Streamlining Environmental Monitoring Networks: Application To Nitrogen Dioxide In North Italy

Abstract

This article presents a methodology for streamlining a monitoring network. The approach shows that risk maps, which provide decision-makers with the probabilities of a pollutant to be above a regulatory threshold, can also be used to identify monitoring stations that are providing redundant information. The method is applied to an air quality network of 102 stations monitoring nitrogen dioxide (NO2) concentrations in north Italy. The limitations of the network have first been identified in order to specify the annual concentration levels for which reliable risk maps can be designed. This has been achieved by a fractal analysis of the spatial correlation for several NO2 concentration levels. Two predefined annual thresholds at 40 and 50 µg/m3 were used as reference to investigate the possible redundancy of the information provided by the existing network. It was shown that the number of stations could be reduced by almost 20% without losing any information with respect to describing the annual pollution levels.