running), i ¼ Grid cell number in digital map, Dt ¼ Time interval inthe i'th grid cell, t ¼ Time of the day measured in seconds(0 < t < 86,400).

The particulate numbers are found as the product of the fuelconsumption, DE, and the #/kg fuel factor, EIn:

PNj;k;l ¼ EInj;k;l$DEj;k;l�t; i

�

Where PN ¼ Number of particles, EIn ¼ Number of particles per kgfuel (#/kg fuel). For main engines and APU EIn is not stratified ac-cording to “k”. For handling equipment EIn is not stratified ac-cording to “k” and “l” (c.f. Section 2.2).

3. Results

Table 6 shows the fuel consumption and emissions of NOx andparticles (mass and numbers) calculated for main engines, APU andhandling equipment for the four days at CPH represented by flightdata. Totals are given for CPH and for the inner apron area as well asderived fuel and emission rates and fuel related emission indices,split into different power modes and types of operation. Detailedemission results per aircraft type and handling equipment type areprovided in Supporting information 3. The calculated emissionresults are generally explained by the size of the fuel relatedemission factors and the calculated fuel consumption (Table 6). Theresults will be discussed in the following.

3.1. Total airport

Fig. 3 shows the total percentage shares of fuel consumption,NOx and particle emissions (mass and numbers) for aircraft mainengines, APU's and handling equipment used at CPH. The per-centage results for main engines are split into landing (descent andrunway), taxi (five operation modes) and take off (runway andclimb out).

The largest shares (percentage values in brackets) of fuel con-sumption (93%), NOx, (87%), PM (61%) and PN (95%) are calculatedfor aircraft main engines at CPH. For APU and handling equipment,the fuel consumption[NOx, PM, PN] shares are 5%[4%, 8%, 5%] and 2%[9%, 31%, 0%], respectively.

For handling equipment (inclusive push back tractors) the highPM emission share of 31% is explained by the generally high PM

emission factors for the diesel engines used in the handling situa-tions (Table 5), given that the fuel consumption share for theseengines is only 2%. During aircraft take off the emissions of NOx

from aircraft main engines are high (60% of total) due to the highcombustion temperature at full engine power and hence high NOx

emission factors. The large PN shares calculated for aircraft mainengines during taxiing (60%), take off (30%) and landing (5%) is dueto the large fuel consumption (Fig. 3). For APU the shares of fuelconsumption and PN are 5%. Due to the very low EIn values, the PNemission share for handling equipment is close to zero.

Spatial maps with a resolution of 5 m � 5 m of the total NOx andparticle (mass and numbers) emissions for the whole airport forone day with preferable use of lane 22 are provided in SupportingInformation 3.

3.2. Inner apron

In relation to the total airport fuel consumption and emissions,the contributions from the limited spatial area comprised by theinner apron examined in this study are small. Derived from theresults in Table 6, the emissions at the inner apron area for NOx, PMand PN comprise about 4.7%, 13% and 2.9% of the total airportemissions, and the total fuel share is 3.8%.

Fig. 4 shows the percentage shares of fuel consumption, NOx andparticle emissions (mass and numbers) for aircraft main engines,APU's and handling equipment used at the inner apron area.

The fuel consumption shares for APU, main engines andhandling equipment are 43%, 33% and 24%, respectively.

Handling equipment has the largest emission shares (percent-age shares in brackets) of NOx (63%) and PM (75%), due to the highfuel related emission factors for the diesel fuelled handling equip-ment (Table 6). Having somewhat lower emission factors, the large

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 225

fuel consumption for APU is the main reason for the high NOx andPM emission shares of 25% and 19%, respectively, in the APU case.For main engines the NOx and PM emission shares are small, 11%and 6%, respectively, due to the very small emission factors duringtaxiing.

In opposition to particulate emissions by mass (PM), handlingequipment is a small source for particle numbers (PN) at the innerapron area, due to the generally very low emission factors (#/kgfuel) for handling engines compared to aircraft engines. The PNemission share for handling equipment amounts to only 2.4%. Thelargest emission share is calculated for APU (54%). For main enginesthe emission share becomes slightly smaller (43%).

