A study led by the Barcelona Institute for Global Health (ISGlobal) has consistently estimated daily ambient concentrations of PM2.5, PM10, NO2 and O3 across a large ensemble of European regions between 2003 and 2019 based on machine learning techniques. The aim was to assess the occurrence of days exceeding the 2021 guidelines of the World Health Organization (WHO) for one or multiple pollutants, referred to as "unclean air days."