Fragmentation of ice particles: laboratory experiments on graupel–graupel and graupel–snowflake collisions

Abstract

Abstract. Until now, the processes involved in secondary ice production which generate high concentrations of ice crystals in clouds have been poorly understood. However, collisions that involve rimed ice particles or crystal aggregates have the potential to effectively produce secondary ice from their fragmentation. Unfortunately, there have only been a few laboratory studies on ice–ice collision so far, resulting in an inaccurate representation of this process in microphysical schemes. To address this issue, experiments were conducted at the wind tunnel laboratory of the Johannes Gutenberg University, Mainz, on graupel–graupel and graupel–snowflake collisions under still-air conditions at −15 ∘C and water supersaturation. The particles were synthetically generated within a cold room through two distinct methods: by riming and vapor deposition for graupel with diameters of 2 and 4 mm and by manually sticking vapor-grown ice which was generated above a warm bath to form snowflakes with a diameter of 10 mm. All fragments resulting from graupel–graupel collisions were collected and investigated by means of a digital optical microscope, while fragments from graupel–snowflake collisions were observed and recorded instantly after collision using a holographic instrument. From these experiments, distributions were obtained for fragment sizes, cross-sectional areas, and aspect ratios. The results showed a higher number of fragments at lower kinetic energy compared to those presented in the literature. A total of 150 to 600 fragments were observed for graupel–graupel with dendrites collisions, as well as 70 to 500 fragments for graupel–snowflake collisions for collision kinetic energies between 10−7 and 10−5 J. Parameterizations for fragment size distributions are provided with a mode at 75 µm for graupel–graupel with dendrites collisions and at 400 µm for graupel–snowflake collisions. We also propose new coefficients fitted on our experiments to parameterize the number of fragments generated by collisions based on the theoretical formulation of Phillips et al. (2017). These results can be used to improve the representation of ice collision breakup in microphysical schemes.