Remote sensing of optical and microphysical properties of cirrus clouds using MODIS channels : methodology and sensitivity to physical assumptions

A methodology for the detection of cirrus clouds and the retrieval of their microphysical and optical properties based on observations of reflected sunlight is introduced. The retrieval method is based on correlation of the bidirectional reflectance of three channels, 0.65, 1.6, and 2.2 µm, that are available on board Earth Observing System (EOS) Moderate Resolution Imaging Spectroradiometer (MODIS). Validation studies using microphysical measurements and MODIS Airborne Simulator (MAS) observations illustrate the nature of the potential errors associated with the retrieved optical depth and mean effective ice crystal size. The effects of the physical assumptions involving ice crystal size distribution and shape employed in the algorithm are subsequently assessed. In terms of the microphysical models used for radiation calculations, the ice crystal shape assumption is found to have the most significant impact on the retrieved parameters. The effect of the background surface reflectance on the retrieval results is further examined and we show that in order to reliably infer thin cirrus parameters from solar reflectance measurements, it is essential to properly account for the background radiation both over land and ocean surfaces. Finally, we present the measured ice microphysical data for tropical cirrus as a function of cloud development and ambient temperature to illustrate the importance of vertical inhomogeneity for validation studies.