Mixed-phase Multilayer Clouds in the Arctic: A Simulation Study using ICON

Multilayer clouds (MLCs), defined as individual, vertically overlapping clouds, are frequently occurring worldwide but have been far less studied than single layered clouds. Earlier studies have suggested a clear abundance of MLCs in the Arctic compared with the rest of the world and with data from the MOSAiC campaign in 2019-2020 we have classified multilayered clouds at a 52% frequency of occurrence. The microphysical interaction between these cloud layers is expected to be complicated, such as the seeder- feeder mechanism, and we thus employ a model to further investigate these clouds. 

Cases from the MOCCHA campaign in 2018 as well as the MOSAiC campaign in 2019-2020 have been selected for MLC occurrences. These cloud systems vary from vertically distinct layers with no potential of seeding (subsaturated layer of >3km) to a doubly layered system within the boundary layer with frequent seeding events. The structure of the former can be simulated at a coarse grid spacing, provided appropriate initial conditions and aerosol concentration, whilst the latter is highly dependent on initial and boundary conditions, resolution, and parameterisation for the boundary layer. 

Together with an analysis of the measurements on board of the ships, the ICON (ICOsahedral Non-hydrostatic) model was deployed. The simulations are run with refined nests down to 75 meters horizontal grid spacing in ICON-LEM. Initial and boundary data are supplied by both ICON Global and IFS. As the Arctic aerosol contribution is yet to be parameterised, we are further making use of the prognostic aerosol module ART (Aerosol and Reactive Trace gases) developed by KIT, set up specifically for cloud condensation nuclei activation for sea salt and sulfate. 

Various sensitivity experiments have been performed on these case studies including (i) sensitivity to microphysical parameters, such as CCN and INP parameterisation and concentration, (ii) sensitivity to horizontal and vertical resolution as well as (iii) initial and boundary condition impacts on resolving the cloud layers. Furthermore, the aerosol concentration has been scaled, in the existing parameterisations in ICON, to represent the measurements on site as well as prognostically run using ICON-ART. 

Preliminary results on the modelled multilayer cloud system highlight a high dependency on the initial and boundary data quality as well as domain resolution while the microphysics have a smaller impact on the formation and detailed structure of the multilayer cloud system.