At temperatures between -40°C and 0°C, clouds can be mixed phase, so called because they consist of a mixture of both liquid cloud droplets and ice crystals. This type of cloud is especially poorly represented in climate models. One of the reasons is that both hydrometeors are assumed to be homogeneously mixed in global models, but observations show that ice and liquid are heterogeneously mixed and exist in separate "pockets". This difference in the 3-dimensional spatial distribution of ice and liquid is important to assess and quantify precipitation, cloud processes, radiative properties, and consequently their impact on climate change. The present study aims to better characterize mixed phase clouds and especially the spatial distribution of the thermodynamic phase and understand how meteorology, air parcel transport and aerosols impact it.

We defined a parameter to describe the spatial distribution of liquid and ice phases within mixed-phase clouds from observations from the space-based lidar CALIOP (Cloud-Aerosol Lidar with Orthogonal Polarisation). We spatially and temporally collocated the satellite measurements with reanalysis retrievals of aerosol concentration and meteorological parameters from ERA5 (European Centre for Medium-Range Weather Forecasts Reanalysis v5) and MERRA-2 (Modern-Era Retrospective analysis for Research and Applications, version 2) and then applied a multi-linear linear regression fit to quantify the influence of the external parameters on the spatial distribution of the cloud phase up to first order. A second part of the study focuses on ground-based measurements from the North Slope Alaska Station (NSA), where the transport of air parcels is analysed according to cloud type.

Focusing on the Arctic region, the results show that temperature is the most important parameter influencing the liquid-ice interface: for example, clouds with a temperature above 265 K have seven times more liquid-ice interfaces and are more homogeneously mixed than clouds with a temperature below 253 K. Black carbon concentration are also important parameters to describe the phase distribution. At NSA, clouds associated with higher transport may be more heterogeneously mixed. The results could be used to refine the parameterisation of clouds in models and their impact on climate change.