An artificial intelligence tool designed by an international team of researchers will enable scientists to more accurately forecast Arctic sea ice conditions months into the future. The improved predictions could underpin new early-warning systems that protect Arctic wildlife and coastal communities.
The team led by British Antarctic Survey and the Alan Turing Institute designed and trained the IceNet forecasting system, which they describe in "Seasonal Arctic Sea Ice Forecasting with Probabilistic Deep Learning," published in the journal Nature Communications.
IceNet is almost 95% accurate in predicting whether sea ice will be present two months ahead — better than the leading physics-based model.
"IceNet has the potential to fill an urgent gap in forecasting sea ice for Arctic sustainability efforts and runs thousands of times faster than traditional methods," says lead author Tom Andersson, data scientist at the BAS AI Lab.
From British Antarctic Survey
View Full Article
No entries found