Number of Residents Exposed to Significant Aircraft Noise

Data Scope

The count of the people exposed to aircraft noise around airports is obtained by computer modeling and not by any census. For almost 30 years, the Federal Aviation Administration's (FAA's) Integrated Noise Model (INM) has been the pre-eminent tool for assessing aircraft noise around airports¹. Using information on aircraft mix, average daily operations, flight tracks, and runway distribution, INM can generate and plot contours of Day Night Sound Level (DNL). According to Federal guidance, levels at or above DNL 65 decibels (dB) are considered significant. With the addition of a digitized population census database, INM can produce the number of residents exposed to DNL 65 and higher. Airport authorities make use of INM to generate noise exposure maps when applying for Federal funding of noise compatibility programs or major construction projects. For major airports, a noise exposure study could cost several hundred thousand dollars.

For as long as there have been predictions of airport noise exposure, the FAA has been interested in the number of people exposed to aircraft noise nationwide. The Environmental Protection Agency (EPA) produced the first estimate when their analysis found that in 1975, 7 million residents are exposed to significant levels of aircraft noise. That number became the "anchor point" for future estimates of the nationwide impact. In 1980 during the development of the rulemaking to "phase-out" the operations of the noisier Stage 1 airplanes, the FAA found it necessary to identify the benefits of the proposed regulation. As it was logistically and economically impossible to calculate the number of people impacted at each U.S. airport, the FAA developed a crude methodology to estimate the change in number of people impacted by noise (from the 1975 anchor value) as a function of changes in both the national fleet and in the FAA's Terminal Area Forecast (TAF). Again in 1990, the FAA created an improved but still simplistic method of estimating the change in number of people impacted (relative to the 1980 estimates) for what became the regulation to phase-out operations of Stage 2 airplanes by the end of 1999.

In 1993, the FAA finished the development of the Nationwide Airport Noise Impact Model (NANIM)². NANIM was an analytical tool for estimating the impact of airplane noise on residential communities surrounding U.S. airports that support jet operations. NANIM calculated the regional and national totals of the number of people, the land area, and the number of housing units exposed to DNL 65 dB or higher. Based upon the limitations of NANIM, the FAA used this model to determine the relative changes in numbers of people and land area exposed to DNL 65 as the result of changes in nationwide aircraft fleet mix and operations. The FAA last used NANIM in 1998 to predict the effect of new information on numbers of "hushkitted" airplanes in the US commercial jet fleet (relative to the 1990 estimates). The finding of that study became the noise exposure measure in the Department of Transportation (DOT) performance report for the Government Performance and Results Act (GPRA).

A little over four years ago, the FAA initiated a project to collect airport noise analysis databases for a large number of the world's airports. This sample database of airports would be the basis for assessing worldwide trends that would occur as the result of stringency, different land-use planning initiatives and operational procedures. The objective was to develop a tool that could be used by the Committee on Aviation Environmental Protection (CAEP) under the International Civil Aviation Organization (ICAO). Previous attempts by CAEP to globally assess aircraft noise exposure had limited success. The proposed FAA methodology had much more promise, as the number of sample databases was large and has since grown to around 200. Furthermore, a generalized methodology was included to account for airports for which noise databases did not exist. Based on the initial success of the FAA activity, the fourth meeting of CAEP (CAEP4) recommended that a task group be formed to complete the development of this tool for CAEP analysis. This group and subsequently the model became known as MAGENTA (Model for Assessing Global Exposure form Noise of Transport Airplanes).

The MAGENTA task group is comprised of CAEP participants including other government regulatory groups and industry participants. It has met several times since CAEP4 to complete the data collection and data review necessary to utilize MAGENTA for CAEP purposes. Issues include adoption of a policy metric, review of processing assumptions and coordination on data exchange among the various other CAEP task and working groups.

Model Data

Documentation Certain details of the MAGENTA model are still under development. This development is intended to support CAEP stringency analysis that must be completed by September 2000. A formal User's Guide and Reference manual have not been produced. The documentation of MAGENTA is contained in ICAO working papers, which are updated every few months. There was a comprehensive description of the model produced and delivered to NASA/FAA for CAEP4. At that time, the model was called WANIM (World Aviation Noise Impact Model).

