A Comparative Analysis of Annoyance From Aircraft Noise At Three European Airports
A Bristow, R Batley, M Wardman, ITS, University of Leeds; C Heaver, FaberMaunsell; T Elliff, EUROCONTROL, UK
This paper reports results from the 5A project (Attitudes to Aircraft Annoyance Around Airports) funded by EUROCONTROL, the European organisation for the safety of air navigation. The aim of the project was to gain a greater understanding of how residents perceive and value annoyance from aircraft noise. As the EU moves towards a standardisation of noise measures and with the underlying principle of common standards across the community, it becomes increasingly important to understand variation in annoyance levels within and between countries. This paper focuses on the findings from a social survey of the factors that influence annoyance; the aim was to determine attitudes to aircraft and other forms of noise, and identify the factors that determine behavioural response to noise.
Annoyance is a function of perceived noise which is itself a function of actual noise. These relationships are modified by a range of factors relating to: quality of life, other sources of noise, exposure to noise and activity patterns, area type, country, personal and socio-economic characteristics and confounding issues. The results presented in this paper include a discussion of descriptive statistics and cross tabulations and the development of models to explain variations in annoyance levels.
The survey work took place at the end of 2002 around the three airports at Bucharest, Lyon and Manchester, achieving a total sample size of 647 respondents (with at least 200 respondents at each airport). A common methodology was used at each airport with the aims of achieving a controlled comparison of attitudes and valuations across countries and permitting identification of any heterogeneity across locations. The design of the social survey was influenced by three key sources: 15 focus group analyses undertaken earlier in this study (5 at each airport site), the body of literature on variables that influence annoyance from noise, and guidelines on standardised noise reaction questions for use in community noise surveys. By drawing on earlier studies, we sought to ensure that no key influential variables were omitted and that, where appropriate, questions were asked in such a way as to permit comparison between studies. In accordance with guidelines on standardised noise reaction questions, responses to attitudinal questions were offered on 5-point scales.
The empirical findings include the following:
*The relative importance of aircraft noise against 15 other factors affecting local quality of life. The scale ranges from not at all important (1) to extremely important (5). A common feature across all three countries is the importance of crime and personal security. In Lyon there is an emphasis on local environmental issues and road traffic and air pollution are rated to be more important than aircraft noise. Air pollution is the top concern in Bucharest, where aircraft noise has the lowest importance rating. In Manchester noise is given a fairly low importance rating. These rating are modified according to specific location around the airport.
*Levels of dissatisfaction with the same set of quality of life issues. Aircraft noise was the aspect that caused most dissatisfaction in Lyon. In Manchester it rated third and lower down the list in Bucharest. Again ratings were dependent on the specific location.
*Aircraft noise was the most frequently noticed noise at Lyon and Manchester and the third most frequently noticed in Bucharest. Responses were influenced by specific location and distance from the airport, exposure to noise (time spent at home), the respondent?s view of the noisiness of the area and their own sensitivity to noise.
*There is a high degree of consistency between rating of the noisiness of noise sources and the frequency of noticing.
*This consistency in response carried over into responses on levels of annoyance from different noise sources. Annoyance levels appear to be influenced by such factors as specific location, double glazing, noisiness of the area, length of residence, exposure, sensitivity and income.
*The most severe disturbance from aircraft occurred when using the garden in both Lyon and Manchester. The ratings at these locations were much higher than for any other activity.
*Aircraft noise was perceived to be most disturbing during weekends, at night and during the evenings.
*Large 747 planes were consistently rated as more annoying than 737/airbus or equivalent planes, which in turn were rated more annoying than smaller turbo-prop aircraft. Respondents were also exposed to a noise simulation, and rated the aircraft involved.
*Confounding factors included easy access to the airport when flying, which is seen as a key benefit of living near an airport in both Lyon and Manchester. In Bucharest access to jobs is seen as the key advantage
As part of Phase 2 of the research project, the original data set was supplemented in late 2003 with noise measurements for the respondents? home locations. This permitted investigation of the relationship between annoyance levels, as perceived by respondents, and measurements of noise. In this phase of work, to be reported in March 2004, models are being developed to explain how different factors modify the way in which a given level of noise translates into a reported level of annoyance. The factors being considered are:
*Cultural and situational variables, such as country, area, settlement type and distance from the airport;
*Socio-economic characteristics such as age, gender, household composition, income, house type, tenure type and possibly exposure;
*Personal characteristics, such as membership of environmental organisations, use of the airport, job at airport, sensitivity to noise, and awareness of other risks;
*Time of day and day of week. Analysis involves an innovative application of advanced choice modelling. Since the relevant responses are distributed across a 5-point scale, the data are being applied to the estimation of an Ordered Generalised Extreme Value (OGEV) model.
Although OGEV was originally developed for the specific context of departure time choice, the key features that make it relevant to the present context are its ability to model ordered responses; in particular, its ability to accommodate correlation between contiguous points on the response scale.
Association for European Transport