Long-term measurements at the U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) at the North Slope Alaska (NSA) site in Utqiagvik, Alaska and from the Multidisciplinary drifting Observatory for the study of the Arctic Climate (MOSAiC) expedition are being used to study the climatology of clouds containing supercooled liquid (SCL) in the Western and Central Arctic. Classification of cloud hydrometeors in the liquid, ice or mixed phase of the cloud is determined by using the Cloudnet algorithm developed by the Finish Meteorological Institute. We apply the Cloudnet processing chain to a set of ground-based remote sensing measurements from the NSA site and the ARM mobile facility and the TROPOS shipborne atmosphere observation suite (OCEANET) on board of the RV Polarstern research vessel during MOSAiC. In order to accurately detect cloud droplets and SCL layers within mixed-phase clouds, Cloudnet relies on lidar observations. Lidars however suffer from total signal attenuation at a penetration optical depth of about three. Conversely cloud radars with their capability to penetrate multiple liquid layers can be used to expand the identification of cloud phase to the entire vertical column beyond the lidar signal attenuation height by using information of the cloud radar Doppler spectrum. The Leipzig Institute for Meteorology (LIM) recently developed a deep learning approach for reVealing supercOOled liquiD beyOnd lidar attenuatiOn (VOODOO) which benefits from the morphological features in cloud radar Doppler spectra to extract further information related to the existence of SCL using Cloudnet’s target classification as supervisor. The current contribution presents a SCL climatology obtained using Cloudnet for the NSA site along with case-studies for MOSAiC where VOODOO results are contrasted with the standard Cloudnet outputs. Advantages and limitations of both methods will be presented to the scientific community.