Advancing the satellite remote sensing of heterogeneous clouds by developing a tomographic technique that uses 3D radiative transfer.
Clouds are one of the dominant sources of uncertainty in our understanding of climate and weather. Much of our information about the global distribution of clouds and their associated radiative feedbacks and interactions with aerosols, is obtained from the inversion of measurements from satellite instruments. A single algorithm has been at the heart of these studies, which is the bi-spectral method for retrieving cloud optical depth and droplet effective radius. This retrieval assumes that clouds form homogeneous slabs during the retrieval process, an assumption which causes errors in the retrieval that vary systematically with cloud heterogeneity. I demonstrate for the first time how these retrieval errors cause bias in the assessment of various aerosol cloud interactions in marine boundary layer clouds. Making progress requires the development of retrievals that relax the strong assumptions of homogeneity and slab geometry. I describe the development and validation of a tomographic retrieval of 3D volumetric properties of clouds. This technique can retrieve cloud microphysics at resolutions comparable to Large Eddy Simulations and does not require strong assumptions about vertical structure like existing techniques. Cloud tomography will provide unprecedented insight into climatically important marine cumulus clouds and other heterogeneous clouds for which current cloud remote sensing techniques are inadequate.
