Detection of supercooled liquid water containing clouds with ceilometers: development and evaluation of deterministic and data-driven retrievals

Abstract. Cloud and aerosol lidars measuring backscatter depolarization ratio are the most suitable to detect cloud phase (liquid, ice, or mixed phase). However, such instruments not widely deployed as part of operational networks. In this study, we propose a new algorithm supercooled liquid water containing clouds (SLCC) based on ceilometers only co-polarization backscatter. We utilize observations collected at Davis, Antarctica, where low-level, mixed-phase clouds, including (SLW) droplets ice crystals, remain poorly understood due paucity ground-based observations. A 3-month set were during austral summer November 2018 February 2019, with variety lidar W-band radar which used build two-dimensional mask distinguishing SLW clouds. This is reference develop single polarization ceilometer operating in vicinity for same period. Deterministic data-driven retrieval approaches evaluated: an extreme gradient boosting (XGBoost) framework ingesting average characteristics was effective method reproducing classification obtained combined radar–lidar approach accuracy high 0.91. study provides SLCC data highlights considerable benefits these provide intelligence polar regions that usually suffer from Finally, two algorithms applied full year retrieve frequency occurrences SLCC: present 29 ± 6 % time T19 24 5 G22-Davis over annual cycle.