Modelling Future Evolution of the Antarctic Ice Sheet With Explainable Machine Learning
Paula Maddigan, Bach Hoai Nguyen, Peter Yu Feng Siew, Bing Xue, Nicholas Golledge, Mengjie Zhang · 2026
Read Modelling Future Evolution of the Antarctic Ice Sheet With Explainable Machine Learning by Paula Maddigan, Bach Hoai Nguyen, Peter Yu Feng Siew, Bing Xue, Nicholas Golledge, Mengjie Zhang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13412399.txt.
Description
Paula Maddigan and co-authors train machine learning models in 2026 to stand in for a heavyweight Antarctic ice sheet simulator. The Parallel Ice Sheet Model predicts ice change well but runs slowly, limiting long forecasts as seas rise. The team taught models on simulator output spanning 2015 to 2100, feeding them climate inputs plus neighbouring and past ice values to predict ice cover, thickness, and speed.
A random forest using an expanding window performed best among the designs tried. Its thickness error averaged 3.36 units, and it mislabelled only about three grid points per year for ice cover out of 2,257 on average. Explainable AI tools show why predictions land as they do, which the authors say builds trust.
Catalog record
| CREATOR | Paula Maddigan, Bach Hoai Nguyen, Peter Yu Feng Siew, Bing Xue, Nicholas Golledge, Mengjie Zhang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |