Statistics of Optical and Microphysical Properties of Ice and Mixed-Phase Clouds on the Antarctic Plateau

The uncertainties in the cloud radiative properties are the main cause of biases in the radiative fluxes both at the top of the atmosphere and at the surface (Di Natale et al. 2022). This study aims at providing an in-depth characterisation of clouds occurrence and properties on the Antarctic Plateau by analyzing an extensive dataset of spectrally resolved downwelling radiances in the far- and mid-infrared region of the spectrum (200 − 1000 cm-1). Observations were performed by the REFIR-PAD (Radiation Explorer in the Far Infrared—Prototype for Applications and Development) spectroradiometer at the Concordia research station on the Antarctic Plateau, during the period from 2013 to 2020. An improved version of the Cloud Identification and Classification (CIC) algorithm (Maestri et al. 2019; Donat et al. 2024) is utilized for the identification of cloud layers and their classification in terms of phase (ice or mixed phase). An extended dataset, comprising about 75000 cloudy spectra, is then analysed to derive geometrical, optical, and microphysical properties of the observed layers. First, the Polar Threshold (PT) algorithm (Van Tricht et al. 2014) is applied to collocated Lidar backscatter profiles obtained from a Lidar system (INO-CNR Istituto Nazionale di Ottica 2024) to derive cloud base and top altitude. Then, the geometrical information is exploited by the Simultaneous Atmospheric and Cloud Retrieval (SACR) physical inversion algorithm (Di Natale et al. 2020), which, applied to the REFIR-PAD radiances, enables the derivation of cloud optical depth, effective dimensions, and atmospheric vertical profiles of water vapor and temperature. Statistics of clouds optical and microphysical properties for different cloud types are derived. Results show that the mean optical depth and effective diameter of ice clouds are 0.58 and 25 μm, respectively. It is found that 90% of the data indicate effective diameters smaller than 52 μm. During the austral summer, both optical depth (0.34) and effective diameter (19 μm) are at their lowest values, while maxima are found in the winter season (0.73 for optical depth and 30 μm for effective diameter). The ice clouds mean temperature is 236 K, with a seasonal cycle showing the highest temperatures in summer. Mixed-phase clouds exhibit a significantly higher mean optical depth of 2.35. Their averaged effective diameter is 8.55 μm, and the mean cloud temperature is 244 K. The base height of mixed-phase clouds is found at mean value of 0.29 km above ground level (agl) which is significantly lower than the one for ice clouds found at 0.62 km agl. Finally, a new parametrization of cloud microphysical properties based on the thermodynamic conditions of the layer is proposed for potential applications in climate and numerical weather prediction models. The effective diameter of ice crystals is parameterized as a function of the mean cloud temperature and ice water content, whereas the effective diameter of water droplets for mixed-phase clouds is expressed as a univariate function of the mean cloud temperature. A comparison with parametrizations widely used in climate and numerical weather prediction models is also provided.