Recent Advances and Implications for Aviation Emission Inventory Compilation Methods

Abstract

With the rapid development of industrialization and urbanization in China, civil aviation plays an increasingly important role in the transportation industry. However, pollutants and greenhouse gas (GHG) emissions from civil aviation are becoming an increasingly concerning environmental problem. In order to mitigate the resulting environmental pollution, such as air quality deterioration, regional and global climate warming, and declining human health, more and more efforts have been devoted to reducing both pollutants and GHG emissions. Among these efforts, emissions inventories from civil aviation provide a basis for quantifying pollutants and GHG emissions, establishing evaluation standards of environmental impact, and formulating management policies for both air quality improvement and climate change mitigation. In this paper, we reviewed both compilation approaches and data collection methods for civil aviation emissions inventories, introduced several typical calculation methods for aviation emissions inventories, and analyzed specific cases of actual application based on typical methods of inventory compilation. We also described in detail the activity level and emission index calculation methods of several pollutants and greenhouse gases. Furthermore, based on the above research methods, four typical application cases were investigated, including a specific airport, the landing and takeoff (LTO) cycle of a nation, the entire period with the LTO cycle and the climb–cruise–descent (CCD) phase of a country, and global emissions inventories from civil aviation. The results suggest that, in addition to quantifying the emissions of both pollutants and GHG produced by civil aviation, the selection of inventory compilation methods is likely to be important for improving aviation emission inventory accuracy and for further reducing the environmental, economic, and health impacts resulting from aviation emissions. Moreover, this paper can also provide a reference and theoretical basis for the development of aviation emission inventory compilation methods in the future.

1. Introduction

It is well known that civil aviation plays a significant role in the transport industry. Especially after the pandemic, global aviation transportation has been gradually recovering, and the demand for air travel is expected to double by 2040 at an average annual growth rate of 3.4%. In the first quarter of 2023, owing to air passenger traffic surging, airline revenue in the Asia–Pacific region accounted for 31.9% of global air passenger revenue [1]. Along with the rapid development of the aviation industry, aircraft emissions have attracted more and more environmental concerns. The main pollutants from aircraft exhaust are carbon monoxide (CO), sulfur dioxide (SO₂), volatile organic compounds (VOCs), particulate matter (PM), nitrogen oxides (NO_x), hazardous air pollutants (HAPs), and other trace components [2]. NO_x and VOCs from aircraft emissions, as the two main precursors of O₃ and secondary PM_2.5, lead to the deterioration of environmental air quality [3], which can directly harm human health via the inhalation of high concentrations of secondary pollutants.

The World Health Organization (WHO) issued the latest revision of the Global Air Quality Guidelines, which emphasized that aviation emissions could harm human life and health [4,5,6,7]. In addition to polluting emissions, carbon emissions have also been gaining more attention. As we all know, the Chinese government has declared that its action goal will be to achieve a carbon peak by 2030 and carbon neutrality by 2060. Thus, all provinces in China must formulate an action plan for carbon peak and carbon neutrality and incorporate it into the country’s comprehensive evaluation system for economic and social development [8,9]. With the substantial increase in aircraft energy consumption and atmospheric pollutant emissions, the impact of aviation emissions on air quality has become increasingly significant, especially around airports and their surrounding areas. Therefore, the compilation of an emission inventory for pollutants from civil aviation is essential. It can provide crucial data support for pollutant control, improvement, and prediction efforts, as well as for the initial design process of engines. The inventory will also enable the proposal of improvements for airport planning and aircraft operation modes. Furthermore, it can drive innovations in engine technology, facilitate the optimization of flight procedures, and serve as input data for dispersion models.

The International Civil Aviation Organization (ICAO) divides the aircraft flight phase into two parts: the landing and takeoff (LTO) cycle and the climb–cruise–descent (CCD) stage. Studies on emissions inventories for single airport or regional airport groups typically focus on the LTO cycle only, and some studies just consider the emissions of the main aircraft engines [10,11,12,13]. However, airport emissions inventories in other studies not only include main engine emissions but also consist of both the emissions of auxiliary power units (APUs) and ground specialty vehicles [14,15,16]. In order to improve the precision of emissions inventories, the altitude of the atmospheric mixed layer and the actual aircraft flight time are corrected in the ICAO standard emission model. Based on the ICAO emission model, Cao et al. refined the impact of thrust setting changes at the different stages of the LTO cycle, and specific information, such as airport runways, annual takeoff, and landing sorties, was used to correct the aircraft taxi time and establish an airport group emissions inventory [17]. In recent years, the emissions during the CCD phase have been increasingly investigated. Dong et al. first used quick-access recorder (QAR) data to build an accurate estimation model for the emissions inventory of a single flight. This method was employed to estimate the emissions of air pollutants resulting from domestic civil aviation activities from 2018 to 2021 in China [18]. Zhang et al. used the spatial allocation method to assign the emissions of a single flight during the LTO cycle and CCD phase to the departure and arrival airports [19]. Cui et al. used the Boeing Fuel Flow Method II (BFFM2), First-Order Approximation (FOA), and the Fuel Percentage Method (FPM) to calculate the emission of CCD phases [20]. Mokalled et al. provided a method to assess airport emissions for a country lacking activity level data and evaluated the LTO emission inventory of an international airport in the Middle East [21]. Winther et al., while considering gaseous pollutants, utilized FOA3.0 to create a detailed emission inventory of particulate mass and number from airport aircraft main engines, APUs, and ground handling equipment, finding that more than half of the particulates originated from sulfur in the airport’s fuel [22]. Pan et al. explored the impact of time-based separation on aircraft pollutant emissions at coastal airports and under headwind conditions, offering practical guidance for flight procedure design [23]. Zhou et al. estimated pollution emissions during the LTO phase of nationwide civil aviation aircraft by implementing mixed-layer height information to calculate climbing and approach phase times [24]. Lawal et al. used the Weather Research and Forecasting (WRF)–Sparse Matrix Operator Kernel Emissions (SMOKE)–Community Multiscale Air Quality (CMAQ) model to simulate air quality in the Atlanta metropolitan area during August of 2019 [25].

The compilation methods for aviation emission inventories still face numerous challenges. Firstly, most studies establish inventories based on LTO cycle emissions, with relatively little research conducted on the CCD phase. The CCD phase is a crucial component of the entire flight journey, and the characteristics of aviation emissions at high altitudes have not been fully elucidated. Compared to the LTO cycle, the CCD phase involves longer flight durations and more complex flight environments. This makes obtaining emission data for the CCD phase more difficult, requiring advanced monitoring technologies and data processing methods. Secondly, aviation emission factors often rely on recommended values, which can deviate from the actual operating conditions of engines. There are significant technical differences among various types of aviation engines, leading to varied emission profiles. However, in the aviation industry, deploying comprehensive emission monitoring equipment and systems is not only costly but also technically challenging. Additionally, most aviation emission inventories utilize historical air traffic data rather than real-time air traffic data. Activity level data are essential for improving accuracy, but QAR data collection is difficult and complex [26]. Automatic Dependent Surveillance–Broadcast (ADS-B) technology relies on a global navigation satellite system to locate targets, so it cannot verify target locations. When promoting ADS-B technology, standards must be formulated to allow for ADS-B airborne equipment installation in factory aircraft [27,28]. Due to the particular nature of the method for obtaining activity levels and the application of different aviation emission cycles, the aviation emissions inventory compilation methods vary greatly.

In this paper, we summarized various methods that contribute to the establishment of aviation emissions inventories by obtaining emission factors and activity levels, analyzed solutions and specific cases of emissions inventory preparation in previous studies, and introduced various calculation methods and application scenarios for emissions inventories on the basis of summarizing the development of aviation emissions inventories. Furthermore, we explained different data collection methods for emissions inventories, including the determination method for emission factors and the collection strategy for activity levels, described and examined classic emissions inventory cases in depth, and proposed future development directions for aviation emissions inventories. Overall, our work aims to provide support for the compilation of aviation emissions inventories and to provide a reasonable basis for the formulation of emission reduction policies.

