Comparison of Dispersion Models by Using Fuzzy Similarity Relations

Aim of this work is to introduce a methodology based on the combination of multiple temporal hierarchical agglomerations by using fuzzy similarity relations for model comparison in a multi-model ensem- ble context. The similarity is a fuzzy relation and it is a generalization of the classic equivalence relation. In this work we consider a generalization of the fuzzy similarity considering a generalized Lukasiewicz structure. Moreover we introduce a mechanism in which multiple temporal hierar- chical agglomerations can be easily combined by using a transitive consensus matrix. The methodology is used to analyze data from a multi-model air quality ensemble system. The models are operational long-range trans- port and dispersion models used for the real-time simulation of pollutant dispersion or the accidental release of radioactive nuclides in the atmo- sphere. We apply the described methodology to agglomerate distributions of predicted atmospheric pollutants from the ETEX-1 experiment.

CIARAMELLA Angelo;  RICCIO Angelo;  GALMARINI Stefano;  GIUNTA Giulio;  POTEMPSKI Slawomir; 

2011-10-17

Springer

JRC65314

0302-9743 (print),    1611-3349 (online),   

https://publications.jrc.ec.europa.eu/repository/handle/JRC65314,   

10.1007/978-3-642-23954-0,