Machine Learning for Sea Ice Forecasting

What IceNet does

IceNet is a deep learning system that forecasts Arctic sea ice. The system is trained on climate simulations and observational data. It currently produces daily forecasts of sea ice concentration up to two weeks ahead, and monthly-averaged forecasts up to six months ahead. IceNet outperforms state-of-the-art dynamical models, especially for extreme sea ice events.

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.

How the project works

IceNet uses machine learning to extend the accuracy and range of sea ice forecasts:

• Training data: combines climate simulations with observational datasets.
• Forecasting: produces daily and monthly averaged probabilistic forecasts of sea ice concentration.
• Performance: demonstrated to exceed traditional models, particularly for predicting summer sea ice loss.
• Applications: Provides a foundation for conservation tools and early-warning systems.

Science objectives

The project aims to:

• advance the accuracy of sea ice forecasts on timescales of days to months
• improve prediction of extreme sea ice events
• provide tools to support conservation and operational planning
• test AI approaches against traditional dynamical models
• deliver open resources for the scientific community

Who is involved

IceNet is led by British Antarctic Survey and the Alan Turing Institute. It involves an international team of scientists and AI researchers.

Science objectives

The project aims to:

• advance the accuracy of seasonal sea ice forecasts
• improve prediction of extreme sea ice events
• provide tools to support conservation and operational planning
• test AI approaches against traditional dynamical models
• deliver open resources for the scientific community

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. BAS Cambridge

   Science, business and operational planning

   Read more of: BAS Cambridge

   

2. Aurora Collaboration Space

   The Aurora Collaboration Space at British Antarctic Survey is a coworking space for members who share our passion for the environment.

   Read more of: Aurora Collaboration Space

   

3. Aurora Innovation Centre

   The Aurora Innovation Centre offers a space and support services for interdisciplinary collaboration to tackle environmental challenges and increase real-world benefit of polar research.

   Read more of: Aurora Innovation Centre

   

4. ICT facilities

   RRS Sir David Attenborough has enhanced satellite communications and data processing and storage systems to meet the increasing requirement for data-intensive operations.

   Read more of: ICT facilities