2. Inventory Compilation Methodology for Aviation Emissions

An emissions inventory refers to the collection of air pollutants emitted by various sources within a certain period and area. The inventory can be used to illustrate emission characteristics and provide a reference for the formulation of emission reduction policies. As an important part of the transportation industry, the civil aviation industry greatly differs from railways, highways, and other means of transportation because it involves a large amount of cross-border passenger and cargo transportation, along with increased global carbon emissions. The aviation industry accounts for 2% of global carbon emissions and exhibits a high annual growth rate [29]. Affected by multiple factors, such as the external atmospheric temperature, pressure, flight altitude, and thrust, the emission characteristics of civil aircraft vary, which increases the difficulty of establishing aircraft emissions inventories. Furthermore, these various influencing factors can be utilized as key indicators to validate the accuracy of aviation emission inventories.

Bottom-up emission evaluation can be obtained by summing the emissions of each flight with specific activity data. Top-down emission evaluation can be obtained based on global jet fuel production [30]. Generally, bottom-up methods have shown different calculation results from top-down methods [31] in addition to a high level of accuracy [32]. In this section, several calculation methods for emission inventory establishment are introduced, such as the ICAO method, the U.S. Environmental Protection Agency (EPA) method, the European Monitoring and Evaluation Program (EMEP) method, the Airport Local Air Quality Studies (ALAQS) method, and the Study of Optimization Procedures for Decreasing the Impact of Noise (SOURDINE) II method. These methods are described and compared in this section in order to determine which method would provide a reliable assessment for evaluating aircraft pollutant emissions. It is also possible to use various methods for different sources in inventory compilation or, given the same emission source, a combination of methods can be employed in cases where different parameters are needed for emission calculation [33].

2.1. ICAO Approach

According to the data requirements and the complexity of the calculation method, the ICAO has developed three methods (simple, advanced, and sophisticated), and their accuracy differs [32].

2.1.1. ICAO Simple Approach

The ICAO simple method requires a minimal amount of data but leads to high uncertainty in the estimation of aircraft emissions. The simple method can be divided into methods A and B according to different data. The flight data needed for method A are the number of LTO cycles during a certain period, and the emission factor can be obtained according to ICAO Doc9889 [32]. Emissions can be calculated using Equation (1) as follows [34,35,36,37]:

$${\mathrm{E}}_{\mathrm{i}}=\sum {\mathrm{E}\mathrm{F}}_{\mathrm{i},\mathrm{j}}\times {\mathrm{N}}_{\mathrm{j}-\mathrm{L}\mathrm{T}\mathrm{O}}$$

(1)

where E_i is the total emission of pollutant i (kg), EF_i,j is the emission factor of pollutant i produced by aircraft type j (kg/LTO), and N_j-LTO is the number of LTO cycles.

NO_x, HC, CO, SO₂, and CO₂ emissions can be determined using the above equation, but there is no provision for PM emissions calculation. The above equation does not account for specific engine types, modes of operation, and time-in-modes (TIMs); it assumes that the conditions are the same or similar to the default data used.

The additional data needed for method B pertain to the aircraft and engine type involved in each flight, and the TIM of each stage can be derived from the ICAO TIM recommendations for different stage thrust settings, as listed in

Table 1. ICAO reference time-in-mode and thrust setting during the LTO cycle.

Mode	TIM (min)	Thrust (%)
Approach	4	30
Takeoff	0.7	100
Climb-out	2.2	85
Taxi/Idle	26	7

. The fuel flow rate and emission index of NO_x, HC, and CO for the different operating phases can be obtained from the Engine Emissions Databank (EEDB). The NO_x, CO, and HC emissions can be calculated with Equation (2) [38,39].

$${\mathrm{E}}_{\mathrm{i},\mathrm{j}}=\sum {\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{j},\mathrm{k}}\times 60\times {\mathrm{F}\mathrm{F}}_{\mathrm{j},\mathrm{k}}\times {\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j},\mathrm{k}}\times {\mathrm{N}\mathrm{e}}_{\mathrm{j}}$$

(2)

where EI_i,j,k is the emission index for pollutant i per kilogram of fuel in mode k for each engine used on aircraft type j (kg/kg), FF_j,k is the fuel flow (kg/s), TIM_j,k is the time for mode k for aircraft type j (min), and Ne_j is the number of engines used on aircraft type j.

Notably, the ICAO does not provide SO_x emission certification standards. The U.S. EPA conducted a survey of the sulfur content in commercial aviation jet fuel, and the results showed that the United States produces an average of 1 g of SO_x per 1000 g of fuel consumed (EI_SOx = 1 g/kg fuel). However, if more accurate results are needed, this average value should not be adopted. The sulfur content can be calculated using the FOA V3.0 (FOA3) method for emissions calculation via Equation (3):

$${\mathrm{E}}_{\mathrm{j}}=\sum {(\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{j}}\times 60\times {\mathrm{E}\mathrm{r}}_{\mathrm{j}}\times {\mathrm{N}\mathrm{e}}_{\mathrm{j}})$$

(3)

$${\mathrm{E}\mathrm{r}}_{\mathrm{j}}=1\times {\mathrm{F}\mathrm{F}}_{\mathrm{j}}$$

(4)

where Er_j is the emission rate of the total SO_x in units of grams of SO_x emitted per second per operational mode for aircraft j (g/s).

2.1.2. ICAO Advanced Approach

During actual aircraft operation, the fuel flow and emission index vary with flight altitude, thrust level, outside atmospheric temperature, humidity, and pressure. The advanced ICAO method entails the use of the BFFM2 to correct the effects of these factors. This ensures that the obtained main engine emissions are more accurate than those calculated by the simple method.

The NO_x, CO, and HC emissions can be calculated using Equation (5). The fuel flow correction model is used to correct the aircraft fuel flow rate at different altitudes. The relationship between the emission index and fuel flow rate is used to modify the emission index with flight altitude [40].

$${\mathrm{E}}_{\mathrm{i},\mathrm{j}}=\sum {(\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{j},\mathrm{k}}\times 60\times \mathrm{f}({\mathrm{F}\mathrm{F}}_{\mathrm{j},\mathrm{k}},{\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j},\mathrm{k}} \mathrm{o}\mathrm{r} {\mathrm{T}\mathrm{h}\mathrm{r}\mathrm{u}\mathrm{s}\mathrm{t}}_{\mathrm{j},\mathrm{k}},{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{d}}_{\mathrm{j}},{\mathrm{N}\mathrm{e}}_{\mathrm{j}}))$$

(5)

where Thrust_j,k is the thrust level for mode k for the aircraft type j, Cond_j is the ambient conditions (forward speed, altitude, pressure, temperature, and height) for aircraft type j movement, and f is a function.

2.1.3. ICAO Sophisticated Approach

To truly characterize the actual emissions from an aircraft, the ICAO sophisticated method proposes that actual and precise data should be obtained from real-time measurements, performance information reports, and sophisticated computer model outputs. These data and information describe the actual fleet composition, including aircraft type and engine combination, TIM, thrust level, fuel flow, and combustion chamber operating conditions at all stages, covering all possible ground and takeoff operations. The sophisticated method allows for more in-depth analysis and optimization of aircraft emission characteristics with the highest accuracy. However, this method heavily relies on the acquisition and conversion of real-time dynamic information from an aircraft, and it is, to some extent, difficult to implement. Therefore, this approach has only been more widely adopted in recent years. Currently, QAR and ADS-B data are used. If the actual fleet engine emission factors, TIM, and fuel flow are known, LTO emissions can be calculated using the same equation as the advanced method [41].

Although the ICAO sophisticated method of calculating emissions using actual monitoring data is limited by data availability and method operability, which would increase research cost, complexity, and application difficulty, it shows the highest accuracy among the methods. Therefore, under the premise of obtaining various actual monitoring data, the application of this method should be promoted.

2.2. EPA Approach

The EPA method is similar to the ICAO simple B method for compiling emission inventories of aircraft LTO cycles. The main difference is that the TIM in the EPA method does not directly match the ICAO reference value but accounts for the influence of meteorological conditions at the different altitudes, which in turn affects the aircraft operating time during the approach and climb phases [42].

