Air pollution represents a global threat leading to large impacts on health and ecosystems and many European
areas still show a poor air quality. Air quality forecasts are instruments, which may support authorities and
citizens in reducing these impacts. This work presents the development and the performance evaluation of an air
quality forecast system for Italy: FORAIR- IT. It operates at both European and Italian scales at 20 km and 4 km
resolution, respectively. The forecasts’ performances are evaluated against measurements, through a comparison
with CAMS_50 validated reanalysis ensemble and by an in depth evaluation during a long lasting PM10 episode,
which occurred in Emilia Romagna region in October 2017. The results show similar skill scores between
FORAIT-IT and CAMS_50 for PM10, PM2.5, NO2 and O3 concentrations. The increase of model resolution over
Italy and the use of the national emission inventory lead to a better reproduction of pollutant concentrations at
urban and suburban sites for PM10, PM2.5 and NO2. The analysis of the pollution episode shows that the correct
forecast of the meteorological forcing is fundamental in capturing both the timing and the occurrence of the
exceedances, while the poor performance of the model in reproducing the magnitude of the exceedances may be
due to the lack of information on some emissions sources such as agricultural biomass burning.

cited By 0