In the upper region, large aerosol concentrations last for a few days; during these events aerosol is often detected up to 7 or 8 kin. Large amounts were detected in mid-May and were very often observed in June. By using meteorological analyses and isentropic backward trajectories, the aerosol behavior above Lampedusa has been related to the large-scale transport patterns and to the source regions. Large aerosol loads are clearly due to dust transport from Africa, occurring through two main paths: from central Sahara, when a high-pressure system was centered over northern Libya, and following the northwestern African coast, often along the Atlas Mountains, when the anticyclone is over Algeria or Libya, at latitudes lower than 30øN. Large aerosol loads are observed even when the air mass trajectories marginally overpass Africa, often up to 5-6 kin. According to the isentropic trajectories, large vertical motions occur when the air masses travel over Africa. Significant differences in the aerosol

No abstract

Desert dust interacts with shortwave (SW) and longwave (LW) radiation, influencing the Earth radiation budget and the atmospheric vertical structure. Uncertainties on the dust role are large in the LW spectral range, where few measurements are available and the dust optical properties are not well constrained. The first airborne measurements of LW irradiance vertical profiles over the Mediterranean were carried out during the Ground-based and Airborne Measurements of Aerosol Radiative Forcing (GAMARF) campaign, which took place in spring 2008 at the island of Lampedusa. The experiment was aimed at estimating the vertical profiles of the SW and LW aerosol direct radiative forcing (ADRF) and heating rates (AHR), taking advantage of vertically resolved measurements of irradiances, meteorological parameters, and aerosol microphysical and optical properties. Two cases, characterized respectively by the presence of a homogeneous dust layer (3 May, with aerosol optical depth, AOD, at 500 nm of 0.59) and by a low aerosol burden (5 May, with AOD of 0.14), are discussed. A radiative transfer model was initialized with the measured vertical profiles and with different aerosol properties, derived from measurements or from the literature. The simulation of the irradiance vertical profiles, in particular, provides the opportunity to constrain model-derived estimates of the AHR. The measured SW and LW irradiances were reproduced when the model was initialized with the measured aerosol size distributions and refractive indices. For the dust case, the instantaneous (solar zenith angle, SZA, of 55.1°) LW-to-SW ADRF ratio was 23% at the surface and 11% at the top of the atmosphere (TOA), with a more significant LW contribution on a daily basis (52% at the surface and 26% at TOA), indicating a relevant reduction of the SW radiative effects. The AHR profiles followed the aerosol extinction profile, with comparable peaks in the SW (0.72 ± 0.11 K d À1) and in the LW (À0.52 ± 0.12 K d À1) for the considered SZA. On a daily basis, the absolute value of the heating rate was larger in the LW than in the SW, producing a net cooling effect at specific levels. These are quite unexpected results, emphasizing the important role of LW radiation.

No abstract

[1] Measurements of the aerosol properties were carried out at the island of Lampedusa, in the Mediterranean, in May 1999, as part of the Photochemical Activity and Ultraviolet Radiation modulating factors II campaign. Data from ground-based lidar and Sun photometer, and particle counters aboard an instrumented ultralight aircraft, are used in this study. Three different cases, when all the measurements were available in cloud-free conditions, were identified to derive the aerosol microphysical and optical properties. In one of these cases (18 May) the airmasses originated from Africa, and were loaded with a large amount of desert dust. In the other two cases (25 May and 27 May) the airmasses passed over Europe before reaching Lampedusa from North. The microphysical and optical properties of the aerosol strongly depend on the origin of the airmasses. The amount of particles in the 1-6 mm range of radii and the average aerosol surface area per unit volume are larger in the desert dust case than on 25 May and 27 May. The real part of the refractive index of the desert dust at 532 nm is between 1.52 and 1.58; its imaginary part is 5-7 Â 10 À3 and the single scattering albedo is about 0.7-0.75. The aerosol layer of 18 May closest to the surface, that probably contains a mixture of desert dust and marine aerosol, displays a smaller imaginary part (1.2 Â 10 À3 ) and a larger single scattering albedo (0.91). The aerosols originating from the North Atlantic and Europe have a real part of the refractive index between 1.35 and 1.49, and an imaginary part ranging from 8 Â 10 À4 to 1.8 Â 10 À2 ; the single scattering albedo at 532 nm (0.78-0.95) is larger than for desert dust values. The smallest value of the single scattering albedo (0.69) corresponds to an airmass originating from North, characterized by a large imaginary part of the refractive index. The asymmetry factor of the desert dust appears consistently larger for the desert dust (0.75-0.8) than for the other cases (0.61-0.72). The extinction-to-backscattering ratio, also derived from the measurements, is about 40 sr for the desert dust, and between 60 and 81 sr for the aerosol of northern origin. Simple estimates of the aerosol average direct shortwave radiative forcing at the top of the atmosphere indicate that all considered aerosol types induce a cooling. The radiative forcing per unit optical depth of the aerosol originating from North is about À37 Wm À2 over ocean and À(12-17) Wm À2 over land, while is À29 Wm À2 over ocean and À8 Wm À2 over land for desert dust. The largest forcing is however produced by the desert aerosols that generally display a considerably larger optical depth.