Unravelling the microphysics of polar mesospheric cloud formation

Polar mesospheric clouds are the highest water ice clouds occurring in the terrestrial atmosphere. They form in the polar summer mesopause, the coldest region in the atmosphere. It has long been assumed that these clouds form by heterogeneous nucleation on meteoric smoke particles which are the remnants of material ablated from meteoroids in the upper atmosphere. However, until now little was known about the properties of these nanometre-sized particles and application of the classical theory for heterogeneous ice nucleation was impacted by large uncertainties. In this work, we performed laboratory measurements on the heterogeneous ice formation process at mesopause conditions on small (r = 1 to 3 nm) iron silicate nanoparticles serving as meteoric smoke analogues. We observe that ice growth on these particles sets in for saturation ratios with respect to hexagonal ice below S-h = 50, a value that is commonly exceeded during the polar mesospheric cloud season, affirming meteoric smoke particles as likely nuclei for heterogeneous ice formation in mesospheric clouds. We present a simple ice-activation model based on the Kelvin-Thomson equation that takes into account the water coverage of iron silicates of various compositions. The activation model reproduces the experimental data very well using bulk properties of compact amorphous solid water. This is in line with the finding from our previous study that ice formation on iron silicate nanoparticles occurs by condensation of amorphous solid water rather than by nucleation of crystalline ice at mesopause conditions. Using the activation model, we also show that for iron silicate particles with dry radius larger than r = 0.6 nm the nanoparticle charge has no significant effect on the ice-activation threshold.