Biogenic ice nucleating particles in agricultural soils: Microbial drivers across different management practices and contribution to bioaerosol emissions

Atmospheric ice nucleating particles (INP) trigger ice formation in supercooled water, influencing cloud microphysics and precipitation. While mineral particles are abundant in the atmosphere, microbially produced compounds have been linked to INP with an ability to nucleate ice at temperatures >-10°C. Agricultural soils, which cover roughly 37% of Earth’s land surface, have been identified as reservoirs of these potent biological INP (bioINP). However, it remains unclear how site conditions and common agricultural practices, such as tillage, cover cropping, and liming, influence the abundance of bioINP and whether these effects are linked to changes in microbial community composition. To address this, we quantified INP concentrations and properties in soils from five Danish long-term agricultural field trials (LT sites) and characterized microbial communities using high-throughput sequencing of 16S rRNA and ITS marker genes. In addition, we established two aerosol sampling sites (AE sites) in agricultural areas in Denmark to monitor soil-derived bioaerosol emissions. We collected aerosol, soil, plant, and faecal samples at these two sites and used high-throughput 16S rRNA sequencing to quantify the fraction of bioaerosols derived from soils.

BioINP concentrations varied over several orders of magnitude between sites, with particularly high levels observed in Flakkebjerg, the most clay-rich of the five LT sites. Soil management practices influenced microbial community composition but had limited and inconsistent effects on bioINP abundance. Bacterial taxa previously reported as ice nucleation active, including Pseudomonas and Lysinibacillus, were detected only at low relative abundances and showed weak correlations with BioINP activity. In contrast, fungal community composition was a stronger predictor, with the relative abundance of Fusarium spp. and taxa like Linnemannia significantly associated with elevated BioINP concentrations. At the AE sites, we applied the source-tracking tool FEAST and found that 8% of the bioaerosols could be tracked to agricultural soils regardless of the season. This suggests that soil-derived particles are efficiently transferred into the atmosphere, even when soils are not directly exposed due to vegetation cover.

Our results suggest that fungi, particularly Fusarium, are the dominant contributors to warm-temperature bioINP activity in agricultural soils and that agricultural soils serve as an important source of airborne bioaerosol particles. This study, therefore, highlights the need to consider fungal ecology when linking agricultural management to atmospheric processes. Future work will quantify the absolute abundance of Fusarium using qPCR, determine fluxes of soil-derived BioINPs under different wind and rainfall conditions, and assess their impacts on cloud dynamics and climate.