Scientists have developed an AI tool, named IceNet, that allows us to predict the melting patterns of Arctic sea ice, which could be a real game changer when it comes to climate change.
Following the release of the IPCC report earlier this year, we’re more aware of the consequences of climate change than ever before. And while we can do our best to minimise the impacts long-term, it is becoming increasingly clear that we are and will continue to see melting in the Arctic.
So if it’s going to happen, we may as well have a way to forecast the sea ice conditions and, in turn, try our hardest to protect wildlife and civilians from copping the worst of it.
The IceNet tool is developed by the British Antarctic Survey (BAS) in collaboration with The Alan Turing Institute and aims to forecast conditions as much as six months in advance.
IceNet is trained on observational data and climate simulations in order to predict melting patterns and, consequently, where the now-melted ice will end up.
According to the study’s lead author, Tom Andersson, the technology runs thousands of times faster than the leading physics-based versions. Not to mention, the system recorded up to 95% accuracy when predicting conditions two months in advance.
“IceNet could run on a laptop, while previous physics-based methods would have to run for hours on a supercomputer to produce the same forecasts,” Andersson explained in a YouTube video.
The technology is particularly useful considering the Arctic is warning 2-3 times faster than the rest of the planet. And considering the amount of ice in the Arctic has halved in the last 40 years and shows no sign of slowing down, it’s pretty imperative that we develop this sort of technology to track it.
There’s no suggestion that IceNet will do anything to slow down the impacts of climate change in the Arctic region, but it may help to minimise the human impact faced by those most vulnerable.