EUROCONTROL Experimental Centre

Transcription

1 EUROCONTROL Experimental Centre Ayce Celikel Nicolas Duchene Emanuel Fleuti Ian Fuller Peter Hofmann Tamara Moore Marcel Silue EUROCONTROL EUROCONTROL Launch Microsoft Office Outlook.lnk Airport Local Air Quality Studies Case Study: Emission Inventory for Zurich Airport with different methodologies EEC/SEE/2004/010 EUROCONTROL

2 Emission Inventory for Zurich Airport with different methodologies Reference: EEC/SEE/2004/010 Nicolas Duchene, Ayce Celikel (ENVISA) Ian Fuller (EUROCONTROL Experimental Centre) Marcel Silue (Brunel University) Emanuel Fleuti, Peter Hofmann (Unique) Tamara Moore (FPIC) ENVISA 38 Rue Des Gravilliers Paris FRANCE UNIQUE (Flughafen Zurich AG) P.O. Box, 8058 Zurich-Airport SWITZERLAND FISCHER & PARTNER International Consultants Schulhausstrasse Hergiswil SWITZERLAND European Organisation for the Safety of Air Navigation EUROCONTROL July 2002 This document is published by EUROCONTROL in the interest of the exchange of information. It may be copied in whole or in part providing that the copyright notice and disclaimer are included. The information contained in this document may not be modified without prior written permission from EUROCONTROL. EUROCONTROL makes no warranty, either implied or express, for the information contained in this document, neither does it assume any legal liability or responsibility for the accuracy, completeness or usefulness of this information.

3 Reference: EEC/SEE/2004/010 Originator: REPORT DOCUMENTATION PAGE Security Classification: Unclassified Originator (Corporate Author) Name/Location: Environment Studies Business Area Sponsor: EUROCONTROL Experimental Centre Centre de Bois des Bordes B.P BRETIGNY SUR ORGE CEDEX France Telephone: Sponsor (Contract Authority) Name/Location: EUROCONTROL EATMP EUROCONTROL Agency Rue de la Fusée, 96 B 1130 BRUXELLES Telephone: Airport Local Air Quality Studies: Case Study: Emission Inventory for Zurich Airport with different methodologies: Authors : Date Pages Figures Tables Appendix References Ayce Celikel Nicolas Duchene Ian Fuller Marcel Silue Emanuel Fleuti Peter Hofmann Tamara Moore EATMP Task Specification - 12/04 Project ALAQS Task No. Sponsor - 7 Period Distribution Statement: (a) Controlled by: (b) Special Limitations: (c) Copy to NTIS: EUROCONTROL Project Manager None YES / NO Descriptors (keywords): Airport ALAQS LASPORT EDMS Emission Inventory Calculation Zurich airport. Abstract: EEC ALAQS Project aims to address strategic, methodological and practical issues surrounding air quality assessment around airports. The project promotes best practice methodology and a supporting toolset that can be applied at Pan-European level. This report discusses the results of the case study of emission inventory for Zurich Airport using three different methodologies; ALAQS, LASPORT and EDMS.

4 EXECUTIVE SUMMARY This report documents the results of a study that compared three Airport Local Air Quality inventory methods using comprehensive airport emissions data from Zurich airport. The models were ALAQS- AV-1, LASPORT and EDMS 4.12 (Emission and Dispersion Modelling Software). The study was one of a series of studies for the EUROCONTROL Experimental Centre ALAQS (Airport Local Air Quality Studies) project. ALAQS aims to promote best practice methods in airport air quality modelling using the GIS based ALAQS-AV tool as a test platform. Since ALAQS-AV is has its origins in a variety of established methodologies (EDMS, LASPORT, COPERT-III) it was the main objective of this study was to compare ALAQS-AV to its peers and investigate the sensitivity of the different input parameters. The scope of the study was to calculate emissions inventories using the candidate models on an asis basis, i.e. as a normal user but at the same time using as much possible the comprehensive empirical data available from the Zurich airport operator. Thus the objective was to reduce the uncertainties that come for using built-in default values except where it was considered that these were more appropriate. The comparison study could only be achieved by rationalising the input data to the three models. A separate section sensitivity analysis is also included that shows the impact of varying the parameters for aircraft in ALAQS-AV from the default set and those used for the comparison. This approach allowed us not only to compare the different methods but to start the process of validating the built-in default values against measured values. The emissions inventory comparison used LASPORT as a reference and showed that, taking into account the differences between the models, their input data requirements and their parameters, the results were comparable for CO, HC and NO x for aircraft, stationary sources and APU/GPU. However, vehicle emissions (landside, GSE) showed a large variation. Considering aircraft emissions, model parameters showed some variation such as aircraft grouping, aircraft profiles etc. As the ideal for the comparison between the models is to use similar parameters; every effort was made to use the same emission factors and where possible aircraft grouping and profiles. From several iterative simulations it was concluded that ALAQS and LASPORT results differences mainly originated from the usage of different aircraft take off profiles and methods, different engine fits for aircraft, different grouping of aircraft. Techniques such as Aircraft grouping with one take-off profile per group are useful for making life easier for the user but internally the calculations should be made on a flight-by-flight, movement-by-movement basis. The practice is well established in the noise modelling community ENHANCE/INM and can be easily employed to good benefit in the air quality modelling world. Vehicle emissions showed a large variation. The EDMS results were a factor of 10 greater than ALAQS or LASPORT for CO and a factor 45% more for NOx. The differences were due to a combination of methodology, emission indices and data assumptions. For example, EDMS is based on round trip journeys where there was no discrimination between type of vehicle. EDMS was preconfigured for US airports and therefore some data are not representative in the European context, e.g. vehicle fleet mix. The present study highlighted the importance of reliable and realistic parameters (especially aircraft engine emission indices and climb out profiles) to provide accurate figures of the pollution around airports. With the exception of the vehicle emissions all the models have similar global results. However, the 4D distribution of emissions requires more robust data and realism. The results of this study should be useful to the airport emissions modelling community in striving for more realism when faced assessing the impact of new projects such as runway extensions and reduced thrust take-offs. 4