Fig. 5 shows the fuel consumption and NOx, PM and PN emis-sions per hour for the inner apron calculated for one day withpreferable use of lane 22 at CPH. The timely variation of thecalculated results can be explained by the development of theaircraft operations during the day also shown in Fig. 5, split intoarrivals and departures. The emission explanations are the same asgiven above.

In Fig. 6 the average daily emissions (per 5 m � 5 m) of NOx areshown separately for handling, APU andmain engine activities for asection of the airport apron area (Not identical with the ‘innerapron area’ in Fig. 2), that has been chosen in order to illustratemore clearly the emission trails of the moving aircraft. Supportinginformation 3 contains activity specific spatial maps for particles

0

10

20

30

40

50

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

No.

hou

r-1

Aircra ac vity - runway 22

Arrival Departure

0

1

2

3

4

5

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

kg h

our-1

NOx emissions - inner apron

APU Handling Main engines

0

2E+18

4E+18

6E+18

8E+18

1E+19

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

# ho

ur-1

PN emissions - inner apron

APU Handling Main engines

Fig. 5. Fuel consumption and emissions per hour for the inner apro

(mass and numbers) and spatial maps for total NOx and particles(mass and numbers). Please note that the emission scale of Fig. 6 isnot linear.

Significant emission contributions are visible on the right side ofthe aircraft, from the handling activities as well as from the push-back truck and APU activities prior to off-block (Fig. 6). These arevisible as red marks on the inner and outer side. The emission trailsfrom the pushback truck and APU are also clear during taxiing fromthe gates to the start-up marks. By the start-up mark the emissionsfrom APU and main engine are visible too. Furthermore, thecontribution from the main engines is clearly visible during taxiingtowards the runways, and when the aircraft arrive at the gates. Thedim APU andmain engine emission trails closest to the gates visibleon Fig. 6 are not located entirely precise. In the digitalizationmodel,the emissions are attached to the nose of the aircraft when theaircraft is moving.

Spatial maps for total NOx and particles (mass and numbers) areprovided in Supporting Information 3.

3.3. Sensitivity of emissions to sulphur content and otherparameters

The impact of usage of jet fuel with zero sulphur content on PMemissions from jet aircraft main engines is examined in this section.Fig. 7 show the percentage reductions in EI PM obtained with the

0

0,05

0,1

0,15

0,2

0,25

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Tonn

es h

our-1

Fuel consump on - inner apron

APU Handling Main engines

0

50

100

150

200

250

300

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

g ho

ur-1

PM emissions - inner apron

APU Handling Main engines

n calculated for one day with preferable use of lane 22 at CPH.

Fig. 6. Total NOx emissions separately for handling, APU and main engines per 5 m � 5 m for a section of the airport apron area. Average day.

Fig. 7. Specific percentage reductions of EI PM for jet aircraft main engines obtained by using zero sulphur jet fuel at CPH.

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 227

FOA3.0 method (Section 2.2), calculated as the percentage differ-ence between the EI's based on the average jet fuel sulphur contentof 942 ppm for CPH and zero jet fuel sulphur content.

Weighted by fuel consumption, the average EI PM reductions formain engines at CPH become 64% and 47% for idle and take offpower, respectively. The total PM emission reductions for mainengines are the same, and the overall PM emission reduction isexpected to be 58%. PM EI's based on zero jet fuel sulphur contentare shown in Supporting information 2.

The PN emissions are expected to be sensitive to the sulphurcontent as well as distance of the measurement equipment to thepassing aircraft, exhaust plume evolution, meteorological condi-tions (wind direction and speed, ambient air temperature andhumidity) (e.g. Wong et al., 2008; Timko et al., 2013; Hsu et al.,2012).

Therefore PNemission sensitivitywas analysed byusing two setsof “low” and “high” EIn values reported in the literature as the inputfactor basis for detailed aircraft/engine calculations at CPH. The“low” and “high” EIn values for jet aircraft main engines have beenmeasured by Herndon et al. (2008) for Hartsfield-Jackson Atlantaairport and by Lobo et al. (2012a) for Oakland International airport.The EIn values have been measured in the exhaust plumes for themost frequently used jet aircraft types during taxi and take off ac-tivities and on the same time the engine types for the passingaircraft have been determined from aircraft registration numbers.Supporting information 2 provides the engine specific values of EInreported by Herndon et al. (2008) and Lobo et al. (2012a), and thelink to the engine types used at CPH. For turboprops, approximateEIn values are derived as average EIn for all jet engines at CPH.

