Super micron Amazonian aerosol particles as efficient catalyzers at cirrus temperatures

The discovery that supermicron particles collected above the Amazon rainforest can act as efficient catalysts for ice formation at cirrus-relevant temperatures adds an important piece to a long-standing puzzle in atmospheric science. Cirrus clouds, which form in the upper troposphere at temperatures below approximately minus 38 degrees Celsius, are difficult to study because the processes that initiate ice within them occur on microscopic scales and under conditions that are challenging to reproduce in the laboratory. Homogeneous freezing of aqueous solution droplets has traditionally been considered the dominant pathway for ice formation in these cold, humid environments, particularly in remote regions far from continental sources of dust and other ice-nucleating particles. The new findings from the Amazon suggest that this picture may be incomplete, even in one of the most pristine continental atmospheres on Earth.

The Amazon boundary layer is a well-known source of biogenic aerosol. Vegetation emits volatile organic compounds such as isoprene and terpenes, which oxidize in the atmosphere to form secondary organic material that condenses onto existing particles. In addition, the forest releases primary biological particles, including pollen fragments, fungal spores, plant debris, and microorganisms, some of which are large enough to fall into the supermicron size range. Previous work over the Amazon, notably the extensive GoAmazon2014/5 campaign near Manaus, documented the rich composition of the regional aerosol population and its transformation as urban pollution interacted with the natural background. The new study extends this knowledge upward, examining particles that are transported aloft and become available for cloud processes at altitudes where cirrus formation occurs.

Supermicron particles have often been neglected in ice nucleation studies, partly because their number concentrations are low compared with the abundant submicron accumulation mode, and partly because instrumentation for studying them is technically demanding. Large particles settle out of the atmosphere relatively quickly and are efficiently removed by precipitation, which the Amazon experiences in abundance. Nevertheless, deep convective systems over the tropical forest are exceptionally effective at lifting surface-derived material to the upper troposphere, where it can be injected into cirrus-generating layers. This vertical transport pathway means that even sparse populations of large biogenic particles may be present at exactly the altitudes and temperatures where cirrus clouds form, giving them a disproportionate influence relative to their concentration.

The term catalyzer in the context of the study refers to the ability of these particles to lower the energy barrier for ice nucleation without necessarily being consumed in the process. In classical heterogeneous ice nucleation, an insoluble surface provides a template onto which water molecules arrange into the crystalline structure of ice. Effective ice-nucleating surfaces share features such as lattice compatibility with ice, hydroxyl-rich functional groups, and nanoscale surface roughness that stabilizes ice-like molecular arrangements. Many biological materials, including certain proteins and polysaccharides found in pollen, fungi, and plant litter, exhibit these characteristics. If the Amazonian supermicron particles contain such biological residues or organic coatings with suitable surface chemistry, their catalytic efficiency at cirrus temperatures becomes physically plausible.

What makes the reported efficiency particularly noteworthy is the temperature regime. At temperatures between roughly minus 40 and minus 60 degrees Celsius, the competition between homogeneous and heterogeneous freezing determines cloud composition. Homogeneous freezing requires high supersaturations with respect to ice and produces large numbers of small ice crystals. Heterogeneous nucleation, by contrast, can occur at lower supersaturations and typically produces fewer ice crystals, because only the small fraction of particles carrying active ice-nucleating sites freezes. When heterogeneous nucleation competes successfully with homogeneous freezing, it depletes water vapor before the homogeneous threshold is reached, resulting in cirrus clouds with lower ice crystal number concentrations and larger crystals. Such clouds have different radiative properties: they tend to be optically thinner and longer-lived, and their net effect on the climate system, which balances the trapping of longwave radiation against the reflection of incoming sunlight, shifts accordingly.

The climate implications of this shift are substantial. Cirrus clouds exert a net warming influence on the planet because their greenhouse effect generally outweighs their albedo effect. Reducing ice crystal numbers in cirrus, as heterogeneous nucleation does relative to homogeneous freezing, tends to thin these clouds and weaken their warming contribution. This is the physical basis for proposed cirrus cloud thinning geoengineering schemes, which would seed the upper troposphere with ice-nucleating particles to suppress homogeneous freezing. The Amazonian results suggest that nature may already perform a version of this intervention over the tropical continents, where biogenic supermicron particles are lofted by convection. Quantifying this natural modulation is essential before any deliberate manipulation of cirrus clouds could be responsibly evaluated, because the background state against which an intervention would act is itself variable and biologically controlled.