5 TABLE OF CONTENTS 1 INTRODUCTION Background Objectives of the study Choice of airport Emission calculation models used ZURICH AIRPORT DATA Airport graphical data Airport operation data Data preparation COMPARISON OF THE PARAMETERS AND ASSUMPTIONS OF THE THREE MODELS Airport geographical display Operational Input data requirements comparison Models assumptions comparison Pollutants considered EMISSION INVENTORY RESULTS ALAQS Emission Inventory LASPORT Emission Inventory EDMS Emission Inventory INTERPRETATION OF THE RESULTS Pollutant emission comparison Aircraft emissions Stationary sources emissions comparison APU and GPU emissions comparison GSE emissions comparison Road emissions comparison Special considerations SENSITIVITY ANALYSIS CONCLUSIONS CONSIDERATIONS FOR THE FUTURE Next Step ALAQS Improvements REFERENCES

6 LIST OF FIGURES Figure 2-1: Zurich airport background map Figure 2-2: Zurich airport orthophoto Figure 3-1: Modelling of Zurich airport in ALAQS Figure 3-2: Modelling of Zurich Airport in LASPORT...15 Figure 3-3: Airport layout in EDMS Figure 3-4: Default aircraft profiles in LASPORT Figure 5-1: Comparison of airport emissions (without road emissions) Figure 5-2: Aircraft emissions comparison Figure 5-3: Zurich traffic 2003 repartition in groups Figure 5-4: Comparison of ALAQS parameters for Small and Business jets (from top left to bottom right: fuel flow in kg/s, CO then NO x and HC in g/kg fuel burnt) Figure 6-1: Overview of the results using various parameters LIST OF TABLES Table 2-1: Zurich data for Emission Inventory Calculation Table 3-1: Operational data requirements for emission inventory programs Table 3-2: EDMS profiles (times in mode) Table 3-3: EDMS parameters per aircraft group...23 Table 4-1: ALAQS Emission Inventory for Zurich Table 4-2: LASPORT (1.3) Emission Inventory for Zurich 2003: Table 4-3: EDMS (4.12) Emission Inventory for Zurich Table 5-1: CO emissions comparison Table 5-2: HC emissions comparison Table 5-3: NOx emissions comparison Table 5-4: Aircraft emissions comparison Table 5-5: Averaged times in mode applied by the models Table 5-6: Comparison of aircraft groups Table 5-7: Stationary sources emissions comparison Table 5-8: APU and GPU emissions comparison Table 5-9: APU and GPU methodology comparison Table 5-10: GSE emissions comparison Table 5-11: Belt Loader Emission Factors Table 5-12: Road emissions comparison Table 5-13: Road emission methodology comparison EDMS-LASPORT

