Theoretical prediction of highway noise fluctuations.

Abstract

A theoretical model is developed to predict fluctuations in traffic noise due to statistical variation in properties of highway traffic, including differences in vehicle type, variations in individual vehicle emissions, clustering of vehicles by vehicle type, and variations in vehicle spacings within and between platoons. Traffic flow is described by a Markov renewal process (MRP), and the noise heard by an off-highway observer as a filtered MRP. Simple formulas for the arithmetic mean, variance, and covariance function of noise intensity are derived for the "shot noise" case (exponentially distributed spacings). A more complicated model with car and truck clustering is investigated numerically, based on traffic data given by Jewell for the southbound curb lane of U.S. 40 in Berkeley, California. Even with only a modest degree of truck clustering and variability in vehicle spacings and noise emission, the variation in noise level is much greater than usually predicted. The coefficient of variation of noise intensity is greater than unity at distances from the roadway less than 700 ft. Peak noise levels may thus be greatly underestimated (2 to 3 dB per highway lane) if these sources of variability are ignored.