The PN emissions at the inner apron area and for total CPH,respectively, become 4.62Eþ 19 and 3.06Eþ 21 based on EIn inputdata from Herndon et al. (2008), and 6.62Eþ 20 and 4.43Eþ 22based on EIn input data from Lobo et al. (2012a). Hence, byswitching from “low” to “high” EIn input data the estimated PNemissions for the inner apron area and for total CPH in both casesincrease with approximately a factor of 14. The possible reasons forthe differences in the measured EIn between studies are given in

Fig. 8. Percentage shares of PN from aircraft main engines, APU's and handling equipment at(Atlanta) and “high” (Oakland) EIn's.

the conclusion part of this paper. Based on the “average” EIn inputdata selected in this study (section 2.2), the PN emissions become2.77Eþ 20 and 2.12Eþ 22, respectively for the inner apron area andfor total CPH (Table 6).

At the inner apron, for main engines and APU the average [low,high] PN percentage shares become 43% [11%, 63%] and 54% [85%,35%], respectively, and handling equipment shares are below 4% inall three cases. For total CPH, the average [low, high] PN percentageshares become 95% [73%, 98%] and 5% [26%, 2%], respectively, formain engines and APU, and handling equipment shares neverexceed 1% (Fig. 8).

3.4. Comparison with other airport results

Only sparse emission inventory information is available fromother airports for comparison with our results obtained for CPH.Table 7 shows the emission share of main engines, APU's andhandling equipment published for other airports.

Main engines are the predominant source of NOx (around 90%emission share for all airports), and PM emission shares vary be-tween 55% (London Heathrow) and 75% (San Diego Airport). APUNOx emission shares lie between 2% (Zürich Airport) and 7% (Lon-don Heathrow) and PM emission shares are between 6% (ZürichAirport) and 19% (San Diego Airport). For main engines and APU,our NOx and PM results for CPH are inside the interval of derivedairport percentage shares.

For handling equipment our NOx share of 9% is somewhat higherthan the around 5% reported for the other airports, whereas our PMshare (31%) is close to the PM share calculated for London Heath-row (33%), and higher than the PM shares derived from San DiegoAirport (6%) and Zürich Airport (20%) results.

For PN, no other complete inventory results could be identifiedin the literature. However, a PN estimate for large aircraft mainengines only is available for Brisbane Airport (Mazaheri et al., 2011).PN data from this study is included in our literature based PN EIn'sas discussed in Section 2.2. Here it should only be noted thataverage idle and take off PN EIn's from Mazaheri et al. (2011)

the inner apron area and as total for CPH, based on average EIn's (this study), and “low”

Table 7Percentage share of NOx, PM and PN for main engines, APU's and handling equip-ment published for other airports.

Airport Source NOx PM PN Reference

Copenhagen Aircraft ME 87 61 94.8APU 4 8 5.0Handling 9 31 0.2

LondonHeathrow

Aircraft ME 87 55 e HeathrowAirport (2011)

APU 7 12 e

Handling 6 33 e

Brisbane Aircraft ME 91 e e BrisbaneAirport (2007)

APU 9 e e

Handling e e e

San Diego Aircraft ME 92 75 e San DiegoAirport (2009)

APU 3 19 e

Handling 5 6 e

Zürich Aircraft ME 93 74 e Zürich Airport(2014)

APU 2 6 e

Handling 5 20 e

3 Measurement value of 5.2Eþ 13 #/km; rough conversion into 6.2Eþ 14 #/kg,by using fuel density ¼ 0.84 kg/litre diesel, and fuel economy ¼ 0.1 L/km.

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229228

derived from Table 4 are around 15% and 300%, respectively, of ourcalculated average PN EIn's.

Generally the differences in airport emission shares are theresult of e.g. the individual inventory methods being used, theemission sources included, the difference in fleet-activity data forthe sources included and the geographical configuration of theairports. Due to lack of detailed information available from thesestudies, the reasons for the differences between studies are notfurther investigated in this paper.

