Evaluating spectral cloud effective radius retrievals against multi-angle polarimetry and in situ cloud probes

Cloud droplet effective radius (CER), defined as the ratio of the 3rd moment of the droplet size distribution (DSD) to the 2nd moment, is a radiative quantity widely used for studies of aerosol/cloud interactions and their impacts on Earth's radiation budget and hydrological cycle. CER commonly is retrieved simultaneously with cloud optical thickness from passive imagers using a bi-spectral technique pairing a non-absorbing visible/near-infrared spectral channel with an absorbing shortwave/mid-wave infrared spectral channel. Multi-angle polarimetry also can be used to retrieve CER, and DSD width, from polarized reflectance observations across the cloud-bow. Spaceborne imagers providing global CER retrievals include MODIS (Terra, Aqua) and VIIRS (SNPP, NOAA-20/21+). Spaceborne polarized cloud-bow retrievals have been limited primarily to POLDER, though several upcoming missions will include multi-angle polarimeters (e.g., ESA's EarthCARE, NASA's PACE and AOS).

Evaluating CER retrievals typically is done via comparisons against CER derived from DSDs measured in situ by cloud probes. However, these comparisons involve numerous confounding factors that influence the interpretation of comparison results, including instrument calibration, retrieval forward model assumptions, information content/sensitivity differences, etc. Here, we show an extensive comparison of airborne imager CER retrievals from the Enhanced MODIS Airborne Simulator (eMAS) against co-located multi-angle polarimetric retrievals from the Research Scanning Polarimeter and in situ cloud probes obtained for marine stratocumulus during the NASA ORACLES field campaign. eMAS is spectrally similar to MODIS and VIIRS and includes channels in the CER-sensitive 1.6µm (2 channels), 2µm (4 channels), and 3.7µm (1 channel) bands that enable a variety of bi-spectral retrieval channel pairings beyond those of any single spaceborne imager. We also show CER and DSD effective variance inferred from eMAS observations of the backscatter glory using peak-matching similar to the polarized cloud-bow. The impacts of numerous confounding factors are explored, and the broader implications of retrieval differences due to differing fundamental sensitivities of multi-spectral imagery and multi-angle polarimetry are discussed.