Comparison of dispersion models by using fuzzy similarity relations

Abstract

Aim of this work is to introduce a methodology, based on the combination of multiple temporal hierarchical agglomerations, for model comparisons in a multi-model ensemble context. We take advantage of a mechanism in which hierarchical agglomerations can easily combined by using a transitive consensus matrix. The hierarchical agglomerations make use of fuzzy similarity relations based on a generalized Łukasiewicz structure. The methodology is adopted to analyze data from a multi-model air quality ensemble system. The models are operational long-range transport and dispersion models used for the real-time simulation of pollutant dispersion or the accidental release of radioactive nuclides in the atmosphere. We apply the described methodology to agglomerate and to individuate the models that characterize the predicted atmospheric pollutants from the ETEX-1 experiment. © 2011 Springer-Verlag Berlin Heidelberg.