The duration of the approach and climb modes highly depends on the height. The EPA guidelines provide approach and climb times for the 3000 ft default altitude and procedures for adjusting these times for different altitudes [43]. The adjustment can be calculated using Equations (6) and (7) as follows:

$$\mathrm{Climb-out}: {\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{a}\mathrm{d}\mathrm{j}}={\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{d}\mathrm{f}\mathrm{l}\mathrm{t}}\times \left[{\displaystyle \frac{\mathrm{M}\mathrm{i}\mathrm{x}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{H}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}-500}{3000-500}}\right]$$

(6)

$$\mathrm{Approach}:{\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{a}\mathrm{d}\mathrm{j}}={\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{d}\mathrm{f}\mathrm{l}\mathrm{t}}\times \left[{\displaystyle \frac{\mathrm{M}\mathrm{i}\mathrm{x}\mathrm{i}\mathrm{n}\mathrm{g} \mathrm{H}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{3000}}\right]$$

(7)

where TIM_adj is the adjustment time in the approach or climb mode and TIM_dlft is the default time.

The weighted average emission factor represents the average emission factor for each LTO cycle across all engine models used on a particular type of aircraft. The weighted average emission index can be calculated with Equation (8) as follows:

$${\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j},\mathrm{k}}={\sum }_{\mathrm{m}=1}^{{\mathrm{N}\mathrm{e}}_{\mathrm{j}}}{(\mathrm{X}}_{\mathrm{m},\mathrm{j}}\times {\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{m},\mathrm{k}})$$

(8)

where X_m,j is the proportion of the j aircraft with the engine type M.

The total emissions of each LTO cycle for a given aircraft type can be calculated using Equation (9) as follows:

$${\mathrm{E}}_{\mathrm{i},\mathrm{j}}={\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{j},\mathrm{k}}\times {\displaystyle \frac{{\mathrm{F}\mathrm{F}}_{\mathrm{j},\mathrm{k}}}{1000}}\times {\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j},\mathrm{k}}\times {\mathrm{N}\mathrm{e}}_{\mathrm{j}}$$

(9)

2.3. European EMEP Approach

The European Environment Agency (EEA) has published the EMEP method using the decision tree hierarchy concept. The EPEM method is subdivided into Tiers 1, 2, and 3 based on complexity, data requirements, and research purposes. In the case of unconstrained data requirements, the higher the activity level, the more accurate the emissions inventory. Two factors should be considered when calculating emissions: as much detail as possible should be used, and if the source category is a critical source, Tiers 2 and 3 must be used for emission estimation [44].

The Tier 1 aviation emission calculation method is based on the amount of aviation fuel consumption, and it can be used to calculate the emissions of domestic and international aircraft during the LTO cycle and the cruise phase (NO_x, CO, CO₂, HC, SO₂, NMVOC, CH₄, N₂O, PM_2.5, and PM₁₀), as expressed in Equation (10) [45,46,47].

$${\mathrm{E}}_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}}={\mathrm{A}\mathrm{R}}_{\mathrm{f}\mathrm{u}\mathrm{e}\mathrm{l}\_\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}}\times {\mathrm{E}\mathrm{F}}_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{l}\mathrm{u}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{t}}$$

(10)

where AR_{fuel_consumption} is the activity rate by fuel consumption for each of the flight phases and trip types.

The Tier 1 method is a fuel-based approach that is only suitable for a rough estimate of emissions; it calculates emissions using the average emission index and the number of aircraft provided in the EEA/EMEP guidebook. If the total emissions of CO₂, SO₂, and heavy metals are estimated, the Tier 1 method is sufficient, as the emissions of these pollutants depend only on the fuel type. In contrast, the PM₁₀ and PM_2.5 emissions depend on the aircraft payload. Therefore, when estimating the total emissions of these pollutants, the Tier 2 method is more appropriate because it considers the aircraft type.

The Tier 3 method is based on actual flight movement data and includes the 3A and 3B methods. The 3A approach accounts for cruise emissions over different flight distances. For domestic and international flights, the use of this method requires a detailed description of the origin and destination airports and aircraft type. The 3A method aims to model the average fuel consumption, LTO cycle emissions, and cruise distance to provide emissions inventories for a series of representative aircraft. The data used in this method account for the emissions variation in different flight phases and the fact that fuel consumption is related to the distance flown. The 3B method differs from the 3A method by the use of aerodynamic performance information of a specific aircraft and engine to calculate fuel consumption and emissions over the flight’s entire trajectory. To use the 3B approach, complex computer models are needed to consider the equipment, performance, and trajectory, as well as calculation parameters, for all flights in a given year.

The above three methods are closely related to fuel consumption and emission index. Tiers 1 and 2 are top-down calculation approaches based on fuel characteristics, while Tier 3 is a bottom-up calculation approach based on flight characteristics. Differences in emissions between aircraft types and modes are ignored in the Tier 1 approach. Moreover, the use of reference emission factors can significantly impact the result uncertainty. The uncertainty can range from 20–30% for the LTO cycle and 20–45% for the cruise phase. In the Tier 2 approach, the uncertainty associated with cruise is higher. The Tier 3 method yields uncertainties ranging from 5–10% and 15–40% during the LTO and cruise phases, respectively.

2.4. Other Approaches

To meet the needs of specific studies and obtain solutions that can help reduce environmental impacts, specific projects are supported internationally to study inventory emission tools, such as ALAQS, SOURDINE II, and MEET. ALAQS and SOURDINE II are introduced in this subsection.

The model method is applied in the calculation of emission inventory. The Emissions and Dispersion Modeling System (EDMS) model is a Gaussian diffusion model that can calculate the emissions of aircraft engines, APUs, GSEs, motor vehicles, and other emission sources in the airport [48]. The FLEXPART model is a Lagrange particle diffusion model that can calculate the transmission, diffusion, settlement, and radioactive decay of particles released from point, line, surface, or body sources at different regional scales, and the diffusion of pollutants from the source to the surrounding area can be simulated by forward operation [49]. The Weather Research and Forecasting (WRF) model is a mesoscale numerical prediction model developed by scientists from several research institutes and universities, including the National Center for Atmospheric Research (NCAR) and the National Center for Environmental Prediction (NCEP) [50]. The CMAQ modeling system is utilized to simulate the physical and chemical processes of aviation-derived pollutant formation and transport [51].

2.4.1. ALAQS

The EUROCONTROL Experimental Centre ALAQS project, initiated in 2003, addresses the strategic, methodological, and practical issues surrounding air quality assessment around airports. The aim of the project is to raise awareness among airport operators and practitioners regarding best practice emissions inventory and dispersion modeling methods, specifically suitable to all European airports. ALAQS achieves this aim through case studies using the ALAQS-AV testbed, resulting in guidelines and methods that can be applied on a pan-European level. The geographical information system (GIS)-based ALAQS-AV tool was specifically developed for the ALAQS project and draws on new and existing methods. ALAQS-AV can be used to conduct airport air quality studies or as a testbed for comparing different emission factors, inventories, and dispersion methods [52,53].

ALAQS-AV is designed to capture airport pollution sources and process different types of emission source estimates into a standard format in preparation for dispersion modeling. At present, the ALAQS-AV database includes the Computer Program to Calculate Emissions from Road Transport (COPERT), the Technical Reference Center for Air Pollution and Climate Change (CITEPA), and the ICAO.

Except for engine start emissions, aircraft engine emissions can be calculated with Equation (11) as follows:

$$\mathrm{A}\mathrm{C}\mathrm{e}={\mathrm{F}\mathrm{F}}_{\mathrm{k}}\times {\mathrm{E}\mathrm{F}}_{\mathrm{k}}\times \mathrm{T}\mathrm{I}\mathrm{M}\times \mathrm{N}\mathrm{e}$$

(11)

where ACe is the total aircraft engine emissions per LTO cycle.