4. Conclusions

This paper presents a detailed emission inventory for NOx andparticles (mass and numbers) for Copenhagen Airport (CPH) basedon detailed activity data for aircraft main engines, APU's andhandling equipment and time durations of their use. Total resultsare shown for CPH and for a section of the airport apron area. Thetemporally and spatially very detailed inventory for the airport andthe inner apron area, and the derived source contributions for mainengines, APU's and handling equipment, serve as a valuable newcontribution to the research field. The precise emission estimatesfor handling equipment in the airport based on equipment type andtechnology specific activity data also bring important new knowl-edge to the research community.

The emission results are spatially distributed in a high resolu-tion 5 m � 5 m grid suited for further dispersion modelling of airquality. Sensitivity calculations are made for CPH in two cases inorder to examine the PM emission impact of using zero sulphur jetfuel, and to assess the impact of PN emissions calculated by usingsets of “low” and “high” EIn from the literature.

As totals for CPH, aircraft main engines have the largest shares(percentage values in brackets) of fuel consumption (93%), NOx

(87%), PM (61%) and PN (95%). For APU and handling equipment,the fuel consumption [NOx, PM, PN] shares are 5% [4%, 8%, 5%] and 2%[9%, 31%, 0%], respectively. At the inner apron area, handlingequipment has fuel consumption [NOx, PM, PN] shares of 24% [63%,75%, 2.4%], whereas APU and main engines shares are 43% [25%, 19%,54%], and 33% [11%, 6%, 43%], respectively.

Some uncertainty exists in the calculated particle numbers formain engines and APU in general. This is due to the large variationin the EIn values reported by different studies as explained belowand for APU especially due to the low number of available data. Thesensitivity calculations for CPH made in this study showed that byswitching between “low” and “high” EIn factor sets reported in the

literature (Herndon et al., 2008; Lobo et al., 2012a), the calculatedPN emissions for aircraft main engines changed with a factor of 14.

It is difficult to make a precise quantification of the emittedparticle numbers based on measurements, due to the physical andchemical conversion that takes places in the exhaust plume andrapidly change the number of particles as the exhaust plumegradually disperses and cools down (e.g. Lobo et al., 2012a; Timkoet al., 2013). For comparison there is essentially no change in par-ticle mass during the dispersion and cooling of the exhaust plume(Wey et al., 2006, 2007). Other factors that influence the EIn valuesis the distance of the measurement equipment to the passingaircraft, exhaust plume evolution, meteorological conditions (winddirection and speed, ambient air temperature and humidity) andthe sulphur content of the jet fuel as outlined in Section 3.3.

For diesel fuelled handling equipment, the selected EIn values of3.1Eþ 15/kg fuel is low compared to the EIn's for aircraft mainengines (3.91Eþ 16/kg fuel) and APU (3.96Eþ 16/kg fuel) used inthis study. The low EIn values for handling equipment is supportedby a measured factor of 3.3Eþ 15/kg fuel by Ban-Weiss et al. (2010),and even values of 6.2Eþ 14 #/kg fuel3 for diesel vehicles measuredby Geller et al. (2005). The low EIn levels are also confirmed bymeasurements in Santa Monica Airport (Hu et al., 2009) andmeasurements made by USEPA on a diesel truck (Kinsey et al.,2009). Further ACI (2012) explains that particles emitted fromaircraft engines tend to be smaller in size but higher in numberscompared to those emitted from diesel engines.

Due to the still somewhat uncertain EIn levels for aircraft mainengines and APU, the relative particle number importance of thesesources might change at CPH if new PN emission information be-comes available. In order to increase the precision of the PN in-ventory for CPH, in situ measurements of EInwould be very useful.However, even with updated EIn values from the literature orspecific measurements for CPH, it is expected, that still handlingequipment only plays a minor role in relation to total particlenumber emissions in the airport.

The combustion of sulphur containing fuels leads to the for-mation of sulphate bound particles. The sensitivity analysis towardsfuel sulphur content in this studywas based on PM emission factorsestimated with the ICAO validated FOA3.0 method and suggeststhat more than half of the PM emissions from aircraft main enginesoriginates from the sulphur in the fuel. This emphasizes theimportance of using local information on jet fuel sulphur content,alongside with precise aircraft fleet/engine activity data, in order tomake PM emission estimates for an airport as precise as possiblebased on the FOA3.0 method.

