Predicting aircraft noise annoyance from single flyby events using acoustical and psychoacoustic metrics

The increasing volume of air traffic has resulted in a rising number of individuals exposed to aircraft noise, which is associated with various adverse health effects, with noise annoyance being among the most prevalent. This study identifies acoustic and psychoacoustic metrics best suited to predict noise annoyance from single aircraft flyby events. Data from four laboratory listening experiments involving 151 participants were re-analyzed for that purpose. The underlying experiments included synthesized and measured sound signals comprising a wide range of aircraft, specifically civil jet aircraft (current and future designs), helicopters, and propeller aircraft. Acoustic and psychoacoustic parameters, including effective perceived noise level, were computed to determine key predictors for annoyance. Our findings show that loudness-based indicators alone are generally adequate to predict noise annoyance effectively across diverse aircraft types. However, in the case of outliers regarding noise annoyance, adding psychoacoustic parameters, particularly tonality, roughness, and fluctuation strength, significantly improved prediction accuracy. Despite utilizing an older loudness model, the effective perceived noise level consistently exhibited excellent predictive performance, demonstrating its robustness as a baseline metric for assessing short-term annoyance from aircraft noise.