3D printed ice crystals could improve weather forecasts

Accurate weather forecasts are important to global business operations, and 3D printed ice crystals may help to improve those predictions.
22 January 2024

Cool tech: 3D-printed ice crystals reveal hidden weather secrets that could improve forecasts made on supercomputers.

Getting your Trinity Audio player ready...

3D printers enable industrial designers to realize all kinds of shapes and intricate patterns by forming custom parts layer by layer. Topologies produced include model weights for optical neural networks, which highlights some of the cutting-edge applications being explored. Most recently, researchers in Europe have successfully used the versatile manufacturing technique to fabricate 3D printed ice crystals.

The tiny non-spherical 3D printed ice crystals open the door to much more accurate weather forecasting when used in combination with a high-speed camera setup capable of capturing their motion as they tumble through the air. Small particles, known as aerosols, have a profound effect on weather systems, particularly on cloud dynamics – which is a major source of uncertainty in weather forecasting.


On TechHQ, we’ve written about how massive supercomputers enable meteorologists to run ensemble models and make probabilistic forecasts based on those results. By clicking on a weather app, smartphone users can quickly see information such as the percentage chance of rain and have probabilistic forecasts to thank for that guidance.

However, the results of those meteorological models are only as good as the assumptions that underpin them. And, in the case of aerosols, some of those assumptions can be quite crude, which is where having 3D printed ice crystals (and other airborne particles) could turn out to be a big help.

In their work, the group – based in France, Germany, and Sweden – showed how non-spherical particles tumble down in the air, rotating first one way, and then the other, in an oscillatory fashion. Weather scientists have long been puzzled as to why dust in the sky is able to travel so far and why the particle size distribution of a column of air doesn’t match up with their theoretical predictions. And it’s noteworthy that models often assume that aerosols are spheres.

“The orientation of nonspherical particles in the atmosphere, such as volcanic ash and ice crystals, influences their residence times and the radiative properties of the atmosphere,” point out the researchers in their paper.

The thinking is that adding more detailed and accurate descriptions of how particles move could improve forecasting. The Earth’s atmosphere is, in essence, a giant fluid, and the evolution of that fluid can be described by physics – the tricky part is getting the inputs right. Even a small error in particle data can dramatically change calculations of how much water clouds carry, how far they travel, and whether that rain falls as a drizzle or a downpour.

3D printed ice crystals in motion

The motion of 3D printed ice crystals revealed using high-speed video cameras could play a role in improving weather forecasts and – given how sensitive supply chains can be to flooding, storms, and other climate events – those gains could buy businesses more time to make contingency plans. Companies are already making greater use of climate data in their longer-term planning, and providers are always looking for opportunities to enhance their offerings to firms.

Better weather forecasting also impacts the energy grid too. Electricity providers are keen to know how much wind and sun to expect so that they can make the most of renewable sources. More accurate meteorological models also help with demand forecasting – for example, by prompting operators to bring more power online to balance increased demand for air-conditioning during a heatwave.

Also, when floods and other disruptive weather events are forecast, energy providers have the opportunity to protect key infrastructure and have maintenance crews on standby to minimize outages.

Weather has been a talking point for centuries, with people picking out colors in the sky and seeing how birds behaved for clues. Today, the anatomy of an atmospheric model has become very complex indeed – so complex that it takes a supercomputer to make sense of the various physical and dynamic elements.

What’s more, adding to the list of necessary equipment could be a 3D printer – if data from 3D printed ice crystals turns out to have merit.