The above parameters can be estimated based on realistic gate scenarios and taxiway and runway operations. To estimate the hourly emissions per LTO mode, the aircraft movement database is linked to a specified aircraft database. Hourly total emission rates are then stored in the database.

2.4.2. SOURDINE II

The SOURDINE II project addresses the new air traffic management (ATM) concept, which aims to mitigate the impacts of aircraft noise and emissions around most airports by defining new approach and departure procedures. TBEC is a Microsoft Access application that has been specifically developed for the SOURDINE II project to calculate aircraft HC, CO, NO_x, SO₂, CO₂, H₂O, VOC, and total organic gas (TOG) emissions. It uses the ICAO EEDB, which provides the fuel flow and emission index at different thrust settings of different engines. The overall principle of TBEC is the calculation by interpolation of emission levels based on the actual thrust [54].

The pollutant emission E_seg can be expressed as follows:

$${\mathrm{E}}_{\mathrm{s}\mathrm{e}\mathrm{g}}=∆{\mathrm{T}}_{\mathrm{s}\mathrm{e}\mathrm{g}}\times \left[{\mathrm{E}\mathrm{F}}_{\mathrm{s}\mathrm{e}\mathrm{g}}\left({\mathrm{P}}_{\mathrm{i}}\right)+{\displaystyle \frac{{\mathrm{P}}_{\mathrm{s}\mathrm{e}\mathrm{g}}-{\mathrm{P}}_{\mathrm{i}}}{{\mathrm{P}}_{\mathrm{i}+1}-{\mathrm{P}}_{\mathrm{i}}}}\left({\mathrm{E}\mathrm{F}}_{\mathrm{s}\mathrm{e}\mathrm{g}}\left({\mathrm{P}}_{\mathrm{i}+1}\right)-{\mathrm{E}\mathrm{F}}_{\mathrm{s}\mathrm{e}\mathrm{g}}\left({\mathrm{P}}_{\mathrm{i}}\right)\right)\right]$$

(12)

where EFseg(P_i) is the emission flow for the segment associated with power setting P_i (g/s), P_seg is the power setting of the segment, E_seg is the emission of the pollutant produced on the segment (g), ∆T_seg is the duration of the flight segment (s), and P_i and P_{i + 1} are the two power setting values bounding P_seg.

2.5. Analysis of the Applicability of Different Compilation Methods

The purpose and need to quantify aircraft emissions determine the inventory resolution and accuracy level, then affect the selection of the compilation method. Each method provides its own advantages and disadvantages, and it should be reasonably selected based on the application scenario [55]. A comparison of the various approaches is provided in

Table 2. Comparison of all the compilation methodologies.

Compare Items	ICAO Simple A Method	ICAO Simple B Method	ICAO Advanced Method	ICAO Sophisticated Method	EPA	EMEP Tire 1	EMEP Tire 2	EMEP Tire 3	ALAQS	SOURDINT Ⅱ
NO, CO, HC	√	√	√	√	√	√	√	√	√	√
SO2	√	√	√	√	√	√	√	√
VOC					√	√	√	√		√
H2O										√
CH4						√	√	√
N2O						√	√	√
PM2.5						√	√	√
PM10						√	√	√	√
CO2	√	√	√	√		√	√	√		√
TOG										√
Scope of application	LTO	LTO	LTO + CCD	LTO + CCD	LTO + CCD	LTO + CCD	LTO + CCD	LTO + CCD	LTO	LTO
Data requirements 1	A	B	C	D	B	A	B	C	C	B
Precision 2	A	B	C	D	B+	B	B+	D	C	B
Emission factor	ICAO Doc9889	ICAO EEDB			FAEED	EEA/EMEP EF databank			ALAQS-AV DB	ICAO EEDB
Country	ICAO				USA	EEA			EUROCONTROL	European Union

¹: A indicates the lowest data requirements, B indicates medium data requirements, C indicates high data requirements, and D indicates the highest data requirements. ²: A indicates the lowest precision, B indicates medium precision, C indicates high precision, and D indicates the highest precision.

.

The ICAO has proposed simple, advanced, and sophisticated methods according to the data acquisition difficulty and calculation complexity; as they increase, the accuracy correspondingly increases. The simple A method assumes that the conditions studied are the same or similar to the default data, so the pollutant emissions are relatively imprecise, and the method does not distinguish flight stages. Only if the study conditions are exactly the same or similar to the default parameters will the ideal results be obtained with the simplest calculation method. This method is widely used to calculate aviation emissions in non-road mobile source emission inventories. The simple B method is more complex and precise than the simple A method, accounting for the TIM, fuel consumption, engine type, and emission index of the different flight stages, although it still uses ICAO reference values. It is more detailed, and the value obtained using method B is more reliable than that obtained by method A when the engine type of each aircraft is known. However, the simple method is only suitable when the need for accurate emission results is not extremely high and when there is no actual flight data or it is difficult to obtain. In the ICAO advanced approach, emissions are calculated using additional meteorological data and existing models, and the parameters of the aircraft emission calculation function can be freely selected. This allows for performance-based calculations using additional information, leading to a more accurate emissions inventory. Due to the consideration of the flight altitude, thrust grade, external atmospheric temperature, pressure, and other factors, the results are closer to the actual values. The sophisticated method is the most accurate method to calculate emissions based on real data. If the emissions inventory involves policies affecting aircraft operations at an airport, the sophisticated method is recommended, but real data are not easily available and the calculations are complex, making this method unsuitable for large-scale calculations. In general, the more complex the method is, the higher the collaboration level needed. These methods can be combined, and the use of the simple method in one part of the inventory does not prevent the use of more complex methods in the rest. When choosing methods to establish an aircraft emissions inventory, a combination of methods can be chosen. The selection is based on the data and information available and the intended inventory accuracy.

The EMEP method comprises three calculation methods: Tiers 1, 2, and 3. Tiers 1 and 2 can be used to construct emissions inventories based on fuel characteristics. In the Tier 1 method, fuel sales data must be acquired, assuming that the amount of fuel used matches that sold each year. A higher accuracy of emission can be obtained for CO₂ and SO₂ in this way. However, to calculate the emissions of other gases, the other two methods are recommended. Tier 2 uses aviation fuel consumption statistics divided into domestic and international components. To divide fuel usage by the LTO cycle and cruising phase, detailed LTO activity and fleet composition knowledge are needed. Tier 3 is based on actual flight data, like the ICAO sophisticated method, so they both attain the highest accuracy.

The EPA method is similar to the ICAO simple B method for calculating aircraft emissions during the LTO cycle. The difference is that the EPA method uses a weighted average emission factor, accounting for the engine composition of each aircraft. Moreover, the TIM does not depend on the ICAO recommended running time but accounts for meteorological conditions, thereby correcting the approach and climb times. Hence, the accuracy is higher than that of the ICAO simple B method. The EPA method should be prioritized in calculating U.S. aviation emissions inventories.

Both the ALAQS and SOURDINE II methods are emissions inventory compilation methods developed for specific cases. The aim of the ALAQS project is to raise awareness among airport operators and practitioners regarding best-practice emissions inventories, and it is applicable to all European airports. The SOURDINE II project aims to alleviate the impact of aircraft noise and emissions around most airports. Both projects were proposed by European agencies, and only LTO cycle emissions are calculated, so the applicability to European airports is higher. If the purpose of the emissions inventory is similar to these two projects, it is recommended to choose these two methods.

In terms of accuracy, the ICAO sophisticated method and the EMEP Tier 3 method are based on real flight data and provide the highest accuracy. Considering the calculation convenience, the ICAO simple A method and EMEP Tier 1 method require the least additional data, and emissions inventories can be established quickly at the expense of accuracy. The ICAO method is the most widely used due to its completeness and systematism. Considering regional applicability, the EMEP, ALAQS, and SOURDINE II methods should be selected for European airports, while the EPA method should be selected for American airports. Considering the various emission gases, EMEP Tier 1 is suitable for the calculation of gases whose emissions are related to fuels only, such as CO₂ and SO₂, while the EMEP method is suitable for the calculation of most gas types, including NO_x, CO, CO₂, HC, SO₂, NMVOCs, CH₄, N₂O, PM₁₀, and PM_2.5. Only the SOURDINE II method can be used to calculate TOG and H₂O emissions.