Structure The MAGENTA system makes use of several databases. They are:

• Population Data
  For the United States, population levels and distribution come from the 1990 US Census database. Other data is developed from the Joint Resources Assessment Database System (JRADS), which is a worldwide population database that includes population estimates for all major cities in 130 countries. Special data collection was used for airports not covered by JRADS including generalized population densities.
• Traffic Distribution
  Major assumptions on traffic patterns come from obtaining INM datasets that were developed for an airport. If an INM database did not exist, the contractor developed one. These databases represent baseline conditions from 1990-1997.
• Aircraft Distribution
  The INM database contains a default assignment of aircraft to routes. These are updated based on the International Official Airline Guide (IOAG) and an aircraft registration database. The IOAG gives general traffic distributions including a time of day parameter, which is important for the DNL metric. This distribution is then "refined" using information from an aircraft registration database. The original aircraft registration database used for MAGENTA was ACAS³. For the current system which is the one used to develop US trends and the final ICAO stringency analysis, aircraft specific information comes from a database developed by the Air Transport Association. It is referred to in MAGENTA working paper documentation as the Campbell-Hill database.
• Forecast Data
  Traffic growth, including traffic levels, and aircraft seat size class, come from the CAEP Forecasting and Economic Support Group (FESG). This involves ICAO route group growth factors that must be applied to individual airports.

Assumptions and Parameter Estimation The model begins with a 1998 baseline developed using the databases described above. The model steps though time retiring and adding aircraft based on a set of rules developed by the ICAO task groups. These groups develop rules for how long an aircraft stays in service and when it is "retired", they provide the aircraft with which it will be replaced. ICAO also developed assumptions on traffic growth for regional analysis. The system contains a method for taking regional growth assumptions and distributing them to the specific airports that compose the MAGENTA sample database.

The MAGENTA system contains a series of programs that link all of the data described in the previous sections together (see Figure 1). A combined input file is then sent to the noise calculation engine (INM) and the resulting contours and population data are loaded into a GIS system.

Peer Review and Evaluation Studies There has been an active CAEP task group that has reviewed the progress and development of MAGENTA. This includes comparisons to locally provided population and contour data. Up to the present, these have addressed the individual MAGENTA components. The item most reviewed and tested was the MAGENTA equivalency processor. This consists of the MAGENTA component that approximates the shapes of hundreds of individual aircraft noise footprints with a reduced set of data and is referred to in MAGENTA working papers as the aircraft equivalency processor.

Non-sampling Errors

Coding/Recording Error There exist the possibility of coding error in each of the individual MAGENTA components (Figure 1). This includes actual program logic to the programming of the "rules". In one case a program may incorrectly assign a number of operations to an airport based on a programming logic error. In the later, the program logic is correct, however, the rule was input incorrectly. As resources allow, the CAEP MAGENTA task group has made efforts to anticipate these problems and correct them through validation studies.

Non-coverage Error MAGENTA data coverage is very comprehensive in terms of its breadth. Databases have been obtained or developed for all airports, domestic and internationally that would have any substantial effect on trend. The coverage issue that could introduce error would include the age of the databases and the appropriateness of applying broad assumptions to individual airports.

The majority of the MAGENTA databases were collected prior to CAEP4. They range in age from 1990 to 1997 with many of the base cases now reaching six years old. It is an assumption that for significantly impacted areas, the runway locations and utilizations used in the older base case remain relevant today. For domestic airports, the population data is for 1990. It is an assumption that the population numbers and distribution around an airport remain unchanged. As US Census data contains the best practical source of information, this could be confirmed or when the 2000 population information is released.

The data does not contain individual local traffic growth forecast data. Individual fleet mix and traffic level is obtained by applying broad growth assumptions to individual airports. This may cause some error in airport-to-airport comparisons and it is an assumption that this type of error will not translate to a bias in the overall national numbers. After completion of the activity for CAEP, FAA will explore substituting the TAF for the ICAO forecast.

¹ Integrated Noise Model (INM) Version 6.0 User's Guide, Report No. FAA-AEE-99-03, September 1999.
² NANIM User's Guide and Methodology Report, prepared for the FAA by the Washington Consulting Group, Contract DTFA01-89-Y-01017, October 1993.
³ ACAS (AirCraft Analytical System) is a leading aviation market information system for Windows PCs covering. commercial aircraft, military transports, business jets, and Russian built aircraft. The database includes principal details on operator, engine, ownership, hours/cycles, history, orders, availability, values, contact names and addresses, maintenance capability and inspection intervals. The product is commercially available through AvSoft Ltd. of Rugby Warwickshire, England.