Evaluating the realism of double moment parameterised particle size distributions in a midlatitude frontal ice cloud with complex microphysics

Abstract Uncertainties in the representation of ice particle size distributions (PSDs) cause inaccuracies in simulations of clouds, which have negative impacts on weather forecast and climate predictions. In this study, a midlatitude frontal case study from the Parameterizing Ice Cloud using Airborne obServationS and triple frequency dOppler radar (PICASSO) field campaign in the UK is analysed. The cloud was up to 6 km deep and persisted for several hours, which allowed ice PSDs at a range of temperatures from C to near C to be sampled by the Facility for Airborne Atmospheric Measurements (FAAM) research aircraft; this allowed information about a range of microphysical regimes to be captured. We compare observed PSDs and gamma PSDs with the same concentration and ice water content. We find that at low temperatures a gamma distribution is an appropriate fit to the observations, predicting PSDs similar to those observed. However, in warmer, microphysically complex regions with multiple ice crystal habits, some of which may be caused by secondary ice production, a single gamma distribution does not provide an accurate parameterisation of the observed PSDs. We explore the impact of these discrepancies on the microphysical evolution of the cloud by computing the corresponding process rates using the observed and parameterised PSDs. Errors of more than 100% in aggregation rate, around 50% in precipitation and vapour growth rates, and more than 10 dB in radar reflectivity are calculated.