Studies on the effects of CO2 emissions from aircraft on radiative forcing require only a knowledge of total emissions. However, to examine the potential effects of other emissions from aviation (e.g., those considered in Chapter 4), estimates of the amount and the distribution of emissions are required. Such 3-D inventories for present and projected future aviation operations have been produced under the aegis of NASA's Atmospheric Effects of Aviation Project (AEAP), the European Civil Aviation Conference's ANCAT and EC Emissions Inventory Database Group (EIDG), and DLR.
Figure 9-4: Comparison of growth rates for civil
Figure 9-5: Aircraft emissions inventory calculation
Figure 9-6: Scheduled aircraft mission profile.
Figure 9-7: Passenger traffic demand growth to 2015.
These inventories consist of calculated aircraft emissions distributed over the world's airspace by latitude, longitude,and altitude. Historical inventories of aviation emissions have been produced for 1976 and 1984 by NASA. Present-day and 2015 forecast inventories (where present-day is taken to be the most recent available-1991-92) have been produced by NASA, ANCAT, and DLR. DLR has also produced emissions inventories of scheduled international aviation only for each year from 1982 through 1992, and for total scheduled aviation for 1986 and 1989. DLR has also constructed a four-dimensional (4-D) inventory with diurnal cycles for scheduled aviation in March 1992.
All of the aforementioned 3-D emissions inventories have a common approach of combining a database of global air traffic (fleet mix, city-pairs served, and flight frequencies) with a set of assumptions about flight operations (flightprofiles and routing) and a method to calculate altitude-dependent emissions of aircraft/engine combinations in the fleet. Figure 9-5 shows how these processes are combined.
ll of the historical, present-day, and 2015 forecast inventories considered in this section assume idealized flight routings and profiles, with no winds or system delays. Thus, minimum fuel burn and emissions possible for each flight operation are implicit, given the onboard load assumed. Simplifying assumptions for military operations vary according to aircraft type.
The NASA, ANCAT, and DLR 3-D inventories adopt a similar overall approach but differ in some of the components and data used. This section describes thecommon approaches and explains the differences. More detailed information appears in the source material for these inventories (Baughcum et al., 1996a,b; Schmitt and Brunner, 1997; Gardner, 1998).
All of the inventories use a "bottom-up" approach in which an aircraft movement database was compiled, aircraft/engine combinations in operation were identified (to differing levels of detail), and calculations of fuel burned and emissions along great-circle paths between cities were made. Flight operation data were calculated as the number of departures for each city pair by aircraft and engine type-which, combined with performance and emissions data, gave fuel burned and emissions by altitude along each route. This approach resulted in data on fuel burned and emissions of NOx (as NO2) on a 3-D grid for each flight. In addition, the NASA inventories provide 3-D distributions of CO and total HC. NASA and ANCAT inventories were calculated on a 1� longitude x 1� latitude x 1-km altitude resolution, whereas the DLR inventory used a 2.8� longitude x 2.8� latitude horizontal resolution. Different approaches were taken for constructing underlying traffic movements databases. The NASA inventories use scheduled jet and turboprop aviation operations for the years 1976, 1984, and 1992 (Baughcum et al., 1996a,b). Movements for charter carriers, military operations, general aviation, and the domestic fleets of the former Soviet Union (FSU) and the People's Republic of China were estimated separately (Landau et al., 1994; Metwally, 1995; Mortlock and Van Alstyne, 1998). Military aircraft contributions to emissions were calculated by estimating the flight activity of each type of military aircraft by country. The 1976 and 1984 NASA inventories were based on operations for 1 month in each quarter of the year, whereas the 1992 inventory compiled movements on a monthly basis to reflect the seasonality of aviation operations.
The ANCAT approach used a combination of air traffic control (ATC) data and scheduled movements, favoring ATC data where available (Gardner, 1998). Where ATC data were unavailable, scheduled data were taken from the ABC Travel Guide (ABC), the Official Airline Guide (OAG), the Aeroflot time table, and a German study of Chinese domestic aircraft movements. Only jet aircraft were represented in the ANCAT/EC2 inventory. The most significant omission of ATC data was the United States, for which data were unavailable for security reasons. Thus, only time table data were used for the United States; so nonscheduled U.S. domestic charters and other flights were not recorded. To compensate for this problem, fuel usage data were factored up by 10% (Gardner, 1998). ATC data accounted for half of the non-U.S. aircraft movements in the database. Military movements were estimated by allocating fuel and emissions to countries' boundaries from an analysis of the world's military fleet composition.
The DLR inventory for 1991/92 (Schmitt and Brunner, 1997) used the ANCAT/EC2 civil movements database.Emissions inventories for 1986, 1989, and 1992 were based on scheduled air traffic only; a 4-D inventory with diurnal cycles for March 1992 was based on ABC data. ICAO data (ICAO, 1997b) were used for emissions inventories for international (only) scheduled air traffic in the years 1982 to 1992.
Calculation of fuel burned and emissions for aircraft differs between the three inventories. NASA used detailedmanufacturers' proprietary performance information on each aircraft-engine combination and the flight profile shown in Figure 9-6. Emissions were calculated from the information in the ICAO Engine Exhaust Emissions Data Bank (ICAO, 1995), through the use of Boeing "Method 2" procedures (Baughcum et al., 1996b, Appendix D), which allow extrapolation of sea-level data in the ICAO data bank to the operating altitudes and temperatures encountered throughout the aircraft flight profile.
The ANCAT/EC2 inventory used commercial software for flight and fuel profiling, along with Project Interactive Analysis and Optimization (PIANO), a parametric aircraft design model. The global civil fleet was modeled with a selection of 20 representative aircraft types. These representative aircraft were assumed to be fitted with generic engines typical of the technology and thrust requirements of each type. PIANO generated fuel profiles covering the entire flight cycle, including steps in cruise for each aircraft. Fuel use during ground operations was estimated from ICAO certification timings (ICAO, 1993).
The DLR inventory used airline data and an in-house flight and fuel profile model (Deidewig et al., 1996). The DLR approach also used different aircraft/engine combinations from those utilized by ANCAT. The aircraft mission was simulated by using a simplified flight modeling code as point-to-point missions with no step cruise. Although the climb was calculated in iterative steps, the cruise segment was treated as one section, applying the Breguet formula to calculate the cruise fuel. Descent was assumed to be a gliding path with minimum engine load; no separate approach procedure was used. A thermodynamic model for design and off-design operation of a two-shaft fan engine was applied. Constant efficiencies and constant relative pressure losses for main engine components were assumed for simplicity.
The ANCAT/EC2 and DLR inventories calculated NOx emissions from the fuel using the DLR fuel flow method. This method has been tested and correlated with information from airlines, flight measurements, and altitude chamber measurements (Deidewig et al., 1996; Schulte et al., 1997).
Other reports in this collection