Statistical tools for data optimization in air quality monitoring networks

In this paper, we present the application of statistical tools for data optimization in air quality monitoring networks, particularly analysing data correlation structure with multivariate statistical techniques and applying a method based on the Shannon index to evaluate the possible exclusion of monitoring stations or measured pollutants appearing as "the least informative". Our goal is the definition of a simple procedure for identifying the redundancy in air quality data sets. The procedure results may be useful both to evaluate effectiveness and efficiency of existing networks, and to select the data sub-sets more suitable for analysing, modelling and reporting AQM data.