A raindrop hits bare soil and, in the fraction of a second before it soaks in, traps tiny air bubbles beneath its surface. Those bubbles rise, burst, and fling thousands of microdroplets skyward, each one potentially loaded with bacteria and fungal spores. It is a process that plays out billions of times during every storm, across every field and forest floor on the planet. And according to a growing body of research, some of the microbes riding those droplets may carry molecular tools that help clouds produce more rain.
A study published in Science Advances identifies a previously unknown class of ice-nucleating proteins produced by Mortierellaceae, a family of fungi found in soils worldwide. In laboratory tests, these proteins triggered ice crystal formation at temperatures warmer than those typically required by mineral dust or other known biological ice nucleators. That temperature range matters because it overlaps with conditions inside mixed-phase clouds, where liquid water and ice coexist and where much of Earth’s rainfall begins.
From dirt to sky: the launch mechanism
For soil fungi to influence clouds, they first have to get off the ground. Two experimental studies published in Nature Communications explain how that happens. In one, researchers showed that raindrop impacts on soil aerosolize bacteria through bubble-driven jetting. When a drop strikes porous ground, trapped air bubbles rise through the thin water film and pop at the surface, launching microdroplets that carry living microorganisms into the lower atmosphere.
A companion study used high-speed imaging to capture the splash dynamics frame by frame, confirming that a single raindrop can loft thousands of tiny droplets, many of them ferrying microbial passengers. The work established raindrop-driven aerosolization as a repeatable, well-documented physical process rather than a theoretical curiosity.
Proteins with a borrowed past
The Science Advances study goes beyond identifying the fungal proteins. Genetic and phylogenetic analyses traced their ancestry back to bacteria, suggesting the genes were picked up through horizontal gene transfer at some point in evolutionary history. That finding implies ice-nucleation ability has been shared across microbial kingdoms for a very long time, possibly because organisms that can catalyze ice gain a survival edge by influencing local moisture conditions around them.
The best-known biological ice nucleator is a protein made by Pseudomonas syringae, a bacterium that has been studied for decades and is even used commercially in snowmaking. The discovery of a structurally distinct class of ice-nucleating proteins in fungi broadens the picture considerably. It suggests the atmosphere’s biological ice-nucleation toolkit is more diverse than researchers assumed, with multiple microbial lineages independently acquiring or sharing the capacity to seed ice.
Storm-chasing for bioaerosols
Field observations add real-world weight to the laboratory findings. A campaign called BACS, conducted in Northern Colorado during May and June of 2022 and 2023, tracked how convective storms reshape bioaerosol populations. Results were published in an open-access paper that appeared recently in the PubMed Central archive (the unusually high PMC identifier reflects its very recent publication date). The study showed that storm dynamics and precipitation altered both the concentration and the cloud-relevant properties of biological particles in the air. The researchers documented bioaerosols acting as cloud condensation nuclei and ice-nucleating particles during active convective events.
The BACS team did not isolate Mortierellaceae proteins specifically. But their data confirmed that living and once-living particles are active components of storm clouds, not passive contaminants drifting through them. That distinction matters: it means biological material is not just present at cloud altitudes but is participating in the microphysical processes that determine whether a cloud produces rain.
Where the evidence trail goes cold
No published study has yet captured the full journey from soil fungus to cloud ice crystal in a single observational dataset. The Science Advances paper proves the proteins work in the lab. The Nature Communications experiments prove raindrops can launch microbes skyward. The BACS campaign proves bioaerosols are cloud-active during storms. But the bridge connecting these findings still relies on inference rather than a continuous measurement chain.
Whether Mortierellaceae proteins reach sufficient concentrations at cloud height to meaningfully influence precipitation rates remains an open question. Mortierellaceae fungi are common in temperate and agricultural soils, but their distribution varies by region, season, and land use. How much their ice-nucleating proteins contribute to rainfall compared with mineral dust, soot, or bacterial nucleators like those from Pseudomonas syringae has not been quantified under field conditions. In some ecosystems, particularly those with rich organic soils and intensive agriculture, fungal ice nucleators could be locally significant while remaining minor players at the planetary scale.
The most provocative implication is a self-reinforcing feedback loop: rain launches fungi, fungi seed more rain. The idea is consistent with the available evidence, but proving it would require simultaneous ground-level emission measurements, vertical atmospheric profiling, and cloud-microphysics observations during the same storm systems. That kind of coordinated campaign has not been reported as of May 2026. Until it is, the “rain begets rain” cycle should be treated as a compelling hypothesis, not an established feature of the climate system.
Why soil health may be weather-relevant
For anyone tracking climate adaptation or agricultural planning in drought-prone regions, the practical thread running through this research is straightforward: soil health and atmospheric science may be more connected than standard weather models account for. Current precipitation forecasts rely heavily on temperature, humidity, and large-scale circulation patterns. Biological ice nucleation is not yet a variable in operational weather models.
If future field campaigns confirm that fungal ice nucleators routinely reach cloud altitudes and significantly influence ice formation, land management decisions that reshape soil microbial communities, such as tilling practices, pesticide application, or conversion of grassland to pavement, could carry knock-on effects for local rainfall patterns. That possibility reframes soil not just as a medium for growing crops but as a living surface that interacts with the atmosphere in ways scientists are only beginning to measure.
The picture emerging from these studies is one in which the ground and the sky are part of a shared biological system. Microscopic fungi and bacteria, organisms most people never think about, may be quietly helping to stitch together the water cycle from below.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