7 1 Introduction This study was one of a series of studies for the EUROCONTROL Experimental Centre ALAQS (Airport Local Air Quality Studies) project. This report describes the results of the latest study in the series based on Zurich airport and comparing three emissions inventory models: ALAQS-AV, EDMS and LASPORT. Airport operations are part of the total Air Traffic Management (ATM) system. One cannot grow without affecting the other. With the current growth of air transportation there are several constraints on the capacity and environment is one of the important ones. Airport activities affect the environment in several ways, such as noise, emissions and local air quality. In the ALAQS project the main focus is air quality around airports and the ways to measure it. Airport stakeholders, in order to estimate the extent to which they contribute to local air pollution, need to conduct air quality studies. To calculate airport pollution, as a first step, all airport related activities that produce emissions should be defined. A reliable emission inventory method should be chosen and necessary data collection for each source should be carried out. The result of the emission inventory will give a global value (i.e. total NOx per year) such as the quantity of the emissions produced from airport operations. However as the public interest increased regarding the health effect of air pollution originating from airport operations, it becomes important to use more realistic inventory and dispersion models that provide input to the city models and to calculate the pollutants concentrations. In order to conduct dispersion modelling, detailed emission inventories should be done. A detailed emission inventory provides information on the spatial distribution and total amounts of pollutants from major pollution sources within a certain geographical area. However, there are few airports operators, that conduct detailed 4D Emissions Inventories, which can be used for dispersion modelling. One of the reasons is that there is no harmonised methodology to conduct such inventory for all airport sources and more time and man resources are needed for data collection. Once the emission inventory is done for an airport then, depending on the scale of the emission inventory and available resources, dispersion modelling can be carried out to estimate the pollutant concentrations plume. The ALAQS project addresses the harmonisation issue by collecting best practice methodologies in one GIS based test platform ALAQS-AV; to promote best practice methods in airport air quality modelling. 1.1 Background With the expansion of Europe and the growing importance of environmental issues, there is a need for harmonisation. For air quality issues, this means the development of tool that will allow conducting emission inventories and dispersion modelling. The emission inventory tools analysed in this report concern in particular airports. They offer a thorough understanding of emissions exhausted at one airport in terms of scope and quantity. The sources considered for the simulations gather all the features that are responsible for pollutants exhaustion in the following areas of one airport: air traffic, handling, infrastructures and landside traffic. Depending on the level of accuracy to reach (and on the availability of data), the appropriate methodology to perform the study varies. The ALAQS project aims to promote best practice in airport emissions modelling. To that end a series of studies is underway to validate the methodologies incorporated into ALAQS-AV by demonstrating the characteristics of the different methodologies and comparing results. The ALAQS (Airport Local Air Quality Studies) project allowed the aggregation of some well known European and international models/default databases (LASPORT, EDMS) in order to (i) develop a test bed tool (ALAQS-AV) and (ii) address best practice recommendations for assessing airport local air quality. An additional objective of the ALAQS project was the implementation of the format of the output emission inventory so that various dispersion models could be applied. Since ALAQS-AV is has its origins in a variety of established methodologies (EDMS, LASPORT, COPPERT-III) it was the main objective of this study was to compare ALAQS-AV to its peers and investigate the sensitivity to the different input parameters. 7

8 It is critical to bear in mind that all airport emissions sources should be accounted for the modelling, even though some assumptions have to be made. That is to ensure that the inventory provides a complete picture of the pollution surrounding the airport. As part of the air quality modelling exercise, emission inventories should be provided in 4D with thorough spatial and temporal data for all sources in order to accurately model the dispersion of emissions. Finally, such airport models can be used to test future scenarios of emission and therefore are important tools for decision-making when dealing with environmental concerns. It is recognised that more realism would allow interested parties to identify the advantages of new projects such as runway extensions, where changes in aircraft operation may incur changes in emissions, e.g. reduced thrust take-offs. 1.2 Objectives of the study The objectives of the study were to: 1. Investigate compatibility of the models with current airport practice, aircraft types and ground procedures. 2. Investigate to what extent the default parameters of the candidate models could be used given that the study airport had extensive data for aircraft movements, stationary sources, airside and landside vehicles. 3. Compare emissions inventory results between the three models for CO, NOx and HC. As CO 2 is not a major airport air pollution concern, calculations of CO 2 emissions are currently not conducted in ALAQS. 4. Use the emission inventory outputs for dispersion modelling (To be reported separately) The aim of this case study is to get a better understanding of methodological and practical issues about the use of emission inventories for airports. A follow-up study will report on a dispersion modelling with different emission inventories. This report, part of the EUROCONTROL Experimental Centre ALAQS project, compares three emission inventory tools (ALAQS, LASPORT, EDMS) with the aim of identifying potential improvement areas in the ALAQS methodology and resolve them. A secondary objective is to provide guidelines towards their use. Within the scope of the Airport Local Air Quality work, Zurich airport was chosen as a case study to perform assessments of gaseous emissions. The present study is expected to further validate the ALAQS emission inventory (and in a smaller extent LASPORT and EDMS ones), which is essential as those programs constitute the basis for dispersion modelling applications. ALAQS emission inventory is conducted using the same input parameters as LASPORT. It is demonstrated in the sensitivity analysis (See Section 6) by changing the ALAQS default emission factors to the LASPORT ones. The objective of the sensitivity analysis is to highlight differences caused by the methodologies such as aircraft grouping, gate scenarios, and roadways methods. 1.3 Choice of airport An emission inventory using ALAQS and LASPORT was conducted in 2003 for Lyon airport over the year 2000 [ref 2]. One of the conclusions from that study highlighted the importance of the input data quality and extent: it has to be exhaustive enough to provide reliable results in output. The differences experienced in that previous report were mainly resulting from the lack of operational data and the differences in default parameters, especially climb out profiles and engine emission indices. Thus the present study required an airport that could provide detailed operational data for the different sources. Zurich airport, which is managed by Unique, was chosen for that study because of: Unique experience in air quality modelling and benefits for goal completion with airport stakeholders Availability of comprehensive empirical data for its emission sources on the airport 8

