Biological ice nucleating particles at tropospheric cloud height

Airborne ice nucleating particles (INPs) promote the freezing of cloud droplets, which is relevant for the radiative properties of clouds and for the development of precipitation. A quantitative assessment of the impact of INPs on cloud processes and on their responsiveness to climate and land use change is still missing. This is particularly true for INPs of biological origin. They are made of molecules produced by bacteria, fungi, plants, lichens, which promote the freezing of droplets at temperatures above -15 °C. Bottom-up modelling studies based on the release of ice nucleation active cells from crops and plants have excluded any chance for biological INPs to impact climate. Nevertheless, recent observations point at the ubiquity across ecosystems of species capable of producing INPs and at the fact that such INPs can be released from cells and maintain their activity for instance when linked to soil particles. Here we employed a top-down approach to improve our understanding of the variability of biological INPs in precipitation. 16 sampling campaign were organised between 2012 and 2014 and over 100 precipitation samples were collected at the High Altitude Research Station Jungfraujoch (3580 m a.s.l., Switzerland). They have been analysed for their content in INPs active at moderate supercooling directly in field with our new immersion freezing apparatus LINDA. Several environmental parameters have been studied to derive more information on the most relevant factors responsible for the variability of INPs. The abundance of bacterial cells and the presence of the nucleation active plant pathogen bacterium Pseudomonas syringae have been determined as well, to know more on the nature of biological INPs in precipitation. By means of stable water isotopes, we demonstrate that INPs are rapidly and selectively removed by precipitating clouds. Focusing on INPs active at -8 °C (INPs-8), their concentrations varied between 0.21 and 434 INPs-8 mL-1. Up to 75% of this large temporal variability can be modelled and predicted by multiple linear regression models based on the combination of a few environmental parameters. These models point at the interaction of “source” (uptake) and “sink” (removal) processes as relevant to determine the variability of INPs-8. Large abundance of INPs-8 can be best expected with a coincidence of high wind speed and little precipitation lost from an air mass prior to sampling. Bacterial cells present more constant concentrations than INPs, from 2.4·103 to 6.8·104 cells mL-1, with their numbers increasing mainly under high wind speed. INPs are more efficiently removed than bacterial cells by precipitation, which implies a larger variability, a shorter residence time in the atmosphere and shorter lengths of dispersal for INPs rather than for bacterial cells. P. syringae has been successfully isolated at high-altitude and its presence seems to be influenced by uptake and removal processes, as it happens for INPs-8. This study constitutes a strong improvement of our understanding of the abundance, variability and nature of biological INPs in precipitation and points at the potential for this group of INPs to impact cloud processes. In fact, a coincidence of high wind speed and first precipitation often occurs at the passage of a front, where the meteorological conditions are also favourable to precipitation. This can be the ideal and frequent context where to expect large numbers of INPs-8 and to study their effects on cloud processes. Furthermore, bacterial cells can contribute to the number of INPs-8, but a large fraction of INPs-8 is potentially due to cellular fragments and macromolecules, both freely floating and attached to mineral and soil dust. This multiplies the possibility for biological INPs to be released and be present in the atmosphere