Determination of uncertainty in environmental noise measurements by bootstrap method

An effective method for real-time evaluation of confidence intervals associated to quantile (L-q) and equivalent (L-eq) levels in environmental noise measurements is presented. The non-parametric surrogate data (or bootstrap) method, is described here in its basic form, valid for independent and identically distributed data, but is readily extendible to the treatment of dependent data. Application to actual measurements are shown which illustrate the practical effectiveness of real-time error evaluation during environmental monitoring. (C) 2003 Elsevier Ltd. All rights reserved.