9 Working with Zurich airport which has already experience in air quality modelling, will ensure a wider data collection so that an evaluation of the applicability of the methodologies is made possible. Through the use of real case study and actual airport data, it is expected to compare the three emission models and to identify potential differences in order to correct them. 1.4 Emission calculation models used Three models were used to calculate the emissions: ALAQS-AV (version 1) developed by the EUROCONTROL Experimental Centre (EEC) LASPORT (version 1.3.4) Emission Calculation and Dispersion Model (developed on behalf of the German Airport Association); version 1.3 EDMS (version 4.12) which is the FAA Emission and Dispersion Model The LASPORT Emission Calculation and Dispersion Model is developed on behalf of the ADV (German Airport Association). LASPORT is a program package for dispersion calculations in the context of air pollutant emissions from airports. In particular, LASPORT calculates emissions for aircraft (complete LTO cycles), auxiliary power unit (APU), ground power units (GPU), airside and landside vehicle traffic. EDMS was developed by the United States Federal Aviation Administration (FAA) in co-operation with the United States Air Force (USAF). EDMS is designed to assess the air quality impacts of airport emission sources, particularly aviation sources, which consist of aircraft, auxiliary power units, and ground support equipment. EDMS also offers a limited capability to model other airport emission sources that are not aircraft-specific, such as ground access vehicles and stationary sources. ALAQS-AV is a customised Geographical Information System (GIS) application for the capture of airport pollution sources and for the processing of the different types of emission sources into a standard format in preparation for dispersion modelling. The ALAQS-AV toolset forms a test-bed of the ALAQS project [Ref 1], it is used for evaluating different emissions and dispersion modelling methods. ALAQS-AV is implemented as a customised Arcview 8 application using Microsoft Visual Basic for Applications. The application is inspired by some of its concept and system data from EDMS (Emission and Dispersion Modelling System), LASPORT, COPERT-III models. ALAQS can import and export EDMS data. However, as in the EDMS some of the technical data and emission calculations methods are US based, they do not necessarily reflect the European situation. ALAQS differs from EDMS on the following points: 1. Aircraft type, LTO profiles and engine emission indices tables are European practice based (optimised on the long term) 2. ALAQS calculates Aircraft Emissions based on aircraft movements table. 3. In ALAQS movement table all the flights are assigned to departure or arrivals with the corresponding gates and runway configurations where in EDMS; movements are represented as a complete LTO cycle, therefore same configurations regardless departure or arrival flight is used. 4. Methodology and emission factors for GSE, APU, GPU. 5. ALAQS supports different methods to calculate road emissions. 6. ALAQS is currently not linked to a dispersion modelling application. ALAQS has inherited some concepts from LASPORT: 1. Aircraft grouping. 2. The aircraft categorisation is different (see Appendix F). 9

10 2 Zurich Airport Data 2.1 Airport graphical data The following background map of Zurich airport (Figure 2.1) was provided by Unique. This schematised map presents all the features of the airport (runways, gates, taxiways and so on). It allowed for a precise geographical localisation of the feature considered and the subsequent emissions. The size of the grid of the map below is 1000 meters. Several other maps were available, each of them detailing a particular group of objects. Figure 2-1: Zurich airport background map The orthophoto presented in Figure 2.2 is a very detailed representation of Zurich airport geometry with a high resolution. This map was imported in ALAQS and allowed the location of emission sources to a sufficient accuracy. The geographical extent of the study includes the entire airport area (represented by a red borderline on Figure 2.1) as well as roadways exclusively associated with the airport. 10

11 Although all the emission inventory models require geographical data, the level of accuracy required varies. LASPORT and EDMS require schematics of the airport rather than orthometric data. As ALAQS-AV is a GIS application the user has the opportunity to make use of ortho-photos (e.g. satellite images). A major implication is that different maps of the area (for example the population repartition) can be added to the background of the emission/dispersion maps contributing to a quick visual localisation of the emissions in reference with the geographical characteristics of the area considered. Figure 2-2: Zurich airport orthophoto 11

12 2.2 Airport operation data The ALAQS study emission inventory of Lyon Saint Exupery Airport [Ref 2] showed that a thorough data collection was essential to ensure a reliable study. It is recognized that many airports may have difficulty in obtaining the level of data available at Zurich. Data required by the programs presented in this report might not be straightforward to obtain from airports. However Zurich airport experience with emission inventories and dispersion modelling allowed comprehensive data to be available as shown in Table 2-1 below. Data for 2003, including all the aircraft movements, was provided by Unique from on site investigations. Where assumptions and default value were made these are discussed later. Table 2-1: Zurich data for Emission Inventory Calculation Source Group Emission Source Operational Data Emission Factor Data Spatial Information Temporal Information Air Traffic Aircraft (year 2003) Handling Aircraft Main Engine Ignition Aircraft APU GPU GSE Airside Traffic De-icing Refuelling Infrastructure Power Generation Plant (including emergency power station) Aircraft Maintenance Airport Maintenance Fuel farm Construction Other: Engine test run Landside Traffic Road traffic (+ parking lots) For the aircraft emissions calculations; additional computation was required so that the real engine/aircraft fits could be used. The results were enhanced by using real engines instead of the ALAQS default engines for the selected aircraft. (See different simulation results with ALAQS default engine in Appendix D). The de-icing category includes both corrective (arrivals) and preventive treatments (departures). It has been reported in previous studies, especially emission inventories of Lyon airport (EUROCONTROL, 2003) and of airport areas in the Region Isle de France (DRIRE, 2003), that emissions due to de-icing are negligible since they represent less than 0,007% of airports global emissions in terms of HC. Zurich airport fuel farm has fixed roof tanks with floating membranes. Emissions from such tanks are much less than the ones from fixed roof tanks. Consequently, fuel tank evaporation emissions were assumed insignificant. It is well worth noticing that the emergency power station emissions are accounted for in the power generation plant. 12

