A New AOD Retrieval in Vicinity of Broken Clouds: A Potential Tool to Study Aerosol Indirect Effects

To investigate aerosol indirect effects, simultaneous and spatially coincident measurements of aerosol and cloud properties are highly desirable. Currently, aerosol optical depth (AOD) could be estimated quite accurately only for a limited number of selected clear pixels located far away (~3 km) from broken clouds and their shadows. Locating such pixels requires statistical analysis of two-dimensional (2D) distribution of reflectance. Here we introduce a novel method for the AOD retrievals under partly cloudy conditions. In contrast to the current retrieval algorithms, which are based on reflectance, a new method exploits reflectance ratios, which are less sensitive to the three-dimensional (3D) effects of clouds. As a result, the new method provides an effective way to minimize the 3D cloud effects, which otherwise would have a large (up to 140%) contaminating impact on the aerosol retrievals. To illustrate the potential for this approach, we use 3D fields of aerosol and clouds obtained from a Large-Eddy Simulation (LES) with explicit microphysics. The LES simulations are performed for typical conditions observed during the Cloud and Land Surface Interaction Campaign (CLASIC) conducted during the summer of 2007. We demonstrate that the new retrieval has the potential for (i) detection of clear pixels outside of cloud shadows, (ii) increased "harvest" of such pixels, and (iii) accurate (~15%) estimation of AOD for the majority of them. Uncertainties of the ratio- based AOD retrieval and their sensitivity to observational conditions are also discussed.