Extreme Compound Events of Air Pollution and Effects on European Mediterranean Cities

Air pollution is a significant concern for society due to its negative impact on human health. The ability to forecast pollutant concentrations and predict extreme pollution events is essential for mitigating their harmful effects. These events are commonly presented as compound and can have serious consequences (Zscheischler et al., 2020) for public health, making it important to identify and understand the variables that contribute to them.

In this study, we used state-of-the-art techniques based on artificial intelligence and machine learning methodologies to analyze data from a selection of European Mediterranean cities in order to identify the meteorlogical and preceding pollution variables that best explain extreme pollution events involving PM10, O₃and NO₂. Furtheremore, the role of the combined effect of these variables and recent climatic conditions is also examined on excess mortality rates using the likelihood multiplication factor parameter (Ridder et al., 2020) as well as the importance given by Random Forest Regression  Models.

Our results show that these events show a clear compound nature, understanding the non-linearity changes in the intensity of compound events. The main driver factors depend on the pollutant specie as well as the season of the year.

 

References
Ridder N. N., Pitman A. J., Westra S., Ukkola A., Do H. X., Bador M., Hirsch A. L., Evans J. P., Di Luca A., Zscheischler J. (2020). Global hotspots for the occurrence of compound events. Nature communications 11(1), 1–10.
Zscheischler J., Martius O., Westra S., Bevacqua E., Raymond C., Horton R. M., van den Hurk B., AghaKouchak A., Jézéquel A., Mahecha M. D., et al. (2020). A typology of compound weather and climate events. Nature reviews earth & environment 1(7), 333–347.