Due to a similar sulphur dependency, jet fuel sulphur reductionis also expected to reduce PM emissions from APU's (Lobo et al.,2012b) and as well the emitted number of particles from aircraftmain engines and APU's (see e.g. Lobo et al., 2012b; Timko et al.,2013).

The methodology presented in this paper can be used togenerate spatial and temporally highly resolved emission in-ventories for aircraft main engines, APU and handling equipment inother airports and serve as an input for subsequent air qualitydispersion modelling.

Acknowledgements

The work was funded by Copenhagen Airport and the authorsacknowledge their support and provision of background data ma-terial. John Kinsey, USEPA North Carolina, Prem Lobo, Missouri

M. Winther et al. / Atmospheric Environment 100 (2015) 218e229 229

University of Science and Technology, RichardMiake-Lye, AerodyneResearch, Ulrich Janicke, Janicke Consulting and Emanuel Fleuti,Zürich Airport must be thanked for technical discussions in relationto particle measurements and for providing data used for the par-ticle emission calculations in the project.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.atmosenv.2014.10.045.

References

ACI Europe, 2012. Ultrafine Particles at Airports e Discussion and Assessment ofUltrafine Particles (UFP) in Aviation and at Airports in 2012. ACI EUROPEEnvironmental Strategy Committee, Autumn 2012, p. 54.

Ban-Weiss, G.A., Lunden, M.M., Kirchstetter, T.W., Harley, R.A., 2010. Size-resolvedparticle number and volume emission factors for on-road gasoline and dieselmotor vehicles. Aerosol Sci. 41, 5e12.

Brisbane, 2007. The Environmental Impact Statement (EIS) and Major DevelopmentPlan (MDP), Chapter D6: Air Emissions. Brisbane Airport Corporation, p. 108.http://www.bne.com.au/corporate/upgrading-your-airport/new-parallel-runway/eismdp.

Carslaw, D.C., Beevers, S.D., Ropkins, K., Bell, M.C., 2006. Detecting and quantifyingaircraft and other on-airport contributions to ambient nitrogen oxides in thevicinity of a large international airport. Atmos. Environ. 40, 5424e5434.

CPH Local Regulations. Chapter 9. Available at: http://www.cph.dk/en/about-cph/b2b/adgang-og-fardsel/local-regulations/.

Diapouli, E., Chaloulakou, A., Spyrellis, N., 2007. Levels of ultrafine particles indifferent microenvironments e implications to children exposure. Sci. TotalEnviron. 388, 128e136.

Ellermann, T., Massling, A., Løfstrøm, P., Winther, M., Nøjgaard, J.K., Ketzel, M., 2011.Investigation of Air Pollution at the Apron at Copenhagen Airport in Relation toWorking Environment (Danish with English summary). DCE e Danish Centrefor Environment and Energy, Aarhus University, p. 148. DCE report no. 5. http://www.dmu.dk/Pub/TR5.pdf.

EMEP/EEA, 2013. Air Pollutant Emission Inventory Guidebook, Prepared by theUNECE/EMEP Task Force on Emissions Inventories and Projections (TFEIP).Available at: http://www.eea.europa.eu//publications/emep-eea-guidebook-2013 (02.09.13.).

Fenger, J., Løfstrøm, P., Winther, M., Kousgaard, U., Oxbøl, A., 2006. Odour in thesurroundings of Copenhagen Airport. Atmos. Environ. 40, 368e374.

Geller, M.D., Sardar, S.B., Phuleria, H., Fine, P.M., Sioutas, C., 2005. Measurements ofparticle number and mass concentrations and size distributions in a tunnelenvironment. Environ. Sci. Technol. 39, 8653e8663.

Heathrow Airport, 2011. Heathrow Air Quality Strategy 2011e2020, p. 58. http://www.heathrowairport.com/static/Heathrow/Downloads/PDF/air-quality-strategy_LHR.pdf.

Herndon, S.C., Onasch, T.B., Frank, B.P., Marr, L.C., Jayne, J.T., Canagaratna, M.R.,Grygas, J., Lanni, T., Anderson, B.E., Worsnop, D.R., Miake-Lye, R.C., 2005. Par-ticulate emissions from in-use commercial aircraft. Aerosol Sci. Technol. 39,799e809.

