Scientists have created a new artificial intelligence tool that will help to more accurately predict the state of sea ice in the Arctic for several months in advance.
Sea ice at the North and South Poles is difficult to predict due to its complex relationship with the atmosphere above and the ocean below. The sensitivity of sea ice to rising temperatures has resulted in the area of summer Arctic sea ice being cut in half over the past 40 years. The affected region is 25 times the size of the UK.
All of these changes affect the climate, Arctic ecosystems, and the lives of indigenous and local communities whose livelihoods are linked to the seasonal sea ice cycle. Experts are confident that improved forecasts will form the basis of new early warning systems that protect Arctic wildlife and coastal communities.
An international team of researchers led by the British Antarctic Survey (BAS) and the Alan Turing Institute has created the IceNet artificial intelligence system. It solves the problem of making accurate forecasts of sea ice in the Arctic for the coming season. According to the authors of the development, IceNet predicts the presence of sea ice two months in advance with an accuracy of almost 95%.
The authors developed IceNet based on the concept of deep learning. Through this approach, AI “learns” how sea ice changes based on climate simulations over tens of thousands of years and over decades.