Learning What Matters: Causal Time Series Modeling for Arctic Sea Ice Prediction
This work addresses the need for more robust and interpretable forecasting in climate science, particularly for Arctic sea ice, though it is incremental as it builds on existing causal methods within a hybrid neural architecture.
The paper tackled the problem of improving Arctic Sea Ice Extent prediction by addressing limitations of correlation-based models, introducing a causality-aware deep learning framework that integrates causal feature selection methods, and achieved improved prediction accuracy and interpretability across varying lead times using 43 years of data.
Conventional machine learning and deep learning models typically rely on correlation-based learning, which often fails to distinguish genuine causal relationships from spurious associations, limiting their robustness, interpretability, and ability to generalize. To overcome these limitations, we introduce a causality-aware deep learning framework that integrates Multivariate Granger Causality (MVGC) and PCMCI+ for causal feature selection within a hybrid neural architecture. Leveraging 43 years (1979-2021) of Arctic Sea Ice Extent (SIE) data and associated ocean-atmospheric variables at daily and monthly resolutions, the proposed method identifies causally influential predictors, prioritizes direct causes of SIE dynamics, reduces unnecessary features, and enhances computational efficiency. Experimental results show that incorporating causal inputs leads to improved prediction accuracy and interpretability across varying lead times. While demonstrated on Arctic SIE forecasting, the framework is broadly applicable to other dynamic, high-dimensional domains, offering a scalable approach that advances both the theoretical foundations and practical performance of causality-informed predictive modeling.