Author(s)

G. Alfaro Degan, D. Lippiello, M. Pinzari

Abstract

A geostatistical solution is presented to assess the acoustic impact of a quarry plant near Rome by drawing acoustic maps all around the selected site. The method utilised to determine the sound pressure level in the area of interest is the Intrinsic Random Function (IRF) Kriging with external drift. This technique allows the value of a regionalized variable to be estimated at a point in which it is unknown through the use of two sets of data. The first one is obtained by carrying out phonometric samples at various points near the site, while the second one is from a deterministic model of sound propagation defined in accordance with norms in force.

In order to test the quality of the estimates and especially the contribution of the external drift, acoustic maps on a local scale are drawn up both with and without drift. The quality of the results obtained is tested by means of cross validation.

The experimental results demonstrate that, providing it is strongly correlated to the first, the introduction of this secondary information, in a stochastic framework of reference, contributes to a more efficient representation of the spatial variability of sound propagation, and thus to an improvement in the quality of the correlated estimate. In the case study analysed here the introduction of drift permits a reduction of estimate variance, for the same number of available measurements. Again from the cross validation data and estimate variance it was observed that the use of external drift allows loss of information (e.g. due to a reduction in the monitored network) to be compensated for without invalidating the quality of the estimate itself.

Keywords

noise assessment, environmental impact, IRF Kriging with external drift, quarries