IceNet – Machine Learning for Seasonal Sea Ice Forecasting

IceNet began in 2019 as a joint development 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 monthly sea-ice forecasts up to six months in advance.

Project status

While active development on IceNet ended in December 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 systems.

To see the ongoing programme of AI forecasting and modelling at the British Antarctic Survey, visit PolarSim – AI forecasting and modelling.

What IceNet showed

IceNet is a probabilistic deep learning system that forecasts monthly-averaged Arctic sea ice concentration up to six months ahead.

• Training data: combined climate simulations with observational datasets.
• Forecasting: produced monthly-averaged forecasts of Arctic sea ice concentration, up to six months ahead.
• Performance: outperformed a state-of-the-art dynamical model for seasonal summer sea ice forecasts.
• Applications: provided a foundation for conservation tools and early-warning systems.

Why this matters

Sea ice is a key indicator of climate change. Accurate forecasts help scientists, conservationists, and Arctic communities prepare for rapid changes in ice conditions.

Improved forecasting supports wildlife protection, shipping safety, and long-term climate research.

Resources

• Paper: Andersson, T. R. et al. Seasonal Arctic sea ice forecasting with probabilistic deep learning. Nature Communications 12, 5124 (2021).
• Code: open-source repositories for the paper and its pipeline, available through icenet.ai.

1. Predicting September 2021 Arctic sea ice using artificial intelligence

   Tom Andersson, a data scientist at British Antarctic Survey Artificial Intelligence (AI) Lab, shares the latest predictions from a new Arctic sea ice forecasting AI tool as this year’s Arctic […]

   Read more of: Predicting September 2021 Arctic sea ice using artificial intelligence

   

1. Artificial intelligence to help predict Arctic sea ice loss

   A new AI (artificial intelligence) tool is set to enable scientists to more accurately forecast Arctic sea ice conditions months into the future. The improved predictions could underpin new early-warning […]

   Read more of: Artificial intelligence to help predict Arctic sea ice loss

   

Dataset URL:

1. PolarSim – AI forecasting and modelling

   PolarSim delivers AI forecasting, modelling, downscaling and emulation tools for the polar regions

   Read more of: PolarSim – AI forecasting and modelling