EXPERIMENTAL AND MODELLING APPROACHES TO INVESTIGATE OPTICAL PROPERTIES OF ATMOSPHERIC AEROSOL

The aerosol scientific community has a strong interest in improving knowledge about aerosol physical-chemical properties and measurement methodologies due to the large uncertainties still affecting estimates of aerosol optical properties and their impact on climate, visibility, and air quality. Therefore, in this PhD work, the contribution to this research field was mainly devoted to the study and implementation of experimental and modelling approaches aiming at filling some gaps in the current knowledge about absorption and scattering properties of atmospheric aerosols. On-line and off-line instrumentation was employed to investigate aerosol optical properties. The set-up and operation of the polar photometer developed by the Environmental Physics research group of the University of Milan (PP_UniMI) were optimised for the analyses of several filter samples of various types and sizes, collected during different campaigns also in the frame of collaborative projects. Several approaches were developed and exploited to investigate biases affecting outputs of widespread on-line instrumentation; truncation error correction in scattering measurements was deeply examined and in-situ techniques to measure aerosol absorption coefficient were analysed and compared to filter-based methods (both on-line and off-line), obtaining results that could be of large interest for the scientific community and deserve further investigation. The application of the widespread IMPROVE algorithm - used to estimate light extinction - at polluted urban sites was investigated, and a tailored approach was developed for typical characteristics of wintertime aerosol in Milan (Italy). The aerosol extinction coefficient reconstructed with this model for a PM1 dataset was apportioned among emission sources identified applying a receptor model. High-time resolved multi-wavelength aerosol optical properties measured in Rome (Italy) during the collaborative project CARE (Carbonaceous Aerosol in Rome and Environs) were investigated to find out an original methodology to distinguish aerosol types based on the combination of several intensive optical parameters. This approach could be useful in monitoring networks to discriminate in near-real-time between pollution vs. natural sources-driven high PM events.