Noise indicators to diagnose urban sound environments at multiple spatial scales
Résumé
The diagnostic of urban sound environments requires noise indicators able to capture its main physical characteristics. However, the more information furnished by indicators, the longer the measurements should be and the lesser immediate their understanding is. In this paper, a methodology in three steps is proposed to diagnose urban sound environments at the neighborhood level, with an increasing level of detail
that offers some flexibility to the decision maker when investigating the sound quality of a neighborhood. The first step consists of the calculation of three sound structure indicators, namely the L50,A, the ÃLeq,A and the SGC[50Hz-10kHz], which allow a continuous spatial categorization of the neighborhood. The second step consists of the calculation of sound events indicators, namely the L1,A, the MILA50+10, and the
MILLF50+15, which are sensitive to the physical dimensions of noise emergences (threshold values, suddenness, occurrences). The third step consists of a map of the emergences, which describes precisely at a given location the number and the duration of the emergences. The procedure is validated over three measurement campaigns achieved in January, April and June, when geo-referenced noise measurements were collected over 18 1h-soundwalks periods in Toulouse (France). A clustering analysis is performed in order to select the subsets of indicators used to describe sound environments. Although these indicators might differ in theory from one location to another, the clustering analysis selects the same indicators than a previous
study achieved on another site, suggesting their possible generalization for further use. Moreover, the stability in the indicators values between the three different campaigns validates their calculation over short time samples. Finally, the procedure enables describing in details the time fluctuations of sound environments at both the daily and the seasonal scale. These fluctuations can be explained by the land use of the site.