Novel AI tool for predicting arctic ice loss

Months later, scientists developed a new artificial intelligence device that could accurately predict Arctic ice conditions. According to the British Antarctic Survey (BAS) and an international team led by the Alan Tourism Institute in the United Kingdom, the improved forecasts could support new early warning systems to protect Arctic wildlife and coastal communities from the effects of frost damage.

In the magazine cribed l Nature relationships, IENET, ISNET, will face challenges to produce future Arctic ice forecasts – something that has alienated scientists for decades.

The researchers said that the freezing temperatures of the northern and southern poles could not be predicted because of the complex interactions between the atmosphere and the ocean below.

He said the risk of sea frostbite has halved in the last four decades, compared to the fact that Britain has lost about 25 times the size of the UK. The researchers said that these accelerated changes have significant consequences for the global climate, the Arctic ecosystem, and the indigenous and local communities.

According to the researchers, it is 95 percent accurate to predict whether ice cream was discovered two months ago.

“The Arctic is a region that is at the forefront of climate change and has been extremely hot in the last 40 years,” said Tom Anderson, lead author of the study at BSS Laboratory. “Aysnet has the potential to fill an urgent gap by forecasting sea ice for Arctic sustainability efforts and is thousands of times faster than traditional methods,” Anderson said.

The new Snow Forecast Framework eliminates data from satellite sensors in a way that traditional systems could not easily achieve by making climate models more effective, said Scott Hosking, senior researcher at BAS AI Laboratory.

Aisnet Experiment – Current Arctic Snow Forecast by Tom Anderson et al. (2021) (

Contrary to the usual predictive systems that try to shape the laws of physics directly, the authors developed an Internet network based on the concept of in-depth learning. With this approach, the model “learns” how the sea ice will change from thousands of years of weather simulation data to predict the future of the Arctic ice months.

“We have now shown that AI can accurately forecast sea ice. Our next goal is to develop a daily model like the weather forecast and make it available in real time,” Anderson said. “This could serve as an early warning system for hazards,” he said.


Leave a Comment