IceNet is a deep learning sea ice forecasting system developed by the British Antarctic Survey and The Alan Turing Institute . The original IceNet research model, published in Nature Communications , was trained on climate simulations and observational data to forecast the next 6 months of monthly-averaged sea ice concentration maps. This version advanced the range of accurate sea ice forecasts, outperforming a state-of-the-art dynamical model (ECMWF SEAS5) in seasonal forecasts of summer sea ice, particularly for extreme sea ice events.

IceNet began in 2019 as a joint initiative between the British Antarctic Survey and the Alan Turing Institute. It was one of the earliest environmental forecasting models to demonstrate that deep learning could outperform traditional physics-based models, in this case producing daily and monthly sea-ice forecasts up to six months in advance. While active development on IceNet ended in 2025, all original research outputs, publications, and core open-source repositories remain fully accessible to the global scientific community.

The teams are now focused on building next-generation, sustainable forecasting models and pipelines through PolarSim at the British Antarctic Survey and CryoCast at The Alan Turing Institute.