Advances in Ceilometer Data Processing—Detecting Atmospheric Icing in the Vertical Profile

Aviation sector is especially vulnerable for variety of weather phenomena; such as strong wind, turbulence and icing. In general, icing is a significant challenge for various areas of every life in the modern society; in addition to aviation, other lines of transport, and even energy, business can be influenced by the phenomenon. The smaller the aviation vehicle, such as drone, the faster icing can cause significant danger. Today number of professional and hobby drone flights has increased. Due to influence of phenomenon dedicated icing forecasts and monitoring are required for safe operation of aviation.

Weather services forecast operationally icing conditions. Skill of every forecast model has to be evaluated, which is best done with observations. Traditionally icing conditions are monitored with in-situ sensors close to the ground. However, icing, which can be caused by super-cooled liquid water droplets in fog, cloud or precipitation, occurs in vertical profile of the atmosphere. Ceilometers were designed for cloud ceiling. Nowadays, more and more ceilometers record profiles of attenuated backscatter coefficient from aerosol and hydrometeors. Additionally, Finnish Meteorological Institute operates more than 100 ceilometers throughout the country. This encouraged us to investigate use of ceilometer as a real-time icing monitor in vertical profile of the lower atmosphere.

We developed an algorithm to identify super-cooled liquid water containing clouds and subsequent precipitation based on ceilometer attenuated backscatter coefficient profiles. We evaluated ceilometer based icing target classification by comparing to observations of in-situ icing and hydrometeor number size distribution sensors at towers in two locations in Finland. Observations utilized in this study cover altogether 8 years. In Finland icing occurs on daily basis between September and April. The results suggest that icing due to super-cooled liquid water containing clouds can be reliably monitored with a ceilometer network up to cloud layers visible to sensor. Additionally, icing fog can also be identified with this method. We are further developing reliability of the algorithm in identifying precipitation. The latest status of algorithm development and results will be presented in the conference.

This work was financially supported by MATINE.