Evaluation of air quality forecasting system FORAIR_IT over Europe and Italy at high resolution for year 2017

Abstract. Air pollution represents a global threat leading to large impacts on health and ecosystems and many European areas still show a poor air quality. Many measures and policies have been adopted in the past decades at European, national, regional and even local level and many tools have been developed to tackle this issue. Among these tools, the European Air Quality Directive places more emphasis on the use of models for air quality assessment and management. Within this context, air quality forecasting systems play an important role in supporting decision makers when short-term actions are required to reduce human health risks by limiting population exposure. In this framework, at European level within the Copernicus Atmosphere Monitoring Service (CAMS), the regional air quality models participating to the service provide 4-day daily forecasts of the main atmospheric pollutants concentrations, in the lowest layers of the atmosphere. This work presents the development and the performances evaluation of FORAIR_IT, an high-resolution air quality forecasting system operating at both European (20 km) and Italian (4 km) scales. Its skill results are compared with CAMS_50 interim ensemble reanalysis (IRA) ones and a long lasting PM10 exceeding event, occurred in Emilia Romagna region in October 2017, is studied in more detail. Results show similar skill scores between FORAIT_IT and CAMS_50. Comparing the annual average of the monthly Root Mean Square Error Difference (RMSED) between FORAIT_IT first forecast day and CAMS_50, over the European domain the RMSED is 0.6, 1.7, 1.4 and 7.4 μg/m³ for daily mean PM10 and PM2.5 and daily maximum for NO₂ and O₃, respectively, while over the Italian domain it is  1.2, 0.3,  4.3 and 3.8 μg/m³. The importance of increasing model resolution in the region of interest is highlighted by the lower values of RMSED over Italy with respect to Europe. The results obtained by the detailed analysis of the PM10 exceeding event suggests the crucial role of the meteorological forcing in capturing both the timing and the intensity of the exceedances. As far as we know FORAIR_IT is the first forecasting system at high spatial resolution at Italian National level.