During the process of compiling an aviation emissions inventory, apart from selecting appropriate compilation methods, the manner of collecting emission factors and activity level data plays a crucial role. Emission indexes serve as key parameters for estimating pollutant emissions, and the accuracy directly impacts the precision of the inventory. Meanwhile, activity level data provide detailed information on the volume of aviation activities, serving as the direct basis for calculating emissions. The accuracy and completeness of these two aspects of data are essential for selecting suitable compilation methods, determining technical routes, and implementing quality control measures. Together, they form the foundation for compiling a high-quality aviation emissions inventory.

3. Data Collection Methods for Emissions Inventories

3.1. Emission Index

The emission index refers to the mass of pollutants emitted by burning 1 kg of fuel, which is an important parameter for calculating emissions.

3.1.1. Emission Index of CO, HC, and NO_x

The emission index of CO, HC, and NO_x under default thrust conditions can be retrieved from the ICAO EEDB, and the weighted average for several engine types is determined as the emission index. The weighted emission index of each pollutant can be calculated with Equation (13) as follows:

$${\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j}}=\sum {\mathrm{E}\mathrm{I}}_{\mathrm{i},\mathrm{j},\mathrm{k}}{\times \mathrm{N}}_{\mathrm{k}}/(1000 \mathrm{N})\times {10}^3$$

(13)

where N_k is the number of aircraft with engine type k and N is the total number of aircraft.

3.1.2. Emission Index of PM and SO₂

The first FOA method, now referred to as FOA1.0, is based on the correlation between the smoke number (SN) in the ICAO EEDB and the limited nonvolatile PM mass emission data, which only estimates the nonvolatile PM component. In early 2005, work related to the FOA method for volatile components began, and two measurement studies conducted by the U.S. Navy and U.S. EPA, as well as theoretical work by MIT researchers [56], were considered the most effective. It led to the development of a simple scaling technique based on the volatile relative mass fraction versus the nonvolatile relative mass fraction of PM, which was referred to as the FOA2.0 method for total PM estimation. Multiplying the obtained nonvolatile PM estimates by a certain multiplier was considered the best method available at the time, with the three suggested multipliers being 2, 3, and 2, respectively [57]. The ICAO CAEP endorsed FOA2.0 as an interim method and suggested that the method for separating the volatile and nonvolatile components of aircraft PM emissions needs to improve. In November 2005, a group of international experts began developing a more robust FOA approach. They completed this work in October 2006, and in February 2007, the ICAO CAEP reviewed and fully accepted the final FOA3.0 method for international use [58].

At present, PM is calculated using the FOA3.0 method. PM emissions are divided into volatile and nonvolatile components [59], and the volatile components include volatile sulfur components and volatile organic components, which can be calculated (after simplification) using Equations (14) and (15), respectively. The emission index of nonvolatile components can be retrieved from the EEDB. However, to obtain more accurate results, Equation (16) can be used to calculate the mass emission factor.

It is noteworthy that PM emissions from aircraft can also originate from their brakes and tires, distinct from those produced by engine combustion. While this aspect holds significant importance, it lies beyond the realm of consideration for PM emission factors addressed in this article.

$${\mathrm{E}\mathrm{I}}_{\mathrm{v}\mathrm{o}\mathrm{l}-\mathrm{F}\mathrm{S}\mathrm{C}}=(\mathrm{F}\mathrm{S}\mathrm{C}\times \mathsf{\epsilon }\times {\mathrm{M}\mathrm{W}}_{\mathrm{o}\mathrm{u}\mathrm{t}}/{\mathrm{M}\mathrm{W}}_{\mathrm{s}})\times 3\times {10}^6$$

(14)

$${\mathrm{E}\mathrm{I}}_{\mathrm{v}\mathrm{o}\mathrm{l}-\mathrm{F}\mathrm{u}\mathrm{e}\mathrm{l}\mathrm{O}\mathrm{r}\mathrm{g}\mathrm{a}\mathrm{n}\mathrm{i}\mathrm{c}\mathrm{s}}={\mathsf{\delta }\times \mathrm{E}\mathrm{I}}_{\mathrm{H}\mathrm{C}\left(\mathrm{E}\mathrm{n}\mathrm{g}\mathrm{i}\mathrm{n}\mathrm{e}\right)}$$

(15)

where EI_vol-FSC is the EI for FSC (mg/kg), FSC is the fuel sulfur content (mass ratio with default 0.00068), ε is the SIV to SVI fractional conversion (default 0.033), EI_{vol-FuelOrganics} is the volatile PM emissions of the fuel organics (mg/kg), and the value of δ is related to the emission index of HC (

Table 3. EI_HC and δ reference table.

Mode	ICAO EIHC (g/kg Fuel)	δ Value
Take-off	0.04	115
Climb-out	0.05	76
Approach	0.08	56.25
Idle	1.83	6.17

shows the specific value).

$${\mathrm{E}\mathrm{I}}_{\mathrm{n}\mathrm{v}\mathrm{o}\mathrm{l}\mathrm{s}}=\mathrm{Q}\times \mathrm{C}\mathrm{I}$$

(16)

where CI is the emission concentration index (mg/m³) and Q is the exhaust volume per kilogram of fuel combustion (m³/kg). The calculation method for CI and Q can be found in previous research [19].

This method is now widely used. Zhang et al. employed this method to calculate the PM emission index, and the specific pollutants include organic carbon (OC), volatile sulfur particles, and BC [60], which is more meaningful for emission inventory [61,62].

3.1.3. Emission Index of CO₂

According to the 2023 Global Carbon Budget report, global carbon dioxide emissions from burning coal, crude oil, and natural gas are expected to reach 36.8 billion tons, an increase of 1.1% relative to 2022, more than the average annual increase of 0.5% over the past decade, and 6% higher than the Paris Agreement in 2015. Although the European Union and the United States have reduced their carbon emissions significantly, by 7.4% and 3%, respectively, global emissions are still rising. All countries need to achieve economic decarbonization faster in order to avoid the more severe impacts caused by climate change.

In some papers, the value of the CO₂ emission index ranges from 3.148 to 3.173 kg/kg. As agreed in the Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA) in Annex 16, Volume 14 of the Convention on International Civil Aviation, the emission index of Jet-A/Jet-A1 fuel is 3.16. At the beginning of 2022, IATA released the latest air passenger carbon emission calculation methodology, and 3.16 is also the recommended CO₂ emission index.

3.2. Activity Level

The calculation of aviation emissions is generally divided into two cycles, which define flight activities occurring below 3000 ft above the ground as the LTO cycle. The LTO cycle is divided into taxi, takeoff, climb-out, and approach phases, and the part above 3000 ft is denoted as the CCD cycle. The CCD cycle contains the climb, cruise, and descent stages, as shown in Figure 1. The main characteristics of LTO and CCD are shown in

Table 4. The main characteristics of the LTO and CCD phases.

Characteristics	LTO	CCD
Definition	Phases of aircraft taxi, takeoff, climb-out, and approach	Phases of aircraft climb, cruise, and descent
Emission Level	Higher, as engines need to generate significant thrust	Higher during climb, lower during cruise, and reduced during descent
Main Pollutants	CO, HC, and NOx	Similar to LTO but with lower emissions during cruise
Emission Altitude	Low, with direct impact on air quality around airports	Close to the ground during climb and descent, higher altitude during cruise
Environmental Impact	Significant impact on air quality around airports, contributing to PM2.5 and O3 pollution	High-altitude emissions may indirectly affect ground-level air quality through atmospheric transport and chemical transformations

.

A wide variety of emission sources can be found at airports. However, not all types of emission sources actually exist depending on the specific activities at individual airports; to better account for this variability, emission sources were divided into four categories: aircraft emissions (aircraft main engine and APU emissions), aircraft handling emissions (ground support equipment, air-side traffic, aircraft refueling-related and aircraft deicing-related emissions), infrastructure-related and stationary-related sources (aircraft and airport maintenance-related emissions), and vehicle traffic sources.

