The Spectral Shape of Shortwave Cloud Albedo and Apparent Absorption

Simultaneous measurements of spectral shortwave irradiance above and below Cirrus clouds during the "Tropical Composition, Cloud and Climate Coupling" experiment (TC4) revealed that the spectral shape of cloud albedo and absorptance can deviate substantially from one-dimensional model calculations. Throughout the experiment, the NASA ER-2 and DC-8 aircraft were flown in vertically stacked formation along extended legs. This allowed the derivation of vertical flux divergence (apparent absorption) from above-cloud and below-cloud irradiance measurements by the Solar Spectral Flux Radiometer (SSFR). At the same time, the MODIS Airborne Simulator (MAS) onboard the ER-2 provided two-dimensional fields of cloud optical thickness and effective radius, which were compared with retrievals based on visible and near-infrared cloud albedo from SSFR. Averaged over the footprint of SSFR (the circle within which 50% of the upwelling irradiance emanates), the optical thickness from MAS was systematically higher than the SSFR-retrieved value, while the effective radius was equal or lower than that from SSFR. The consistency of the retrievals with the spectral measurements was tested by calculating the whole spectrum of shortwave albedo and apparent absorption from the retrieved cloud properties along each individual point of the flight tracks. Between 400 nm and 700 nm, the measured albedo decreased more than modeled with increasing wavelength. The measured apparent absorption was non-zero, in contrast to the one-dimensional model which predicts zero cloud absorption for wavelengths shorter than about 1000 nm. In optically thick regions, it generally increased with wavelength. Most of the discrepancies were resolved by using the MAS-retrieved cloud fields as input to three-dimensional model calculations: The spectral shape of measured albedo and apparent absorption was generally reproduced by resolving cloud heterogeneities within the model. In addition, irradiance contributions from beyond the swath of MAS and spherical effects were examined to explain the remaining differences between three-dimensional model and measurements.