The Amazon atmosphere is not static, and the efficiency of this particle-driven ice nucleation likely varies with season, meteorology, and human influence. During the wet season, the boundary layer is washed by frequent rain, and the aerosol population is dominated by biogenic material with relatively low number concentrations but substantial mass in the coarse mode. During the dry season and especially during biomass burning episodes, smoke particles dominate, and the supermicron fraction may include ash and charred organic material whose ice-nucleating properties differ from those of fresh biological particles. Land-use change, deforestation, and agricultural expansion alter the emission of primary biological particles, potentially changing the supply of ice-nucleating material to the upper troposphere. A future Amazon that is more fragmented, more fire-prone, and less humid may deliver a different mix of particles to cirrus-forming levels, with consequences for regional cloud radiative forcing that are only beginning to be explored.

Methodologically, studies of this kind typically combine offline analysis of collected particles with controlled laboratory exposures. Electron microscopy coupled with energy-dispersive X-ray analysis reveals particle morphology and elemental composition, while immunochemical or spectroscopic techniques can identify specific biological markers. Ice nucleation experiments are conducted in cloud chambers or on continuous-flow diffusion chambers, where particles are exposed to controlled temperature and humidity while ice formation is monitored optically. The distinction between immersion freezing, in which a particle is fully embedded in a liquid droplet before freezing, and deposition nucleation, in which ice forms directly from the vapor on a dry surface, is important because the mechanism operative in cirrus conditions has been debated for decades. Evidence that organic-rich particles can nucleate ice through pore condensation and freezing, in which water condenses in narrow surface pores and freezes there, has unified some previously conflicting observations and may be relevant to the porous structure of biological particles.

The Amazonian findings also connect to a broader reassessment of the global ice-nucleating particle budget. Global models have historically relied on parameterizations anchored to desert dust and, to a lesser extent, marine organic aerosol. Dust dominates the ice-nucleating particle population in many regions, but its transport to the remote tropical upper troposphere is episodic and seasonally variable. Biological particles, though far less numerous, can be active at warmer temperatures than dust and may contribute disproportionately in dust-poor environments. The recognition that supermicron biogenic particles from a rainforest can catalyze ice at cirrus temperatures implies that biosphere-atmosphere coupling extends into a temperature regime previously considered the domain of pure physical chemistry. In effect, the forest canopy influences not only low clouds through its emissions of condensable vapors and cloud condensation nuclei, but also high clouds through the upward transport of its own biological debris.

This vertical linkage depends on convective detrainment, the process by which rising air parcels shed moisture and aerosol into the surrounding environment near cloud tops. Tropical deep convection over the Amazon routinely penetrates to the level of neutral buoyancy near the tropical tropopause, and anvil clouds spreading from these storms persist for hours to days. Particles released into anvil outflow are subject to further transport by the upper-level winds, and some fraction survives the journey because supermicron biological particles are generally insoluble and resistant to the freezing and drying cycles they experience in convective updrafts. The residence time of such particles in the upper troposphere is uncertain, but even a few days of survival is sufficient for regional dispersal across much of the South American continent and beyond.

Future research directions follow naturally from these results. Direct measurements of ice-nucleating particle concentrations in the Amazonian upper troposphere, ideally from aircraft during both wet and dry seasons, would constrain the vertical distribution that models require. Molecular characterization of the active surfaces would identify which biological components carry the catalytic activity, enabling more targeted parameterizations. Long-term monitoring at mountaintop or balloon-borne stations could track seasonal variability in bioaerosol transport. Finally, coupling emission inventories of primary biological particles to global climate models with prognostic cirrus schemes would allow the radiative consequences of forest-derived ice nucleation to be quantified, and would reveal how sensitive this pathway is to the ecological changes now sweeping through the world’s largest tropical forest.

Subject of Research: Super micron Amazonian aerosol particles as efficient catalyzers at cirrus temperatures

Article Title: Super micron Amazonian aerosol particles as efficient catalyzers at cirrus temperatures

Article References: Ladino, L. A., Mendoza-Téllez, S., Ramírez, E., Negrete-Harper, E., Raga, G. B., Ruiz, A., Barrera, J., Wang, B., Vitar, J., Ramirez, O., Russy-Velandia, L., Alvarez-Ospina, H., Martínez, B., González del Castillo, E., Alvarado, C., Miranda, J., Meza, D., Reynoso-Cruces, S., Rosas, I., … Cortes, D. (2026). Super micron Amazonian aerosol particles as efficient catalyzers at cirrus temperatures. Communications Earth & Environment. https://doi.org/10.1038/s43247-026-04019-4

Image Credits: AI Generated

DOI: 10.1038/s43247-026-04019-4

Keywords: Super, micron, Amazonian, aerosol, particles, efficient, catalyzers, cirrus, temperatures, scientific research