Analysis of long-range transport of aerosols for Portugal using 3D Chemical Transport Model and satellite measurements

Abstract

The objective of this work is to assess the contribution of long-range transport of mineral dust from North Africa to the air pollution levels in Portugal based on a combination of a modelling approach and satellite observations. The Comprehensive Air Quality Model (CAMx) was applied together with the updated Dust REgional Atmospheric Model (BSC-DREAM8b) to characterise anthropogenic and natural sources of primary aerosols as well as secondary aerosols formation. The modelling results, after their validation and bias removing process, have been used in combination with aerosol measurements provided by Ozone Monitoring Instrument (OMI), using OMAERUV Level-2 v003 product, aiming to better understand the advantages and shortcomings of both, satellite and modelling aerosol data. The data analysis is presented for Portugal for July 2006 focusing on aerosol optical depth (AOD) at 500 nm and aerosol type. Based on the modelling results, the importance of the long-range transport of mineral dust was demonstrated for the simulation days, achieving a 60% contribution to AOD levels. The mineral dust is affecting atmospheric layers up to 6 km but peak concentrations are presented at layers below 2 km. The model predicts a complex mixture of different types of aerosol for the pixels classified by OMI as “mineral dust” and “sulphates”. Although a good agreement between the model outputs and OMI observations has been found in terms of the spatial pattern and AOD correlation is about 0.48 for mineral dust, several problems were identified. The model is systematically underestimating the aerosol concentration at near ground level in comparison with the air quality monitoring stations, while OMI is in general overestimating AOD for the analysed period based on the comparison with AERONET data. Additionally, misclassification of mineral dust for some geographical locations and discontinuity in AOD values along the coastal line at water/land interface in the OMI data are discussed.