Herndon, S.C., Jayne, J.T., Lobo, P., Onasch, T.B., Fleming, G., Hagen, D.E.,Whitefield, P.D., Miake-Lye, R.C., 2008. Commercial aircraft engine emissionscharacterization of inuse aircraft at Hartsfield-Jackson Atlanta InternationalAirport. Environ. Sci. Technol. 42, 1877e1883.

Hertel, O., Jensen, S.S., Ketzel, M., Becker, T., Peel, R.G., Ørby, P.V., Skjøth, C.A.,Ellermann, T., Raaschou-Nielsen, O., Sørensen, M., Br€auner, E., Andersen, Z.J.,Loft, S., Schlünssen, V., Bønløkke, J.H., Sigsgaard, T., 2013. Utilizing monitoringdata and spatial analysis Tools for exposure assessmentof atmospheric pollutantsin Denmark. In: McConnell, L.L., Dachs, J., Hapeman, C.J. (Eds.), Occurrence, Fateand Impact of Atmospheric Pollutants on Environmental andHumanHealth, ACSSymposium Series, vol. 1149. American Chemical Society, pp. 95e122.

Hsu, H.-H., Adamkiewicz, G., Houseman, E.A., Vallarino, J., Melly, S.J., Wayson, R.L.,Spengler, J.D., Levy, J.I., 2012. The relationship between aviation activities andultrafine particulate matter concentrations near a mid-sized airport. Atmos.Environ. 50, 328e337.

Hu, S., Fruin, S., Kozawa, K., Mara, S., Winer, A.M., Paulson, S.E., 2009. Aircraftemission impacts in a neighborhood adjacent to a general aviation airport inSouthern California. Environ. Sci. Technol. 43, 8039e8045.

ICAO, 2011. Airport Air Quality Manual (doc. 9889), first ed. International CivilAviation Organization. ISBN 978-92-9231-862-8.

Johnson, G.R., Mazaheri, M., Ristovski, Z.D., Morawska, L., 2008. A plume capturetechnique for the remote characterization of aircraft engine emissions. Environ.Sci. Technol. 42, 4850e4856.

Kinsey, J.S., Dong, Y., Williams, C., Logan, R., August 2009. Evaluation of the Emis-sions from Low-sulfur and Biodiesel Fuel Used in a Heavy-duty Diesel Truckduring On-road Operation e Final Report. U.S. Environmental ProtectionAgency, Washington, DC. Contract No. EC-C-09e027, Work Assignment No. 0-5,EPA Project.

Kinsey, J.S., Dong, Y., Williams, D.C., Logan, R., 2010. Physical characterization ofthe fine particle emissions from commercial aircraft engines during theAircraft Particle Emissions eXperiment (APEX) 1-3. Atmos. Environ. 44,2147e2156.

Lobo, P., Hagen, D.E., Whitefield, P.D., 2012a. Measurement and analysis of aircraftengine PM emissions downwind of an active runway at the Oakland Interna-tional Airport. Atmos. Environ. 61, 114e123.

Lobo, P., Rye, L., Williams, P.I., Christie, S., Uryga-Bugajska, I., Wilson, C.W.,Hagen, D.E., Whitefield, P.D., Blakey, S., Coe, H., Raper, D., Pourkashanian, M.,2012b. Impact of alternative fuels on emissions characteristics of a gas turbineengine � part 1: gaseous and particulate matter emissions. Environ. Sci.Technol. 46, 10805e10811.

Mazaheri, M., Johnson, G.R., Morawska, L., 2009. Particle and gaseous emissionsfrom commercial aircraft at each stage of the landing and takeoff cycle. Environ.Sci. Technol. 43, 441e446.

Mazaheri, M., Johnson, G.R., Morawska, L., 2011. An inventory of particle andgaseous emissions from large aircraft thrust engine operations at an airport.Atmos. Environ. 45, 3500e3507.

Oberd€orster, G., Sharp, Z., Atudorei, V., Elder, A., Gelein, R., Kreyling, W., Cox, C.,2004. Translocation of inhaled ultrafine particles to the brain. Inhal. Toxicol. 16,437e445.

Pope, C.A., Dockery, D.W., 2006. Health effects of fine particulate air pollution: linesthat connect. J. Air & Waste Manag. Assoc. 56, 709e742.