Theoretically, the total airport emissions should be calculated by accounting for all emissions sources. However, most studies on airport emissions inventories initially only considered the main engine emissions, while some recent airport emissions inventories account for the main engine, APU, and ground support equipment (GSE) emissions. The scope will become increasingly comprehensive in the future.

A large amount of data is involved in the compiling of emissions inventories, and the data may comprise aircraft takeoff and landing data, including the information for each flight departure airport, arrival airport, aircraft model, departure time, arrival time, takeoff time, landing time, total operating time, total flight time, airline name, and airport longitude and latitude, among others. These data can be acquired from the official websites of the Civil Aviation Administration of China (CAAC) and major airports. Aircraft configuration data can be obtained from the civil aviation statistics published by the CAAC. The data on the aircraft type, International Air Transport Association (IATA) code, engine number, engine model, and wake class can be obtained from the Aviation Emissions Inventory Code (AEIC). Matching data on the aircraft type and engine can be derived from the official website and literature provided by the aircraft manufacturing company. Engine fuel consumption rate, SN, emission factors, and pressure ratio can be retrieved from the ICAO EEDB. The correlation between the air pollutant and carbon emissions of different aircraft models and the flight time during the CCD cruise is recorded in the Base of Aircraft Data (BADA) database of EUROCONTROL, as shown in

Table 5. Data summary for emissions calculations.

Data Type	Data Composition	Data Source
Aircraft takeoff and landing data	Departure/arrival airport Aircraft model Departure/arrival time Takeoff/landing time Total operating/flight time Airline name Longitude/latitude information	CAAC official website Major airports official website
Aircraft configuration data	Aircraft type IATA code Engine number Engine model Wake class	Civil Aviation from statistics AEIC
Matching data of aircraft and engine	/	Official website of the aircraft manufacturing company
Emissions calculation data	Engine fuel consumption rate SN Emission factors Pressure ratio	ICAO EEDB
Correlation between air pollutants of different aircraft types and flight time during the CCD cycle	/	BADA

.

At present, there are two types of actual flight data involved. QAR data encompass more than 2000 parameters related to flight information, including operating status, spatial position, fuel flow, and total amount of remaining fuel, and its flight data is processed at intervals of seconds. The data source is the Civil Aviation of China flight quality monitoring base station. The interval of the ADS-B data signal is 30~70 s, including real-time longitude, latitude, altitude, speed, and other flight information. This data type may include additional information, such as wind speed, wind direction, and outside temperature of the aircraft. The missing data due to device signal loss or other reasons can be replaced by data from similar flights, which can be selected according to the origin and destination airports and aircraft type. The standard for distinguishing domestic and international routes is that routes with domestic takeoff and landing terminals are domestic routes, and those with foreign takeoff and landing terminals are international routes.

Regarding the aircraft pollutant emission measurement method, Heland and Schafer employed ground-based FTIR (Fourier transform infrared spectroscopy) technology to conduct remote measurements of emissions from actual operating aircraft engines, obtaining data on CO, H₂O, CO₂, NO, and NO₂. However, ground-based measurements are susceptible to interference from other pollutants or background radiation present in the atmosphere. Additionally, the parameters for high temperatures may be incomplete, thereby affecting the accuracy of the data [64]. Melamed et al. utilized visible spectroscopy aboard aircraft to measure the NO₂ in the atmosphere. In conjunction with in situ chemiluminescence measurements of NO and NO₂ performed by onboard instruments, they estimated NO_x emissions. However, the accuracy of this method can be impacted by various factors, including environmental interference, uncertainties in air mass factors, and poor data matching [65]. Anderson et al. conducted offline analysis of the exhaust plumes collected from aircraft engines at ground level to test for carbonaceous species, including CO, CO₂, CH₄, and non-methane hydrocarbons (NMHCs). Due to the rapid dilution of the exhaust, precise dilution correction methods were employed to ensure the accuracy of the measurements [66]. Schumann et al. utilized open-path equipment to measure real-time concentrations of NO, NO₂, CO, and CO₂ and studied the spatiotemporal variations in VOCs within airport regions by collecting air samples and analyzing the mixing ratios. Capturing the transient characteristics of aircraft emissions accurately demands data with high spatial and temporal resolution, which increases the complexity and computational load of data processing [67]. Agarwal et al. collected and analyzed PM directly from the exhaust plume of aircraft engines, further examining their physicochemical properties. The extremely high temperatures and velocities of aircraft engine exhaust pose stringent requirements on the heat resistance and sampling rate of the sampling equipment [68].

From the analysis of the aforementioned application cases, it can be inferred that to enhance the accuracy of aircraft emissions measurements, considerations should be given to aspects such as measurement methods and equipment, data processing methods and model calculations, and environmental influencing factors. Employing high-precision sampling and analysis methods and appropriate data processing and correction techniques, along with ensuring stable engine operating conditions and instrument performance, can all contribute to ensuring the accuracy of the results.

3.3. Challenges of the Data Collection

The collection of aviation emission data and activity level data poses a series of difficulties and challenges, primarily stemming from the complexity and diversity of the data, as well as technical and administrative constraints.

Firstly, the diversity of data sources is a significant challenge. Aviation emission data involve multiple departments and agencies, such as the CAAC, airlines, airports, and environmental protection departments. Currently, the sharing mechanisms among these data sources are not well-established, resulting in considerable difficulties in data acquisition.

Secondly, there is uncertainty surrounding the emission index. Emission indexes vary across different aircraft types and flight phases and are influenced by numerous factors such as weather conditions, flight altitude, and speed, making them difficult to accurately quantify. In addition to direct fuel combustion emissions, aviation activities also generate indirect emissions, such as energy consumption by ground service equipment and carbon emissions during airport construction and maintenance. The calculation of these indirect emissions is even more complex.

Furthermore, the complexity of real-time data collection presents another challenge. In practical applications, ADS-B data may contain potential biases that could adversely affect flight safety and efficiency. If GPS signals are interfered with or obstructed, it may lead to inaccuracies in position information. Inconsistencies between barometric altitude and geometric altitude, as well as altitude measurement errors, can cause height biases. Influences such as wind speed and changes in aircraft performance can introduce biases in speed information. To mitigate these potential biases in ADS-B data, future efforts could focus on enhancing equipment performance, optimizing data transmission, improving equipment’s environmental adaptability, and establishing monitoring and early warning mechanisms.

In summary, when selecting methods for compiling aviation emissions inventories and data collection, it is essential to comprehensively consider factors such as data availability, accuracy, and timeliness, as well as the reasonability and operability of the compilation methods. Based on the purpose of compiling the inventory, a reasonable boundary for the inventory should be chosen, such as a single airport, regional airports, a national-level inventory, or a global-level inventory. Simultaneously, attention must be paid to the requirements of relevant regulations and standards for inventory compilation. Furthermore, with the continuous advancement of technology and the accumulation of data, more advanced and efficient compilation methods and data collection methods may emerge in the future.

4. Specific Cases of the Actual Application of Emission Inventories

In order to provide a more intuitive introduction to the selection and application of emission inventory compilation methods and data collection techniques, this section analyzes four typical cases. Depending on the research objectives, appropriate methods were chosen for compiling emission inventories at the levels of a single airport, an airport cluster, a national scale, and a global scale. Case 4.1 demonstrates the emissions inventory of a single typical airport, which accounts for the main engines of the aircraft, the APUs, and the special ground vehicles in the calculation rather than only the main aircraft engine emissions as in the previous study. Case 4.2 investigates the LTO cycle of national air routes, which aims to calculate weighted emission factors based on fleet allocation data for determining the air pollutant emissions of the LTO cycle instead of directly obtaining them from the ICAO EEDB. Case 4.3 studies the LTO cycle and CCD cruise phase of national air routes, which use actual flight trajectory records to establish a high-resolution aviation emissions inventory. Case 4.4 examines emissions inventories of the global civil aviation industry, using the Open Aviation Emissions Model (openAVEM 1.0) to simulate each flight by three sources of flight movements, which is not only highly accurate but also has the largest study area.

