Satellite-based remote sensing of cloud properties is commonly realized by the bispectral solar reflective method, where cloud reflectance observations at two different spectral bands and pre-calculated lookup tables are used to infer the cloud optical thickness (τ) and effective droplet radius (reff). Clouds within a sampled pixel are assumed to be horizontally homogeneous and are approximated by one–dimensional plane–parallel radiative transfer. However, satellite observations with a lower spatial resolution cannot resolve heterogeneous cloud structures within a pixel, which introduces significant biases in retrieved τ and reff based on the bispectral method.
This talk gives an overview of previous studies on the impact of unresolved cloud variability on the cloud property retrieval and presents a unified mathematical framework that explains the bias in retrieved τ and reff. Experimental validation of this framework is presented by means of synthetic cloud fields from a large-eddy simulation and real observations from the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) aboard NASA’s Terra satellite. ASTER retrievals are provided by a new, research-level retrieval setup, which utilizes the operational MODIS Data Collection 6 (C6) retrieval algorithm and yields τ and reff at a horizontal resolution as high as 30m.
The presented results provide an advanced theoretical understanding of how sub-pixel reflectance variability affects τ and reff retrievals and could guide future imager requirements for cloud remote sensing.
613 Seminar Series Committee: