A comparison and summary of aerosol optical properties as observed in situ from aircraft, ship, and land during ACE-Asia

[1] During the ACE-Asia campaign in March-May 2001, in situ measurements of aerosol optical properties were made from multiple airborne and land- or ship-based platforms. Using a suite of direct interplatform comparisons and a campaign-wide statistical comparison, we test the precision of these measurements, and we determine whether the platforms sampled similar aerosol. Data included in the study are from the National Center for Atmospheric Research C-130 aircraft; the CIRPAS Twin Otter aircraft; the National Oceanographic and Atmospheric Administration (NOAA) ship R. V. Ronald H. Brown; and the Gosan surface station on Jeju Island, located off the southern tip of South Korea. Comparisons were made of total and submicron light scattering at 450, 550, and 700 nm; total and submicron absorption at 550 nm; the <(A)over circle>ngstrom exponent; single scatter albedo of the total aerosol, submicron and supermicron aerosol at 550 nm; hemispheric backscatter fraction at 550 nm; and light scattering hygroscopic growth at 550 nm. For the campaign-wide comparison, the data are broken down by light scattering fine mode fraction since the aerosol in the ACE-Asia study region were a variable mix of pollution, dust, and sea salt. Finally, we calculate how the observed uncertainties in the aerosol optical properties propagate to uncertainties in top-of-atmosphere radiative forcing. Single scatter albedo showed excellent agreement among all platforms other than the Twin Otter, with discrepancies generally < 0.02. These data sets combine to give campaign-wide values of single scatter albedo of 0.885 +/- 0.023 for the submicron aerosol (i.e. pollution) and 0.957 +/- 0.031 for the supermicron aerosol (which, for these data, was predominantly dust). The data also indicated that, as expected, the Low Turbulent Inlet on the C-130 produced enhanced concentrations of coarse mode aerosol. There also may have been significant coarse mode particle losses on the other platforms. These effects combined to produce generally lower fine mode fractions and <(A)over circle>ngstrom exponents on the C-130 than on the other platforms. Large discrepancies in hemispheric backscatter fraction and light scattering hygroscopic growth were observed in both the side-by-side and statistical comparisons. We are not able to explain these differences, though possible causes are discussed. Studies of the TSI, Inc. nephelometer backscatter measurement and of the two methods used here to measure hygroscopic growth are needed to clarify the source of these observed discrepancies. A better understanding of the effects of nonsphericity on hemispheric backscatter fraction is also needed.