13 2.3 Data preparation Data provided for Zurich airport was detailed and exhaustive. Most of the emission indices, that are source-specific, were available inducing the use of only few default parameters. Thus the model was able to perform a simulation as close as possible to the actual situation of the airport, which ensured reliable results. The aircraft movement table respected the format established in the ALAQS requirements document therefore no additional treatment was needed. For the application of EDMS, the available data has been reformatted whenever necessary to meet the specific data input specifications. For EDMS it was necessary to pre-process the input data into compatible formats, eg. The movement table had to be pre-processed into an LTO cycle table. LASPORT did not require any pre-processing as Zurich have already used this tool in previous campaigns. ALAQS-AV also did not require any pre-processing. 13

14 3 Comparison of the parameters and assumptions of the three models 3.1 Airport geographical display ALAQS The first step in an ALAQS study is to capture all the relevant airport features into the GIS together with the precise location of the various emission sources. This was accomplished by importing the Zurich airport orthophoto (Figure 2.2) into ALAQS. The orthophoto had sufficient resolution to allow the capture of runways, taxiways, queues, aircraft stands, roadways, parking lots, and stationary sources. Once the airport map was available in the ALAQS application, features have been captured using the creating/editing emission sources menu. After drawing a source on the map, its specific characteristics were entered in attributes field to allow for projection in the ALAQS window. A table can be consulted in Annex A that summarises ALAQS requirements to locate objects. The final airport layout used in the ALAQS study is depicted below in figure 3.1 showing gates, taxiways, runways, queues, stationary sources, roadways and parking lots in ALAQS tool. Figure 3-1: Modelling of Zurich airport in ALAQS LASPORT According to the LASPORT programme package, the first step is in geographically setting up the airport, using an airport map and the necessary information on runways, position areas, taxiways, 14

15 departure routes, aircraft stands and roads. A table can be consulted in Annex A that summarises LASPORT requirements to locate objects. Figure 3-2: Modelling of Zurich Airport in LASPORT 15

16 3.1.3 EDMS In order to spatially locate and graphically depict Zurich Airport, the overall airport infrastructure had to be created in EDMS. Effort was focused on the essential objects:- active runways and their associated taxiways, aircraft gate positions, vehicle parking lots, roadways and stationary sources. For each of the active runways, Runway 10-28, Runway and Runway 16-34, the runway orientation along with x, y coordinates for each runway end was identified. The peak queue time for each runway were identified, as were queue hourly operational profiles. X, Y coordinates for queue ends at peak length paralleled the runway x, y coordinates, as entry of any other coordinates into this EDMS parameter impacted the runway orientation and the subsequent graphic depiction. Approximately 16 gate areas (> 50 individual gates) were identified and entered into the system. These positions accommodate parked, serviced and or handled aircraft, and for each gate positions the corresponding x, y coordinates and the position height were identified. X, Y coordinates and the subsequent facility heights were also identified for 6 structured and or surface parking lots serving the travelling public and airport employees, and 21 stationary sources. A table can be found in Annex A that summarises EDMS geographical requirements. Figure 3-3: Airport layout in EDMS Geographical requirements comparison Tables shown in Annex A summarise the spatial requirements for every emission inventory program. Examining those tables led to the observation that the main difference between LASPORT and 16

17 ALAQS concerns the departure taxiing route of aircrafts. ALAQS assumes that aircraft will leave from the beginning of the runway while in LASPORT the starting point can be specified anywhere on a runway. The starting point is defined as the point where starts the take-off run. In addition the departure taxiing route has to be specified for every flight in the case of a monitoring calculation in LASPORT and thus can vary from one aircraft to another. On the contrary this route is fixed in ALAQS depending on flight type (Departure or arrival), aircraft type and gate assigned. In EDMS as the flights are considered as total LTO cycles; a standard taxing/queuing time is assumed for emission inventory calculations. Taxiways were constructed in individual segments in EDMS and ALAQS. Since EDMS and ALAQS could not accommodate changes in taxiway direction, several taxiway segments were created to spatially locate and graphically depict the airport s taxiway systems. On the contrary, LASPORT copes with taxiway a direction change, which reduces greatly the amount of work necessary. EDMS required the drawing of 1212 taxiway segments compared with 228 in ALAQS and 204 in LASPORT for depicting Zurich airport. As with the taxiway systems, public access and airport perimeter roads were constructed in segments, as EDMS and ALAQS do not allow bi-directional roadway sections. In LASPORT a roadway can be set to two-way or one-way. For Zurich airport, EDMS required drawing of 166 roadways (both airside and landside) compared with 76 for ALAQS and 34 for LASPORT. The airport was drawn using the LASPORT tool as the number of segments required to depict roadways and taxiways was smaller than in ALAQS or EDMS. ALAQS required twice as much road segments as LASPORT (34 and 76 resp.). Meanwhile the largest number was experienced around 166 segments for EDMS and the differences were even higher in the case of taxiways. This was a consequence of LASPORT ability to (i) deal with two directions roads and (ii) allow the drawing of multi-segments roads. It was decided to implement those features in the next version of the ALAQS- AV tool as they were the best practise experienced. 3.2 Operational Input data requirements comparison The input data requirements for the three emission inventories differ from a program to another. ALAQS, LASPORT and EDMS require similar dataset although some of the assumptions can be different.. It was then decided to focus on the comparison of the methodologies in the body of the report. Thus the differences in requirements and in assumptions for the three programs will be shown on a case-by-case basis rather than an exhaustive list. However, a complete set of the data available and the related sources can be found in annex A. Table 3.1 summarises the input data requirements for the three programs. It allows for an in depth comparison of the methodologies through the analysis of the input required and the related assumptions. 17

