Discriminating between "Drizzle or rain" and sea salt aerosols in Cloudnet for measurements over the Barbados Cloud Observatory

Abstract. The highly sensitive Ka-band cloud radar at the Barbados Cloud Observatory (BCO) frequently reveals radar reflectivities below -50 dBZ within updrafts and below the cloud base of cumulus clouds. These so-called haze echoes are signals from hygroscopically grown sea salt aerosols. Within the Cloudnet target classification scheme, haze echoes are generally misclassified as precipitation ("Drizzle or rain"). We present a technique to discriminate between "Drizzle or rain" and sea salt aerosols in Cloudnet that is applicable to marine Cloudnet sites. The method is based on deriving heuristic probability functions utilizing a combination of cloud radar reflectivity factor, radar mean Doppler velocity and ceilometer attenuated backscatter coefficient. The method is crucial for investigating the occurrence of precipitation and significantly improves the Cloudnet target classification scheme for the measurements over the BCO. The results are validated against the amount of precipitation detected by the Virga-Sniffer tool. We analyze data for the measurements in the vicinity of the BCO covering two years (July 2021–July 2023) as well as during the ElUcidating the RolE of Cloud–Circulation Coupling in ClimAte (EUREC4A) field experiment that took place in Jan–Feb 2020. A first-ever statistical analysis of the Cloudnet target classification product including the new "haze echo" target over two years at the BCO is presented. In the atmospheric column above the BCO, “Drizzle or rain" is on average more frequent during the dry season compared to the wet season, due to the higher occurrence of warm clouds contributing to the amount of precipitation. Haze echoes are identified about four times more often during the dry season compared to the wet season. The frequency of occurrence of "Drizzle or rain" in Cloudnet caused by misclassified haze echoes is overestimated by up to 16 %. Supported by the Cloudnet statistics and the results obtained from the Virga-Sniffer tool, 48 % of detected warm clouds in the dry and wet season precipitate. The proportion of precipitation evaporating fully before reaching the ground (virga) is higher during the dry season. During EUREC4A, precipitation from warm clouds was found to reach the ground more frequently over the RV Meteor compared to the BCO.