4.1. Emission Inventories of Zhengzhou Xinzheng International Airport Based on the Localized Emission Factors

This paper aims to promote the study of airport emissions, taking Xinzheng International Airport, an important hub airport, as an example. The emissions calculation technical route is shown in Figure 2.

Previous studies have only considered the emissions from the main aircraft engines in their calculations. In contrast, this study not only takes into account emissions from the main engines but also includes emissions from APUs and specialized ground vehicles. Emissions from the main aircraft engines are estimated based on the landing and LTO cycle defined by the ICAO, encompassing emissions from the ground to the top of the atmospheric mixing layer, using the EPA methodology. This study also revises the actual climb and approach times within the LTO cycle. Additionally, actual aircraft taxiing data provided by the airport are used to calculate pollution emissions during the taxi-in and taxi-out phases.

Emissions from the APU are estimated based on operating times under various workloads and emission indices of pollutants, utilizing the ICAO simplified A method, as shown in Equation (17). The numerous specialized ground vehicles and equipment at the airport also contribute to emissions from the use of fossil fuels. Based on an algorithm provided by the FAA, pollution emissions are obtained based on the operating times and emission parameters of various types of vehicles during a single LTO flight. Furthermore, by considering the annual LTO cycles for each aircraft type, the annual operating times for each specialized ground vehicle are calculated. Multiplying these by the emission factors for each type of vehicle yields the annual emissions from specialized ground vehicles at Xinzheng Airport.

$${\mathrm{E}\mathrm{F}}_{\mathrm{i},\mathrm{k}}={\sum }_{\mathrm{k}}{\mathrm{E}\mathrm{C}}_{\mathrm{i}}\times ({\mathrm{T}\mathrm{I}\mathrm{M}}_{\mathrm{k}}/60)\times 1000$$

(17)

The emission inventory constructed for Xinzheng International Airport in 2019 by this study reveals that the total emissions of NO_x, CO, HC, SO₂, and PM are 1207.7, 921.2, 123.7, 268.3, and 36.2 t, respectively. Aircraft emissions significantly surpass those from GSE. Notably, NO_x emissions are concentrated during aircraft climb/descent phases, exhibiting higher levels in summer and autumn compared to winter and spring, with peak emissions occurring at 11:00. An analysis of localized emission factors for typical aircraft types (A320, A321, B737, and B738) indicates that, except for NO_x emissions in summer, the annual emission factors for other pollutants are lower than the reference values provided by the ICAO for each aircraft model. Consequently, the use of actual data is recommended in such cases. Among these aircraft, the A321 and B738 exhibit the highest fuel consumption, with the B738 accounting for a substantial portion (56.5%) of operational flights, warranting particular attention. Temporal distribution characteristics of airport emissions, together with the contribution of various flight stages and aircraft types to the emission inventory, can provide significant support for improving the basic emission data of localized pollution sources, and analyzing the environmental impact of airport operations can provide a scientific basis for achieving synergy in pollution and carbon emission reduction [69].

4.2. Emission Inventories of Both Air Pollutants and GHG at Chinese Civil Aviation Airports

It is well known that the reference emission factors used in aircraft pollutant emissions inventories differ from the actual operation emission factors, which results in high uncertainty of aircraft emission inventories. Therefore, in this section, we aim to calculate weighted emission factors based on fleet allocation data for determining the air pollutant and carbon emissions of the LTO cycle at airports. Based on 2017~2020 data, taking into account that emissions from the LTO cycle are more than 80% of the total emissions, we calculated aircraft emissions from just the LTO cycle. By means of the calculation method, which is determined according to the characteristics of different emission gases, and on this basis, we established an emissions inventory model. In previous studies, most emission factors were based on the Technical Guidelines for the Compilation of Air Pollutant Emissions Inventories from Nonroad Mobile Sources or the average factor of a single aircraft type. In this study, the average emission factors for national-scale aircraft during the LTO cycle are calculated according to the configuration of the fleet, and the particulate matter emission factor is determined using the FOA3.0 method while the fuel flow rate and SN are also weighted. Emission inventories of gaseous pollutants, such as HC, NO_x, CO₂, and CO, and particulate matter are obtained based on the ICAO simple B method, while SO₂ emissions are calculated using the mass balance algorithm, as shown in Figure 3.

This study has been conducted to develop emission inventories using the LTO cycle at Chinese airports from 2017 to 2020, calculating the annual emissions of pollutants and carbon. Additionally, it analyzed the top 10 airports with the highest year-on-year growth in pollutant and carbon emissions in 2020. Examining temporal changes in pollutant emissions aids in assessing the impact of the pandemic on the aviation industry. The taxiing phase, characterized by incomplete combustion and extended durations, is identified as the primary source of emissions for CO, HC, SO₂, and CO₂ due to incomplete burning and increased fuel consumption. Conversely, the climb phase is recognized as the main contributor to NO_x and PM emissions. This is attributed to the high engine thrust during climbing, which results in greater fuel consumption rates, a lower air–fuel ratio, and higher combustion chamber temperatures and pressures, thereby generating more NO_x and PM. Based on the spatial distribution of pollutant emissions, it is evident that in 2018, the total emissions from East China, Central and South China, Southwest China, North China, Northwest China, Northeast China, and Xinjiang accounted for 26.96%, 25.17%, 16.28%, 14.31%, 7.29%, 6.42%, and 3.56% of the national total, respectively. The emissions are concentrated mainly in East and Central–South China. In response, we propose reducing the fuel consumption rate during the LTO cycle, reducing the emission factor of each pollutant and the LTO cycle time, selecting high-quality fuels (low content of sulfur and aromatic hydrocarbons) and low-emission engines, and optimizing the LTO cycle to reduce aircraft emissions [70].

4.3. High-Resolution Emission Inventories of Chinese Civil Aviation Using Real-World Flight Trajectory Data

Zhang et al. conducted a study on Chinese aviation emission inventory based on the LTO cycle and CCD phases. The model’s methodological framework was established, as shown in Figure 4. The research team developed an aviation emission model grounded on actual flight trajectories, providing a detailed characterization of emission features across the four-dimensional space (time, longitude, latitude, and altitude) throughout all flight phases. By incorporating real-world aviation flight trajectory data, the team significantly enhanced the accuracy and four-dimensional spatial resolution of the aviation emission inventory, markedly reducing the emission simulation errors inherent in traditional simplified great-circle trajectory methods. The fuel consumption was calculated using the ICAO EEDB for the LTO and the BADA dataset for the CCD. The emission index of CO₂, H₂O, and SO₂ were from the ICAO advanced method, and the NO_x, HC, and CO were from the ICAO emission dataset.

The study aimed to leverage more authentic aviation flight information, particularly actual flight trajectory data, to develop a national high-resolution (hourly, 0.001°) aviation emission inventory. This inventory would further be applied to quantify the impacts of China’s rapidly developing aviation industry on air quality and climate. Through an investigation of all flights taking off and landing at Chinese airports, representative ADS-B data from 64 days in 2018 were selected as the data source for the aviation emission inventory. Emission factors were determined based on flight phases and emission gas types, and meteorological data were used to refine the results, thereby establishing the aviation emission inventory model.

The model examined the four phases of the LTO cycle and the three sub-phases of the CCD phase, utilizing actual flight trajectories instead of the traditional great-circle method to achieve a high-precision emission inventory and quantify the environmental benefits of mitigating aviation pollution. The study collated and analyzed key activity level parameters for aviation emissions, such as aircraft takeoff and landing flight data, flight phases, fuel consumption, emission indices, and regional distribution, establishing a methodology framework model. For the emission factor correction in the CCD phase, the Boeing Method II from the ICAO advanced methodology was adopted. Considering the greater significance of BC number concentration compared to mass concentration, calculations were performed for both concentrations.