18 Table 3-1: Operational data requirements for emission inventory programs Emission Source Data required by ALAQS LASPORT EDMS Aircraft (1) - Aircraft identification Y Y Y - Departure profile Arrival profile D or A for Arrival or Departure Y Actual time of arrival Y Y - - Actual time of departure Y Y - - On-block-time - Y - - Off-block-time - Y - - Queuing time - Y Y - Runway used Y Y (2) - Taxi route - Y (2) - Gate Y Y Y APU - Running time arrival Y Y Y - Running time departure Y Y Y GPU - Running time arrival Y Y - - Running time departure Y Y - Main engine start - No data required - - (3) Handling (GSE) - Population and operating time Y Y Y Vehicles airside - Activity profiles Y Y Y - Movements per road (and speed) Y Y Y - Scenario - Y - - Vehicle fleet mix Y Y (3) Vehicles landside - Activity profiles Y Y Y - Movements per road (and speed) Y Y Y - Scenario - Y - - Vehicle fleet mix Y Y (3) Other sources - Activity profiles and emission factors Y Y Y (1) excl. Helicopters (2) Yes but for dispersion modelling only (3) Not considered in the model From Table 3-1 it can be seen that the main difference between the input data requirements for the three methodologies relate to aircraft emissions. ALAQS requires the departure (or arrival) time without considering the period spent on taxiways or queues. Those are computed during the emission calculation. On the contrary, LASPORT needs to have information about both the time at which the aircraft leaves the stand and the actual time of departing (or arriving). Contrarily to ALAQS and LASPORT, which rely on movement table to perform aircraft emission calculation, EDMS is based on the count of the yearly number of LTO cycles. This data specification required additional preprocessing of the existing airport data. Another major difference between EDMS and LASPORT/ALAQS concerns the definition of the runway used by an aircraft (arriving or departing). EDMS inventory mode does not use runway (in dispersion mode it is used). Both ALAQS-AV and LASPORT require a runway linked to each movement to be specified in the movement tables to permit display of the geographical distribution of the emission inventory. In addition to that, EDMS does not consider emissions from main engine start. Considering the Ground System Equipment methodology, ALAQS and EDMS approach is based on an inventory of the GSE population accompanied with its operating time (which is gate specific and 18

19 aircraft group specific). On the contrary LASPORT GSE method relies on the movement table and a system of aircraft grouping. As far as Ground Power Unit emissions are concerned, EDMS uses fixed parameters. On the contrary the running times for both arrival and departure can be set up on an aircraft group basis in both ALAQS and LASPORT. Finally the LASPORT roadways method is the only one relying on the definition of scenarios (eight types). EDMS doesn't discriminate between vehicle types (that is the percentage of passenger cars, light duty vehicle and high duty vehicle) for a road segment and a identical vehicle fleet mix as well as a two way traffic is assumed for any road segment considered in the study. A general property of EDMS Version 4.12 is that the unit system is U.S. based. As an example, all emission results are displayed in short tons as opposed to metric tons, even if some input data is required in the metric system. Moreover some emission factors are required in unconventional units such as kiloliters. 3.3 Models assumptions comparison ALAQS assumption details Default departure and arrival profiles Arrival and departures profiles are important parameters to calculate aircraft emissions. Take-off, climb-out and arrival emissions are calculated with those procedures Initially ALAQS as a default used EDMS aircraft profiles originating from EPA. During the validation process of the study, the aircraft profiles were reiterated in order to obtain a reliable representation. Various types of climb out profiles have been used through the study. Profiles derived from EPA times in mode were replaced with the ones from the INM 6.1 (Integrated Noise Model) software, which were more realistic. Additionally those profiles have already been validated and are used worldwide. Results from those interim simulations can be found in annex D. Finally, in order to validate the ALAQS method, ALAQS profiles have been replaced with the LASPORT ones that are created for Zurich Airport. The aim of such simulation was to obtain results as close as possible with different programs, providing thus a reliable validation of the methods. For the arrival profile the 3-degree approach profile was selected in order to be consistent with LASPORT and EDMS. Runway procedures Runway exits from ALAQS methodology using INM landing roll and touchdown offset is specified in the runway parameters and it is assumed that aircraft exit runways as soon as possible. Departing aircraft always take off from the beginning of the runway. From Zurich airport operational data the hourly number of departures per runway is 7,3; this value was rounded since this parameter must be an integer in ALAQS. Hence 7 aircraft per hour was chosen as the departure capacity. Time of departure and time of arrival In ALAQS, only one time for arrival and one time for departure are considered. The time spent by an aircraft in taxiing or queuing is computed inside the program. Default engine 19

