Multiplex Visibility Graphs as a complementary tool for describing the relation between ground level O3 and No2

Özet

The usage of multilayer complex networks for the analysis of correlations among environmental variables (suchas O3and NO2concentrations from the photochemical smog) is investigated in this work. The mentioned techniqueis called Multiplex Visibility Graphs (MVG). By performing the joint analysis of those layers, the parametersnamed Average Edge Overlap and Interlayer Mutual Information are extracted, which accounts for the microscopicaltime coherence and the correlation between the time series behavior, respectively.These parameters point to the possibility of using them independently to describe the correlation betweenatmospheric pollutants (which could be extended to environmental time series). More precisely the first one ofthem is considered to be a potential new approach to determine the time required for the correlation of NO2andO3to be observed, since it is obtained from the correlation of the pollutants at the smallest time scale. As for thesecond one, it has been checked that the proposed technique can be used to describe the variation of the correlationbetween the two gases along the seasons. In short, MVGs parameters are introduced and results showthat they could be potentially used in a future for correlation studies, supplementing already existing techniques.