A team of atmospheric and microbiological researchers, led by Alex Huffman of the University of Denver, Anna Kunert, and Janine Fröhlich-Nowoisky of the Max Planck Institute for Chemistry, has identified proteins produced by common soil fungi that can trigger ice crystal formation inside clouds at temperatures warm enough to initiate rainfall. The findings, published in Science Advances in April 2026, center on fungi in the genus Mortierella, one of the most abundant fungal groups in agricultural and forest soils worldwide. The discovery challenges a decades-old assumption that bacteria, particularly species like Pseudomonas syringae, are the dominant biological agents of ice formation in the atmosphere.
“We were surprised to find that these fungal proteins are not only potent ice nucleators but also completely independent of the living cell,” Huffman said in a Virginia Tech press release. “That changes the picture of how biology might influence clouds and precipitation.”
For climate scientists refining precipitation models and farmers tracking drought forecasts, the research introduces a biological variable that has gone unaccounted for: tiny, water-soluble proteins that do not need a living cell to function and that may be entering the atmosphere in vast numbers wherever soil is disturbed by wind, tilling, or erosion.
What the research established
The Science Advances paper, with contributions from researchers at the Max Planck Institute for Chemistry, the University of Denver, and Virginia Tech, used ice nucleation assays combined with genomic and molecular analysis to identify a new class of ice-nucleating proteins, or INPs, from Mortierella fungi. These proteins catalyze ice formation at relatively warm subzero temperatures, in the range where mixed-phase clouds produce most rainfall.
What distinguishes these fungal proteins from their well-studied bacterial counterparts is structural. Bacterial INPs, such as the InaZ protein in Pseudomonas syringae, are anchored in cell membranes and generally require intact bacterial surfaces to nucleate ice effectively. The fungal proteins identified in this study are water-soluble and cell-free. They function after separating from the organism that made them, which has significant implications for how far they can travel.
A related study published in the Proceedings of the National Academy of Sciences provided the mechanistic explanation. Fröhlich-Nowoisky and colleagues showed that fungal ice nucleators exist as small protein subunits that self-assemble into larger functional aggregates outside the cell. The size and structure of these aggregates directly influence the temperature at which ice first forms. Because the proteins retain their potency after the parent fungus has died or fragmented, they can persist in soil and potentially in the atmosphere long after the organism is gone.
Separate research published in Scientific Reports in 2015 demonstrated that fungal ice-nucleating activity persists in nanoscale fragments far more numerous than intact spores or hyphae. That earlier work established a key physical principle: a whole fungal spore is relatively heavy and unlikely to reach cloud altitude in large numbers, but nanoscale protein fragments are light enough to be lofted by wind into the lower atmosphere, where they could serve as ice nuclei without requiring the organism itself to be airborne.
Ecological fieldwork ties the laboratory results to real-world conditions. A 2015 study in Biogeosciences confirmed that ice-nucleation-active fungi, including Mortierella alpina, are widespread across multiple soil environments. Mortierella species thrive in temperate agricultural soils, boreal forests, and grasslands, meaning their protein fragments could be entering the atmosphere continuously across large swaths of the planet’s land surface.
What remains uncertain
The most significant gap is the absence of direct atmospheric measurements. Every piece of verified ice nucleation data in this body of research comes from laboratory assays and soil sampling, not from instruments capturing fungal protein concentrations inside actual clouds. Scientists have demonstrated that these proteins work under controlled conditions and that the source fungi are abundant in the ground. The critical middle step, quantifying how much fungal INP material reaches cloud altitude across different regions and seasons, has not been documented.
A second open question involves competition. Clouds contain a complex mix of potential ice nuclei: mineral dust, soot, pollen, bacterial proteins, and now fungal proteins. No field campaign or integrated simulation in the current evidence base has measured how fungal INPs interact with these other nucleators under real atmospheric conditions. The Science Advances paper frames potential roles for fungal INPs in climate models, and institutional press coverage from Virginia Tech has described possible applications ranging from improved rainfall prediction to bio-inspired cloud seeding. Those remain projections, not validated outcomes.
Scale presents its own challenge. Translating soil abundance into atmospheric flux requires data on emission rates, particle size distributions, and vertical transport. The Scientific Reports work on nanoscale fragments suggests that fragmentation dramatically increases the number of potential ice-active particles, but it does not quantify how many actually reach cloud-forming altitudes. Targeted sampling campaigns using research aircraft or high-elevation observatories would be needed to close that gap.
Biological variability adds another layer of uncertainty. Fungal communities differ from field to field and season to season, shaped by soil type, crop rotation, moisture, and land management. Whether all Mortierella strains produce equally potent ice-nucleating proteins, or only certain lineages are strongly active, is not yet resolved. Environmental stresses such as drought, heavy fertilization, or pesticide application could alter fungal abundance and protein output in ways that either amplify or suppress atmospheric effects. None of the cited studies systematically map these ecological controls.
What fungal ice nucleators mean for precipitation models
Four primary sources form the backbone of this research as of May 2026: the Science Advances paper supplies the core discovery and genomic identification; the PNAS study adds the self-assembly mechanism; the Scientific Reports paper explains why whole organisms are not required; and the Biogeosciences ecology study anchors the lab work in real-world fungal distribution. Together, they build a coherent chain from molecular mechanism to environmental plausibility.
For climate modelers, the practical implication is that existing precipitation simulations may be missing a biological input. If fungal INPs prove to be as widespread in the atmosphere as they are in soils, models that account only for mineral dust and bacterial nucleators could be systematically underestimating ice formation over agricultural regions. That underestimation would affect drought frequency projections and water resource planning. Correcting it would require field campaigns to measure fungal INP concentrations at altitude, followed by integration into regional and global models with sensitivity tests comparing scenarios with and without fungal contributions.
For anyone tracking water supply forecasts or agricultural planning, the research does not yet change operational guidance. What it does is identify a specific, testable gap in how scientists predict rain and snow: the potential role of fungal proteins as overlooked ice starters in the sky.
Over the next several years, the developments that will matter most are not technological spin-offs but basic measurements. How many fungal protein particles are actually present in clouds? How often do they trigger freezing compared with mineral dust or bacterial nuclei? How does their abundance shift with land use and seasonal cycles? Until those data arrive, fungal ice nucleators represent a biologically grounded hypothesis that extends the known connections between soil ecosystems and the water cycle. The proteins are real, the mechanism is clear, and the source organisms are everywhere. What scientists still need to prove is that the journey from soil to sky happens at a scale large enough to shape the weather.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
