Recent Improvements in EPIC Cloud Mask Over Snow/Ice and Sunglint Regions

Abstract:

The Earth Polychromatic Imaging Camera (EPIC) onboard the Deep Space Climate Observatory (DSCOVR) satellite has been taking continuous measurements of the entire sunlit side of the Earth’s atmosphere since its launch in 2015. A suite of algorithms has been developed to generate the standard EPIC Level 2 Cloud Products that include cloud mask, cloud effective pressure/height, cloud optical thickness, etc. Initial comparison between version-1 of EPIC cloud products with GEO/LEO observations shows generally good agreement, but also reveals some problematic retrievals such as cloud mask over the snow and ice-covered Antarctic regions. The EPIC cloud mask algorithm considers ocean, land and snow/ice surfaces separately. Over snow and ice surfaces, it relies on two oxygen band ratios to detect clouds. We have implemented a dynamic threshold scheme for the oxygen band ratios as a function of zenith angles and altitude. Over the ocean, we included empirical parameterization of reflectance and oxygen band ratios near the Sunglint regions. These methods have greatly improved the cloud mask performance over snow/ice and Sunglint regions.