Key points are not available for this paper at this time.
The predictability of deep moist convection is subject to large uncertainties, mainly due to inaccurate initial and boundary data, incomplete description of physical processes, or uncertainties in microphysical parameterizations. In this study we present results from a large 108-member ensemble focussing solely on the perturbation of microphysical uncertainties. We perturb the cloud condensation nuclei concentrations, the ice nucleating particle concentrations, the graupel sedimentation velocity as well as the width of the cloud droplet size distribution, all of which are not well constrained by observations. The model simulations are conducted with a convection-permitting configuration of the ICON model using a double-moment microphysics scheme. Results from four convective episodes during the Swabian MOSES field campaigns conducted in the summers of 2021 and 2023 show a large spread in convective precipitation in Germany. Based on convection-related parameters and microphysical process rates, the sensitivities of convection initiation, cloud and precipitation formation to the microphysical uncertainties are discussed. An important finding is e.g. the large sensitivity of hail formation on all analysed days. These results demonstrate the benefits of using an aerosol-aware double-moment microphysics scheme and that the use of microphysical uncertainties for ensemble modelling strategies can produce a sufficiently large ensemble spread for convective-scale predictability.
Czajka et al. (Mon,) studied this question.