20 ALAQS calculates emissions on an aircraft by aircraft basis. The source data identifies the aircraft type and ALAQS relates it to a default engine. Default emission factors for engine originated mainly from the ICAO databases and few directly from manufacturer data. A report presented in Annex G summarises consideration about default engine. After comparing the results of a first simulation with LASPORT results, it was concluded that ALAQS default engine combinations were not representative of Zurich airport traffic (See Annex C for details of engine types). Further investigation of Zurich traffic data, including for each aircraft type a repartition of the mostly used engine, lead to the definition of new engine/aircraft combinations in ALAQS Zurich study. It was decided that for each aircraft type the engine considered in ALAQS would be the one most used in LASPORT. Annex C shows the new engine/aircraft combinations as they were used in ALAQS and the subsequent results are presented in annex D. However, for the last simulation; emission factors derived from Zurich Airport data (based on the real aircraft/engine fleet population and weighted average of emissions factors for each aircraft group) are used in ALAQS in order to provide a reliable comparison. For some propeller category aircraft, the engine fits were not available. Thus it was decided to assign as propeller default engine. The popular TIO-540-J2B2 was chosen. Similarly the default propeller profiles were GASEPV-D-1 (for departure) and GASEPV-A-1 (for arrival). Default engine emission factors The ALAQS-AV default emission factors originate mainly from ICAO Engine emission data but also from other sources (manufacturer database, FAA, FOI etc.) for example in the case of turbo-propeller aircraft. However, for the last simulations Zurich weighted average of emissions factors for aircraft groups are used to allow better comparison between the models. The actual factors are given in Appendix B Table 33. APU, GPU and GSE ALAQS-AV uses the Gate scenario concept, which defines for each gate type aircraft group combination the amount of pollutants released at the gate during departure and arrival operations. Gates can belong to one of the following categories: freight, open or terminal. The gate type will determine the emission factors to be used. A gate scenario includes GSE (Ground Support Equipment), APU (Auxiliary Power Unit), GPU (Ground Power Unit) and aircraft engine start emissions. For each gate type aircraft group combination; the operational times of APU and GPU were implemented in ALAQS as well as the breakdown between arrival and departure for both GSE and Start Engine Emissions. Yearly operational values and emission factors were provided by Zurich Airport LASPORT assumptions details Departure and Arrival Profiles In LASPORT the flight profiles for departures are based on the German noise classification (AzB). For each aircraft group there is one departure profile as shown in Figure 3-4 below. The arrival profile of all aircraft groups is defined as a linear approach 3 degrees slope. Take-off runs vary as function of type and thrust cut-back for jets occurs at 1500ft. 20

21 Level above Ground [m] Arrival Departure Large Departure Medium Departure Small Departure Regional and Business Departure Turboprop Departure Piston Distance [m] Figure 3-4: Default aircraft profiles in LASPORT (Based on the German noise classification AzB) Time of departure and time of arrival LASPORT movement table has to contain the time at which the aircraft is on-stand (on-block-time or off-block-time) and the runway times (take-off-time or landing-time). Departure/Arrival status is calculated by comparison of the runway and stand times. Aircraft / engine match Due to the extensive data available at Zurich airport, it was possible for LASPORT to use the real match between aircraft and engine using the registered tail number. LASPORT thus overcame the other programs by reducing drastically the assumptions about engine/aircraft combination. APU, GPU and GSE Operational data was obtained for aircraft, GSE, APU, vehicle and stationary source activity at Zurich Airport. This data was based on annual operations and in most cases reflected actual airport activity. The GSE methodology is linked to the movement table, as for each aircraft group and gate, APU (time and emission factors), GPU (time and emission factors) and GSE emissions (amount per operation) are predefined and used when this particular aircraft performs the operation EDMS assumptions details Aircraft departure and arrival profiles 21

22 The flight profile, annual average taxi and queue time, and the runway time in mode was required for each aircraft category. The approach profile angle was set at the EDMS system default of 3. The annual average taxi and queue times were determined from Zurich Airport data. Runway time in mode for each aircraft category was determined using EDMS default flight profile parameters. Table 3-2 provides further detail for each of the four modes, take off, climb out and approach and landing roll per aircraft category. Table 3-2: EDMS profiles (times in mode) Time of departure and arrival EDMS does not use Departure/Arrival times since it is not based on a movement table but on a LTO count system. Note. In EDMS each aircraft is assigned the annual number of landing and takeoff cycles (LTO), and touch-and-go (TGO) operations. On the contrary in ALAQS each flight/aircraft is either an Arrival or a Departure). Aircraft / engine match Approximately 5,000 aircraft related data records representing 265,882 aircraft movements (LTO cycles) were provided for entry into the EDMS system. These data records were consolidated into 7 aircraft categories in order to streamline the EDMS data entry process. These aircraft categories, which reflect established Zurich Airport categories, are based on the type and size of aircraft operating at Zurich Airport and are identical in LASPORT. For each of these seven categories, a representative aircraft, engine, APU combination, and overall aircraft classification was selected from the EDMS system (for determination of a flight profile). 22

