资源环境科技发展态势分析平台

Between -37 and 0 degrees C, clouds are liquid, ice, or mixed phase. Nearly all retrieval algorithms for passive instruments provide binary phase informationice or liquidmaking it difficult to retrieve mixed-phase cloud properties. Based on measurements from the geostationary space-based instrument Spinning Enhanced Visible and InfraRed Imager (SEVIRI), we show that the retrieved ice crystal effective radius is smaller than the liquid droplet effective radius for 48% of 230 analyzed cloud thermodynamic phase transitionsphase transition from liquid to ice of rising convective cloudswhile ice crystals are expected to be larger than cloud droplets. We simulate mixed-phase cloud radiances with the numerical model Santa Barbara DISORT Atmospheric Radiative Transfer for which we compare simulated effective radius retrievals with observations. The phase retrieval algorithm from SEVIRI does not represent well mixed-phase clouds, and categorizing clouds by only ice and liquid is not enough to accurately represent mixed-phase cloud optical properties. We conclude that the mixed-phase nature of clouds explains that retrieved cloud droplet radii are larger than ice crystal radii directly before and after the phase transition. However, from a cloud tracking algorithm perspective, the variation of the effective radius enables the detection of mixed-phase convective clouds from binary phase information.