Cambridge scientists have created an artificially intelligent algorithm that can beat scientists at predicting ice crystal formation.
Climate change models could be improved thanks to the IcePic program. The details are in the journal.
When water becomes ice, it expands. Understanding water and how it is frozen around different Molecules has wide-reaching implications in a wide range of areas, from weather systems that can affect whole continents to storing biological tissue samples in a hospital.
The Celsius temperature scale is based on the idea that it is the transition temperature between water and ice, but water doesn't freeze at 0C. Predicting the ice nucleation ability of different materials is one of the main goals of the field.
A tool developed by researchers at the University of Cambridge was able to beat scientists in an online quiz in which they were asked to predict when ice would form.
Artificial intelligence learns from raw data. It can process results faster and more precisely because it finds its own patterns in the data. Different ice crystal formation properties can be inferred by IcePic. Icepic has been trained to look at thousands of images so that it can make accurate predictions.
Scientists were asked to predict when ice crystals would form in different conditions by looking at different images. The performance was measured against these results. IcePic was more accurate in determining a material's ice nucleation ability than researchers from all over the world. Where humans were going wrong was identified by it.
The first author of the study, Michael Davies, is a PhD student in the ICE lab at the University College London.
Humans have a 75 year head start in terms of science, but IcePic was still able to do something we couldn't.
Climate change research uses determining the formation of ice as an important factor.
In the form of rain and snow, water continually moves within the Earth's atmosphere. Smoke particles from pollution are different from smoke particles from a volcano. Understanding how different conditions affect our cloud systems can help us make better weather predictions.
The nucleation of ice is an important part of atmospheric science. There isn't a viable way to predict ice nucleation at the moment. Icepic should allow more applications for discovery.
More information: Accurate prediction of ice nucleation from room temperature water, Proceedings of the National Academy of Sciences (2022). DOI: 10.1073/pnas.2205347119. Journal information: Proceedings of the National Academy of Sciences
All Rights Reserved © 2024