23 Table 3-3: EDMS parameters per aircraft group Engine emission factors for each of the 7 aircraft categories were entered into the EDMS system. The indices for each group were calculated using the exact emission factors from all aircraft within that category. Data pertaining to the indices that included CO, HC and NOx were derived from Zurich Airport's movement database for For this particular study, the emission factors for LASPORT and EDMS are thus identical. APU, GPU and GSE Operational data were obtained for aircraft, GSE, APU, vehicle and stationary source activity at Zurich Airport. This data was based on annual operations and in most cases reflected actual airport activity. In cases were airport related data was not available or where it was determined that the EDMS data may be more appropriate, EDMS default data was used. The EDMS default values for APU, GPU and GSE were chosen based on an evaluation of Zurich equipment. Thus the EDMS defaults were as realistic as possible. 3.4 Pollutants considered The main pollutants considered in the three programs are CO, HC and NO x. Airports are mainly concerned about NO x (as it forms NO 2 which is regulated): which is emitted mostly by aircraft but also by every other source present on the airport. Calculations were also made for PM10 and SO x for some sources. As there are still holes in the knowledge concerning the PM emissions (ICAO defined a smoke number as opposed to a PM10 emission factor) the results given in the report are only indicative. 23

24 4 Emission Inventory Results This section presents the global emissions results for CO, HC and NOx for the three models. The parameters used and more detailed results for the results presented in this section can be found in Annex B. 4.1 ALAQS Emission Inventory Table 4-1: ALAQS Emission Inventory for Zurich 2003 Emissions per source Source Group Emission Source CO HC NOx (t/a) (t/a) (t/a) Air Traffic Aircraft Handling Infrastructure Aircraft Main Engine Ignition Aircraft APU GPU GSE Refuelling Aircraft De-icing Power Generation Plant (including emergency power station) Heat Plant Aircraft maintenance Airport maintenance Construction Other: Engine test run Helicopters Helicopters Landside Traffic Roadways (airside + landside + parking) Total Features for which there were no available data are the fuel farm and the de-icing activities. SO x and PM10 emissions were most of the time approximately zero due to the missing data concerning those pollutants. Execution time of the annual emissions inventory calculation was approximately 7 hours. To avoid overloading the PC, emission calculations were made separately for each source aircraft calculations took approximately 2 hours, stationary sources approximately 5 minutes. This procedure was also adopted to isolate each source because results for individual sources were not implemented in ALAQS-AV version 1. Roadways and GSE emissions were calculated with version 1 of the ALAQS-AV vehicle module, which was a stand alone application based on COPERT III. ALAQS result files were too big for the basic post-processor available in the tool. As a consequence, they had to be imported in Access so that the total emissions per source could be compiled. However this problem will be solved in the next version of ALAQS. 24

25 4.2 LASPORT Emission Inventory Table 4-2: LASPORT (1.3) Emission Inventory for Zurich 2003: Source Group Emission Source Emissions per source CO HC NOx (t/a) (t/a) (t/a) Air Traffic Aircraft 1' ' Handling Infrastructure Aircraft Main Engine Ignition Aircraft APU GPU GSE Roadways (airside) Refuelling Aircraft De-icing Power Generation Plant Heat Plant Emergency Power Stations Aircraft Maintenance Airport Maintenance Fuel Farm Fire Exercise Construction Other: Engine test run Helicopters Helicopters Landside Traffic Roadways (landside + parking) Total Setting up parameters for the studies presented in the current section can be consulted in Annex B. 25

26 4.3 EDMS Emission Inventory Table 4-3: EDMS (4.12) Emission Inventory for Zurich 2003 Source Group Emission Source Emissions Per Source 1 CO HC NOx SOx PM10 (t/a) (t/a) (t/a) (t/a) (t/a) Air Traffic Aircraft Handling Infrastructure Aircraft Main Engine Ignition Aircraft APU GPU GSE Roadways (Airside) Refuelling Aircraft De-icing Power Generation Plant Heating Plant Emergency Power Stations Aircraft Maintenance Airport Maintenance Fuel Farm Fire Exercise Construction Other: Engine test run Helicopters Helicopters Landside Traffic Roadways (Landside + Parking) Total N.B. All emission values in table 4.3 are short tons (= kg) 0: value is 0 / -: value is not calculated The Air Traffic source group noted above was derived using the 7 aircraft categories created for operational aircraft data input. These categories, which ranged in size from large jet to piston propeller aircraft, generated approximately 1,417 tons of CO, 170 tons of HC and 1,072 tons of NO x (including APU operations). Setting up parameters for the studies presented in the current section can be consulted in Annex B. 26