Someone taking a walk in the snowy hills near the Swiss capital of Bern last winter would have been perplexed by a giant white balloon rising into the low-lying clouds. Near the hilltop where the balloon was tethered, they would have found a group of researchers launching a drone, then flying it upwind to sprinkle crystals of silver powder into the fog.
鈥淲e use the natural cloud as a laboratory,鈥� says 聽at ETH Zurich in Switzerland, a member of the research group behind the unusual scene. Called , the group has found ways to use cloud-seeding methods normally employed to boost snow or rainfall to measure how ice crystals form in clouds with new precision.
Such information is key to improving weather and climate models. The rate at which water droplets in clouds form ice crystals determines when forecasts predict a cloud will begin to snow or to rain. Ice in clouds is also an important variable in climate models for estimating how much sunlight they reflect and thus how much they cool the planet. But the exact behaviour of ice in clouds remains a major uncertainty.
Advertisement
Lohmann and her colleagues have been trying to get better measurements for more than a decade. They have watched ice crystals grow in laboratory cloud chambers, but this lacks the variability of turbulence and temperature found in an actual cloud. They have taken measurements on cloud-shrouded mountaintops, but these were confounded by blowing snow. They鈥檝e even attached instruments to cable cars in the Alps, only to see them shut down during the bad weather they were trying to study.
鈥淭hen we thought, you know, we could actually be smarter,鈥� says Lohmann. 鈥淲e have this persistent low stratus in Switzerland.鈥� These steady, famously glum clouds regularly form above the Swiss plateau, funneled between the Jura mountains to the north and the Alps to the south. The researchers realised these clouds would enable them to perform the same kind of controlled experiments from the lab in an actual cloud 鈥� a kind of cloud chamber in the sky. 鈥淲e can perturb the cloud in a way that we want,鈥� says Lohmann.
In late 2022 and again the following winter, the researchers set up shop on a hilltop near the small town of Eriswil, east of Bern. When conditions were right, they flew a drone to release a small amount of silver iodide particles into the cloud. These particles act as seeds around which supercooled water droplets begin to grow ice crystals, eventually becoming heavy enough to fall out. The purpose of the balloon, several kilometres downwind of the drone, was to hoist a device to record 3D images of the crystals as they passed to distinguish different types of growth. Ground-based radar instruments also tracked the plume of ice as it fell through the cloud.
Normally, 鈥渋f you measure ice crystals in a real cloud, you never know if the ice crystal has formed there or if it comes from elsewhere,鈥� says Lohmann. This approach solved that problem, enabling them to make precise measurements of the rate at which ice crystals formed after the drone released its silver.
A key from these measurements is that ice crystals in reality may grow substantially faster than weather and climate models currently expect, says Lohmann. That could explain forecasts that underestimate how soon a cloud will rain or snow. It could also mean climate models are overestimating the cooling effects of some clouds, because clouds with more ice will reflect less sunlight than clouds with more liquid water, she says.
Their 3D measurements of individual crystals have also enabled the CloudLab researchers to the different way droplets of supercooled water form ice under different conditions, which also could have implications for the effect clouds have on the climate. 鈥淭hey would love to be ice,鈥� says Lohmann.
