SPECTRAL EXTINCTION COEFFICIENTS OF RURAL AEROSOL IN SOUTHERN ITALY -A CASE-STUDY OF CAUSE AND EFFECT OF VARIABILITY OF ATMOSPHERIC AEROSOL

The spectral extinction coefficient of the atomospheric aerosol at a r ural location in Southern Italy was determined by means of a telephoto meter: the radiance of a target at a horizontal distance of 12 km was measured at nine different wavelengths in the visible. Therefore the e xtinction coefficient of the aerosol contained within a conical volume of about 450,000 m(3) and a length of 12 km was measured. All measure ments were performed in summer 1993 during a period of stagnant air, p resumably always with the same air mass and thus a similar type of aer osol. The daily variation usually followed a similar pattern: decrease in extinction coefficients in the morning and early afternoon and an increase towards the evening. This variation correlated well with the change in humidity. The Angstrom exponent of the spectral extinction c oefficient, was lowest at high humidity and highest at low humidity. C omparing the horizontal attenuation measurements and vertical transmis sion measurements done with a solar photometer, a considerable vertica l extent of the aerosol was found. This is important for climate consi derations, since additional light absorption and scattering by the aer osol can lead to an increase or a decrease of the temperature. Inversi on of the spectral extinction coefficient data to obtain particle size distributions shows that the dry (<50% r.h.) particles have a peak of the volume size distribution at 0.38 mu m, which increases to 0.81 mu m at 80% r.h. Light absorption by the aerosol varied little on a day to day basis, amounting to 20 to 40% of the light extinction coefficie nt. Values this high are common in Europe, also outside densely popula ted areas. The average single scattering albedo of the dry aerosol was 0.76, thus the aerosol will have a heating effect due to its light ab sorption. Copyright (C) 1996 Elsevier Science Ltd