The use of stratification to improve the design efficiency of community noise surveys

A community noise survey was executed in the Philadelphia area. Temporal and spatial stratification was utilized in both the design and analysis of the monitoring. Measurements of the energy equivalent noise level L eq, the L x values (noise level exceeded x% of the time), the noise pollution level L NP, and traffic noise index (TNI) were obtained. Analysis of variance was applied to the data, revealing large differences between the variances of several noise descriptors at most levels of stratification. Recommendations are advanced for: (a) stratification of noise monitoring; (b) use of a presurvey to guide stratification; (c) analysis of partial results to improve monitoring efficiency; and (d) measurements of background noise level. It is suggested that neighborhood‐averaged L 99 may be a good predictor for neighborhood‐averaged L eq.