Global distributions of Ice-nucleating particles using ICON-ART

Ice nucleating particles (INPs) are aerosols that lower the energy barrier for ice formation in mixed phase clouds, and therefore impact the liquid water fraction in these clouds. Hence, the prevalence of INPs affects the radiative properties of the cloud, as well as the precipitation formation.  In addition, the liquid water fraction of clouds is one factor thought to be contributing to the consistent Southern Ocean radiation bias found across CMIP6 models.  Cloud-aerosol interactions remain a major area of uncertainty in climate modelling, and non-aerosol-aware models use INP parameterisations that are purely temperature dependent and do not take into account the regional variation in aerosol concentration and type.

 Here, we will present modelled global distributions of mineral dust and marine organic INPs – and their relative contributions to the total INP – calculated from simulations with the Aerosol and Reactive Trace (ART) gases module of the ICOsahedral Nonhydrostatic NWP model (ICON).  The INPs are calculated offline using the Ice-nucleating active site (INAS) densities for immersion freezing provided by Ullrich et al 2014 for dust INP and McCluskey et al 2018 for marine INP. The simulations run for one year at 80km horizontal grid spacing.  A comparison to the temperature-only parameterisations and observations of INPs is made, with a particular focus on Antarctica (data from Wex et al 2025) and the Southern Ocean (Antarctic Circumnavigation Expedition ship campaign).

Furthermore, we calculate INPs using a preliminary version of the seamless ICON-ART (ICON-SmART) model and evaluate this. Ultimately, the aim is to improve online INP modelling in ICON-SmART, in part to address the Southern Ocean radiation bias found in seasonal to decadal simulations.