Advancing climate model interpretability: Feature attribution for Arctic melt anomalies Conference: Make sure to enter the correct conference title from your rights confirmation email; ;
Tolulope Ale, Nicole-Jeanne Schlegel, Vandana P. Janeja · 2025
Read Advancing climate model interpretability: Feature attribution for Arctic melt anomalies Conference: Make sure to enter the correct conference title from your rights confirmation email; ; by Tolulope Ale, Nicole-Jeanne Schlegel, Vandana P. Janeja (2025) 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/arxiv-2502-07741.txt.
Description
Tolulope Ale, Nicole-Jeanne Schlegel, and Vandana Janeja work on explaining why climate models flag unusual Arctic melt events. Their 2025 paper compares two models of Greenland surface melt: ERA5, a broad reanalysis product, and GEMB, a model with richer snow physics. Knowing which factors drive a melt anomaly helps scientists trust or fix each model.
The team built an unsupervised method that explains anomalies through counterfactual reasoning, asking what would have to change for the event to look normal. Ground-truth data checked the anomaly calls. The explanations were weighed against standard ranking tools such as Shapley values and random forests. The approach, they report, exposes the physics each model leans on.
Catalog record
| CREATOR | Tolulope Ale, Nicole-Jeanne Schlegel, Vandana P. Janeja |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-07741