This case evaluated the current status of aviation emissions, tracked and analyzed pollutants such as CO₂, SO₂, NO_x, HC, BC, OC, PM, and BC_n, and obtained the total aviation emissions and their horizontal distribution in 2018. In addition, the differences in the flight time, fuel consumption, and nitrogen oxide emissions between the results based on the real ADS-B trajectories and those based on the great cycle aviation performance (GCAP) model were compared, and it was determined that the improvement in the measurement method could help to significantly improve the accuracy of the calculated emissions. This study has revealed significant variations in emissions per LTO cycle among different airports, with differences reaching 2 to 4 times. The primary factor contributing to these discrepancies lies in the composition of the aircraft fleet. Consequently, relying on a single average indicator, as suggested in the Technical Guidelines for the Compilation of Emission Inventories for Non-Road Mobile Sources, fails to adequately capture the specific characteristics of individual airports. It is proposed that future studies should utilize real flight data. By analyzing the fuel consumption and pollutant emissions of the top 10 airports in terms of passenger traffic in China, it was found that the fleet composition and taxi time are important factors that cause differences between airports. Hence, the airport operation efficiency should be improved to reduce aviation emissions [19].

4.4. Emission Estimates of Global Civil Aviation Using ADS-B Data

This study is the first to generate spatially resolved global aviation emissions estimates based on ADS-B data, and the impact of the new ICAO nvPM measurements on global aircraft LTO and CCD cycle emissions is reported. The model developed in this study for calculating emissions, referred to as the openAVEM, can simulate each flight by aircraft type–origin–destination combination data, and it forms a corresponding flight phase–duration–fuel flow. The three data sources used are 2018 timeline data from the Market Intelligence Corporation (OAG), ADS-B activity data over the 2017–2020 period from a commercial service named Flightradar24, and a noncommercial crowdsourcing platform called OpenSky. Because LTO cycle emissions are related to TIM and fuel flow, the ICAO simple B method can be adopted. Emissions during the CCD phase are relevant to aircraft performance, including thrust, fuel consumption, and environmental conditions, so they can be calculated based on the ICAO advanced method. Nonvolatile particulate matter mass (nvPMm) and number (nvPMn) emission indices are obtained from the ICAO database when available; otherwise, they are estimated from the smoke number using the FOA4.0 method. If smoke number data cannot be acquired, the nvPMm emission indices for the LTO cycle and CCD phase are 0 and 30 mg/kg, respectively, based on the values suggested by the U.S. Federal Aviation Administration’s (FAA) emissions model and ignoring the nvPMn emission indices, as shown in Figure 5.

Through the analysis of emissions from 2017 to 2020, this case shows that the emissions in 2019 were the highest and decreased during the epidemic in 2020. Although the contribution rate of APUs to NO_x and fuel combustion emissions at the LTO stage is low, APUs are responsible for 34% of nvPM mass and 25% of nvPM number emissions, which are important contributors to low-altitude nvPM emissions [71]. This case examines the feasibility of open and ADS-B data in the generation of global emissions calculation, and it is concluded that ADS-B data can provide a more up-to-date emissions evaluation as well as a more transparent calculation process. This approach increases flexibility and makes it easier to quantify the influence of activities on emissions and monitor the progress of the aviation industry toward its sustainability goals in a transparent manner [72].

5. Limitation of the Study

This study presents a comprehensive methodology for compiling aviation emission inventories, yet due to a lack of related research and the limited scope of the article’s conception, it overlooks several practical challenges that may arise during the inventory compilation process. These challenges are of utmost importance and require resolution in the future development of aviation emission inventories. The key areas of concern are outlined below.

(1) Data diversity and accuracy: The compilation of aviation emission inventories relies on a multitude of data sources, encompassing flight data, aircraft type information, fuel efficiency data, and more. These data may originate from various channels, including airlines, airports, and international aviation organizations, posing challenges in data integration due to their diversity and format discrepancies. Furthermore, as data may contain errors or be incomplete, ensuring data integrity and accuracy becomes a significant hurdle in the compilation process.

(2) Model selection and parameter setting: Compiling aviation emission inventories necessitates the use of sophisticated mathematical models to estimate emissions, where the choice of model and setting of parameters exert a profound influence on the results. The task of selecting an appropriate model and configuring parameters reasonably presents a technical challenge.

(3) Increasing demands for spatial-temporal resolution: As understanding of aviation emissions deepens, there is a growing demand for higher spatial-temporal resolution in emission inventories. A significant challenge in the compilation process lies in enhancing the spatial-temporal resolution of emission inventories while ensuring data accuracy.

(4) Unique nature of high-altitude emissions: Aviation emissions primarily occur at high altitudes, exhibiting distinct characteristics compared to ground-level emissions. The dispersion and transformation processes of high-altitude emissions are more complex, and their impacts on the environment and climate differ accordingly. Accurately capturing the traits and impacts of high-altitude emissions in emission inventories poses a technical challenge.

(5) Security of aviation data: Due to security concerns and other reasons, actual flight data are currently non-public, encompassing multiple aspects such as flight safety, passenger privacy, commercial competition, and regulatory compliance. This poses a certain degree of obstacle to the further development of emission inventories.

6. Conclusions and Outlook

Along with the rapid development of the aviation industry, aircraft emissions are becoming a larger environmental concern. An inventory compilation of aviation emissions is conducive to analyzing emission characteristics, identifying emission sources, and evaluating the emission contributions of pollutants and greenhouse gases, which is beneficial for policymakers to formulate emission reduction policies and, in turn, mitigate climate change. In this paper, we compiled various methods for aviation emission inventories and the data collection of both emission parameters and activity levels in detail. The inventory calculation results suggest that a bottom-up approach can provide a higher level of accuracy relative to a top-down approach. Furthermore, accuracy also hinges on the volume of data input into the calculations. When compiling detailed emission inventories for specific airports, it is imperative to take into account data from aircraft main engines, APUs, and GSE in order to obtain high-resolution pollutant emission inventories. While LTO cycle emissions constitute a significant portion of aviation emissions, neglecting emissions from CCD phases and focusing solely on LTO emissions would result in discrepancies between the calculated outcomes and the actual contributions of aviation emissions.

Numerous studies have shown that a relationship exists between the relatively high uncertainty and the calculation processes of aviation emission inventories due to the high difficulty of actual data acquisition, especially for pollutants emitted during the CCD cruise phase. To establish accurate emissions inventories, along with further improving the accuracy and credibility of aviation emission evaluations, we suggest that steps should be taken to make full use of QAR data, ADS-B data, and actual data from the entire flight process. Moreover, we also demonstrated via some cases of aviation emissions that the ICAO method, especially the advanced method, is widely used in the compilation of aviation emission inventories because it takes into account aircraft performance and meteorological conditions. Currently, the acquisition of aviation emission and activity level data still faces a series of challenges. In the future, it will be necessary to establish a data-sharing mechanism across departments and institutions. At the same time, it is important to accumulate a large amount of real aviation emission data to form a diversified and localized emission factor database. Additionally, there is a need to further enhance the accuracy of real-time aviation data acquisition, reduce errors from monitoring equipment, and improve resistance to environmental interference.

In view of the shortcomings of the compilation methods for existing emission inventories, we propose further development of inventory technology for aviation emissions and greater aviation emission control in the future. Firstly, it is very necessary to establish a compilation method for aviation emissions inventory, which should be suitable for the local emission status. Secondly, an accurate and referenceable emission index should be obtained according to the actual emission data of each airport, and an emission index library should be developed to provide a reference for policy formulation regarding airport emission reductions. Furthermore, the calculation scope should be extended to airport clusters, an emission index library of airport clusters should be established, and data support should be provided for the implementation of environmental measures in designated areas. Thirdly, it is expected that newly developed equipment will overcome the shortcomings of QAR and ADS-B systems, make flight data easier to obtain, promote the accuracy of aviation emissions inventories, and facilitate location-specific policy formulation. Finally, for security and other reasons, the actual flight data are not public at present, which, to some extent, hinders the further development of emission inventories. Cross-departmental coordination mechanisms need to be established to ensure the availability of activity level data.