A Novel Arctic Cloud Phase Classifier Based on Ship-Based Remote Sensing Measurements During the MOSAiC Expedition

Accurately measuring the phase of Arctic clouds is of great significance for understanding the process of Arctic climate change. A novel algorithm for classifying Arctic cloud phase is constructed based on ship-based remote sensing measurements during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition. Based on complementary information by Ka-Band ARM zenith radar (KAZR), high spectral resolution lidar (HSRL), microwave radiometer (MWR), and temperature soundings, cloud pixels in the vertical direction can be classified as clear sky, aerosol, ice, mixed-phase, liquid, drizzle, rain, and snow. The data set collected from September 2019 to September 2020 is analyzed. Results indicate that the proposed method can effectively distinguish Arctic mixed-phase clouds with low depolarization ratio from liquid clouds. Even if the mixed-phase clouds contain the lower concentration of ice crystals, the proposed algorithm can accurately identify them. Simultaneously, the cloud phase classification is less affected by environmental factors. It provides a reference for the subsequent cloud microphysical retrievals, reducing significant deviations in droplet microphysical retrievals caused by misidentifying mixed-phase cloud as liquid cloud. The algorithm proposed in this paper makes appropriate modifications to cloud classification boundaries and can also be applied to other sites.