Tailored coefficients in the algorithm to assess reconstructed light extinction at urban sites : a comparison with the IMPROVE revised approach

Estimates of light extinction and visibility are routinely performed by the U.S. Interagency Monitoring of Protected Visual Environments (IMPROVE) network using a simple algorithm which assesses light extinction coefficient (bext) at remote and rural sites from concentrations of major particulate matter (PM) species, NO2, and Rayleigh scattering from clear-air gaseous components. Following the same approach, in this paper an equation with tailored (i.e. site-specific) coefficients was implemented with the aim of reducing uncertainties and assumptions of the IMPROVE algorithm for applications at polluted urban sites. Major differences compared to IMPROVE algorithm are: 1) dry mass extinction efficiencies calculated using a discrete dipole approximation code with aerosol size distributions measured at our monitoring site as input data; 2) site-specific water growth functions computed separately for ammonium sulfate, ammonium nitrate, and organic matter; 3) fine soil evaluated using an equation previously adopted at our urban site; 4) aerosol absorption component assessed directly through filter-based optical measurements. Details about the calculations performed are reported in the paper and the comparison with the IMPROVE revised algorithm is discussed. The tailored approach here proposed to estimate reconstructed light extinction was applied to PM2.5 test samples collected on purpose in Milan (Italy), where heavy pollution episodes occur during winter periods. In addition, visual range was calculated applying the Koschmieder equation and compared to visibility measured at the nearby Milano-Linate airport obtaining a fairly good correlation.