Raaschou-Nielsen, O., Andersen, Z.J., Jensen, S.S., Ketzel, M., Sørensen, M., Hansen, J.,Loft, S., Tjønneland, A., Overvad, K., 2012. Traffic air pollution and mortalityfrom cardiovascular disease and all causes: a Danish cohort study. Environ.Health A Glob. Access Sci. Source 11 (1), 60.

Rindlisbacher, T., 2009. Guidance on the Determination of Helicopter Emissions,first ed. FOCA, Bern. Reference: 0/3/33/33-05-20.

San Diego Airport, 2009. San Diego County Regional Airport Authority Air QualityManagement Plan, Final Draft. San Diego County Regional Airport Authority,p. 36. http://www.san.org/sdcraa/airport_initiatives/environmental/air_quality.aspx.

Schürmann, G., Sch€afer, K., Jahn, C., Hoffmann, H., Bauerfeind, M., Fleuti, E.,Rappengluck, B., 2007. The impact of NOx, CO and VOC emissions on the airquality of Zurich airport. Atmos. Environ. 41, 103e118.

Schwarze, P.E., Ovrevik, J., Lag, M., Refsnes, M., Nafstad, P., Hetland, R.B.,Dybing, E., 2006. Particulate matter properties and health effects: consis-tency of epidemiological and toxicological studies. Hum. Exp. Toxicol. 25,559e579.

Timko, M.T., Fortner, E., Franklin, J., Yu, Z., Wong, H.-W., Onasch, T.B., Miake-Lye, R.C.,Herndon, S.C., 2013. Atmospheric measurements of the physical evolution ofaircraft exhaust plumes. Environ. Sci. Technol. 47, 3513e3520.

Westerdahl, D., Fruin, S.A., Fine, P.L., Sioutas, C., 2008. The Los Angeles InternationalAirport as a source of ultrafine particles and other pollutants to nearby com-munities. Atmos. Environ. 42, 3143e3155.

Wey, C.C., Anderson, B.E., Hudgins, C., Wey, C., Li-Jones, X., Winstead, E.,Thornhill, L.K., Lobo, P., Hagen, D., Whitefield, P., Yelvington, P.E., Herndon, S.C.,Onasch, T.B., Miake-Lye, R.C., Wormhoudt, J., Knighton, W.B., Howard, R.,Bryant, D., Corporan, E., Moses, C., Holve, D., Dodds, W., 2006. Aircraft ParticleEmissions Experiment (APEX). NASA/TM-2006e214382, ARL-TR-3903. U.S. Na-tional Aeronautics and Space Administration, Glenn Research Center, Cleveland,OH. Available at: http://gltrs.grc.nasa.gov.

Wey, C.C., Anderson, B.E., Wey, C., Miake-Lye, R.C., Whitefield, P., Howard, R., 2007.Overview on the aircraft particle emissions Experiment. J. Propuls. Power 23,898e905.

Winther, M., Kousgaard, U., Oxbøl, A., 2006. Calculation of odour emissions fromaircraft engines at Copenhagen Airport. Sci. Total Environ. 366, 218e232.

Winther, M., Ellermann, T., Massling, A., Løfstrøm, P., Nøjgaard, J.K., Ketzel, M.,Kousgaard, U., 2014. Inventory of Emissions and Particle Numbers at the Apronat Copenhagen Airport (Danish with English summary). DCE - Danish Centre forEnvironment and Energy, Aarhus University, p. 84. DCE report no. 21. http://www2.dmu.dk/Pub/TR21.pdf.

Wong, H.-W., Yelvington, P.E., Timko, M.T., Onasch, T.B., Miake-Lye, R.C., Zhang, J.,Waitz, I.A., 2008. Microphysical modeling of ground-level aircraft-emittedaerosol formation: roles of sulfur-containing species. J. Propuls. Power 24,590e602.

Zhu, Y., Fanning, E., Yu, R.C., Zhang, Q., Froines, J.R., 2011. Aircraft emissions and localair quality impacts from takeoff activities at a large International Airport.Atmos. Environ. 45, 6526e6533.

Zürich Airport, 2014. Personal Communication with Emanuel Fleuti. Head ofEnvironmental Protection Services, Flughafen Zürich AG.