Researchers have developed a novel method to detect and study how ice forms in mixed-phase clouds, significantly boosting scientists' ability to forecast weather and model climate change.
"Clouds are vital to Earth's climate and water cycle, influencing both rainfall and the planet's energy balance," said Devendra Pal, postdoctoral researcher and seasonal course lecturer at McGill's Department of Atmospheric and Oceanic Sciences. "But mixed-phase clouds are hard to understand and model, partly because researchers still don't fully know how smaller ice crystals form in them or behave over time."
"By observing how these nano-crystals form, grow and scatter sunlight, we can improve the accuracy of weather forecasts and climate models," said Pal, who co-led the study with Parisa Ariya, James McGill Professor of Chemistry and Atmospheric and Oceanic Sciences.
"This discovery matters for everyone affected by changing weather patterns - from farmers and city planners to disaster response teams and climate policymakers," Pal said.
Had been too small to observe directly
Mixed-phase clouds contain both ice crystals and liquid supercooled droplets, or water droplets that remain liquid at temperatures below 0°C.
Ice formation in clouds begins with ice-nucleating particles (INPs) - tiny airborne particles, often at the nanometre scale, that act as seeds for freezing. These INPs trigger the formation of ice crystals that can grow from nanometres to millimetres.
Until now, these nanoscale ice crystals were too small to observe directly or distinguish from their liquid counterparts in real time. Instead, scientists were left to infer whether a particle was ice based on its size limit or analyze how the particle scattered light using a method called polarization ratio.
A new instrument to observe ice formation in real time
To overcome this challenge, the team developed the McGill Real-Time Ice Nucleation Chamber (MRINC). It simulates cloud conditions in the lab through precise control over temperature and humidity.
"Into this chamber, we introduced silver iodide (AgI) particles - known to trigger ice formation -and carefully adjusted the environment to replicate cloud-like conditions," Pal said.
The researchers then used a laser-based imaging technique called digital holographic microscopy to observe the formation of ice crystals in real time-previously impossible at this scale. The resulting holographic images were analyzed using AI-powered software, which could instantly determine whether each particle was an ice crystal or a liquid droplet, and assess its size, shape and surface texture.
"This is the first time such detailed microphysical information has been captured at this scale, and with such fine detail," said Pal.
Added Ariya: "Because of a lack of technologies, none of the climate change models could accurately predict these smaller ice crystals. MRINC enables us to advance our understanding of the interaction processes between ice nucleation, radiation and climate change."
Next stop: natural clouds
The findings open a new window into the earliest stages of ice formation in clouds, with implications for satellite cloud observations, precipitation studies and even cloud seeding, a form of manual weather modification.
As a next step, the team members are adapting the MRINC system for field studies aboard research aircraft and mountaintop observatories. They are also refining the instrument to make it even more precise and adaptable to different atmospheric conditions. These upgrades will allow researchers to track how various particles -such as pollution, dust or wildfire smoke - affect ice formation.
"Ultimately, our goal is to deliver real-time data that strengthens both climate science and operational weather forecasting," Pal added.
About the study
"Microphysical Detection of Nano-Ice Nuclei to Ice Crystals: A Platform for Ice Nucleation Research," by Devendra Pal, Parisa Ariya, Ryan Hall, Yevgen Nazarenko and Leonard Barrie, was published in npj Climate and Atmospheric Science, a Nature Partner Journal.
This study was funded by the Canadian Foundation for Innovation, the McGill Tomlinson Award, the European Space Agency, the Natural Sciences and Engineering Research Council of Canada and the National Research Council.
