Selective Simulating Seeding of Hailstorms - a Summertime Case Study Over Switzerland

Abstract Ice Nucleating Particles (INPs) influence the microphysical properties of hailstorms. However, their impact on surface precipitation and hail size distribution remains controversial. This controversy primarily arises because hailstorms are relatively rare phenomena with numerous non-linear interactive processes. This study investigates the effects of INP perturbations using silver iodide (AgI) on a hailstorm observed over the Swiss plateau on July 6, 2019. The focus is on graupel particles, ice, surface precipitation, and hail size distribution. On this day, the synoptic situation is characterized by a short-wave trough moving southward over central Europe, which gradually, during the day, turned the atmospheric flow over Switzerland to westerly-southwesterly. This situation in the middle and upper troposphere, in conjunction with a cold front moving south from the northern Swiss borders, led to the initiation of convection over the Swiss plateau, particularly near Basel. To simulate this case, we employ the Consortium for Small-Scale Modeling (COSMO) regional weather and climate model (Steppeler et al., 2003; Baldauf et al., 2011). The model runs on a rotated latitude-longitude grid with a 0.01° horizontal resolution (corresponding to approximately 1.1 km grid spacing), 80 hybrid vertical levels up to approximately 23 km, and a temporal resolution of 6 seconds. One prerequisite for our experiments is the successful simulation of a thunderstorm comparable to the observed ones based on that day's initial and boundary conditions. We consider this comparison successful if: i) the simulated cell is within a 40 km and 3-hour range of the observed cell, and ii) the hail diameter at the cloud base in the simulated cell exceeds 1 cm at one or more grid points. Once these conditions are met, we seed this hailstorm at the cumulus stage. The seeding is conducted in the updraft region near the cloud base. The additional AgI particles are treated as a prognostic variable, following the AgI freezing parameterization of Marcolli et al. (2016). We use a combination of indicators to pinpoint the region where the hailstorm was at this distinct stage. These include the maximum updraft in the grid column, precipitation rate, and the mass of graupel and hail. The whole process is repeated ten times following the time-lagged ensemble approach (Vogel et al., 2013) to analyze the effect of seeding and its significance in the context of model variability. Preliminary results show that seeding with AgI increases the mass and the concentration of graupel particles in the studied convective storm in all the ensemble members. We also found a strong link between storm dynamics and microphysical properties, highlighted by increased maximum updraft speeds in the simulated multicell.