Resum

Challenge 3 of PIKSEL project aims to design objectives tools to help air quality managers in a given region to predict the incidence of episodes when tropospheric ozone immission concentration exceeds the legal concentration limits of ozone.

This tool is based on the use of data analysis models, based on Machine Learning (ML) techniques. The models developed are fed by the series of data from the Catalan Meteorological monitoring network (METEOCAT) and the Air Quality monitoring network (XVPCA) distributed throughout the territory of Catalonia. Moreover, a synoptic classification (of types of Synoptic maps) is being developed in this project from the reanalysis maps distributed by ERA 5, introducing thus the synoptic forcing in the developed models.

Based on these data, we are building the models to predict two parameters of special relevance at the time of monitoring risk levels for atmospheric pollution: the maximum hourly O3 and the 'maximum daily 8 h concentration. These models will be elaborated thorough all the relevant stations of air quality, and later the results will be extrapolated to the level of the whole of Catalonia.

Learn more at:

CIMNE Coffee Talk (PIKSEL Special Series): Challenge 3

Resum

Air pollution in urban areas has become a topic of major concern for all relevant institutions. It is well established that living in areas with high levels of air contaminants is linked to adverse health effects. The concentration of a pollutant like NO2 is closely related to traffic emissions, and thus city halls can typically reverse unfavorable air situations by implementing traffic restrictions. This work intends to design, develop and validate a numerical tool that can be used to predict and quantify the concentration of NO2 in urban areas at a micro-scale level. Such a tool is designed to be both accurate and fast because its final goal it to inform decision-making about the suitability of implementing proper traffic restrictions in a daily basis.