This semester's project aims to explore the potential of the Vaisala CL61 ceilometer with polarization capacity for characterizing atmospheric observations. To achieve this, a Python pipeline was developed to classify atmospheric targets. The noise was filtered out from the data, enabling an efficient application of K-means clustering. Each cluster was classified by fixing thresholds on attenuated backscatter and linear depolarization ratio. The results show differentiation between liquid and solid scattering targets as well as the identification of aerosol plumes, demonstrating the extended use of ceilometers through their polarization capabilities.