A complex aerosol transport event over Europe during the 2017 Storm Ophelia in CAMS forecast systems: analysis and evaluation

Abstract

In mid-October 2017 Storm Ophelia crossed over western coastal Europe, inducing the combined transport of Saharan dust and Iberian biomass burning aerosols over several European areas. In this study we assess the performance of the Copernicus Atmosphere Monitoring Service (CAMS) forecast systems during this complex aerosol transport event and the potential benefits that data assimilation and regional models could bring. To this end, CAMS global and regional forecast data are analysed and compared against observations from passive (MODIS: Moderate Resolution Imaging Spectroradiometer aboard Terra and Aqua) and active (CALIOP/CALIPSO: Cloud-Aerosol LIdar with Orthogonal Polarization aboard Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations) satellite sensors and ground-based measurements (EMEP: European Monitoring and Evaluation Programme). The analysis of the CAMS global forecast indicates that dust and smoke aerosols, discretely located on the warm and cold fronts of Ophelia, respectively, were affecting the aerosol atmospheric composition over Europe during the passage of the Storm. The observed MODIS aerosol optical depth (AOD) values are satisfactorily reproduced by the CAMS global forecast system, with a correlation coefficient of 0.77 and a fractional gross error (FGE) of 0.4. The comparison with a CAMS global control simulation not including data assimilation indicates a significant improvement in the bias due to data assimilation implementation, as the FGE decreases by 32 %. The qualitative evaluation of the IFS (Integrated Forecast System) dominant-aerosol type and location against the CALIPSO observations overall reveals a good agreement. Regarding the footprint on air quality, both CAMS global and regional forecast systems are generally able to reproduce the observed signal of increase in surface particulate matter concentrations. The regional component performs better in terms of bias and temporal variability, with the correlation deteriorating over forecast time. Yet, both products exhibit inconsistencies on the quantitative and temporal representation of the observed surface particulate matter enhancements, stressing the need for further development of the air quality forecast systems for even more accurate and timely support of citizens and policy-makers.