Ice nucleation by soil dust compared to desert dust aerosols

Abstract:

A minor fraction of atmospheric aerosol particles, so-called ice-nucleating particles (INPs), initiates the formation of the ice phase in tropospheric clouds and thereby markedly influences the Earth’s weather and climate systems. Whether an aerosol particle acts as an INP depends on its size, morphology and chemical compositions. The INP fraction of certain aerosol types also strongly depends on the temperature and the relative humidity.

Because both desert dust and soil dust aerosols typically comprise a variety of different particles, it is difficult to assess and predict their contribution to the atmospheric INP abundance. This requires both accurate modelling of the sources and atmospheric distribution of atmospheric dust components and detailed investigations of their ice nucleation activities. The latter can be achieved in laboratory experiments and parameterized for use in weather and climate models as a function of temperature and particle surface area, a parameter called ice-nucleation active site (INAS) density. Concerning ice nucleation activity studies, the soil dust is of particular interest because it contains a significant fraction of organics and biological components, both with the potential for contributing to the atmospheric INP abundance at relatively high temperatures compared to mineral components.

First laboratory ice nucleation experiments with a few soil dust samples indicated their INP fraction to be comparable or slightly enhanced to that of desert dust. We have used the AIDA (Aerosol Interaction and Dynamics in the Atmosphere) cloud simulation chamber to study the immersion freezing ability of four different arable soil dusts, sampled in Germany, China and Argentina. For temperatures higher than about -20°C, we found the INP fraction of aerosols generated from these samples by a dry dispersion technique to be significantly higher compared to various desert dust aerosols also investigated in AIDA experiments. In this contribution, we will summarize the experimental results, introduce related INP parameterizations for use in weather and climate models, and briefly discuss possible reasons for the discrepancy between the INP fraction of desert and soil dust aerosols.