Deep learning identification of SST teleconnections driving early-winter North Atlantic climate
Seasonal predictability over the North Atlantic-European (NAE) sector is strongly modulated by the background climate state, particularly in early winter. In this season, different El Niño-Southern Oscillation (ENSO) teleconnections have been reported before and after the 1990s. However, these studies rely on linear analysis, and the reasons for this lack of stationarity and its implications for seasonal forecasting have not yet been fully assessed. Here, we evaluate the non-stationarity of early-winter ENSO teleconnections over the NAE using a non-linear framework, the Neural Network forecast application (NN4CAST). The NN4CAST model predicts November-December mean sea-level pressure anomalies from October sea surface temperature (SST) anomalies. The neural network achieves skill comparable to the ECMWF seasonal forecasting system (SEAS5), and even exceeds it in certain regions like the NAE, despite relying only on October SST predictors. The model skilfully predicts the dominant NAE atmospheric variability modes, with a predictability that is strongly period-dependent. In particular, significant skill is found over the North Atlantic Oscillation centres of action during 1970–1995, especially for the Iceland region, whereas for the East Atlantic pattern, the improvement comes in recent decades (1996–2019). This teleconnection shift is mainly associated with Pacific SST anomalies, with recent decades marked by an intensified Southeast Asian Rossby wave source and a southward displacement of the North Atlantic jet stream. Our results demonstrate that explainable artificial intelligence can identify state-dependent SST precursors of NAE predictability, bridging data-driven and mechanistic understanding of climate teleconnections.
Authors
- Belén Rodríguez‐Fonseca (ORCID: https://orcid.org/0000-0002-5261-7083)
- Magdalena Balmaseda (ORCID: https://orcid.org/0000-0002-9611-8788)
- Irene Polo (ORCID: https://orcid.org/0000-0002-6250-6109)
- Marta Martín‐Rey (ORCID: https://orcid.org/0000-0001-6234-0447)
- Víctor Galván Fraile (ORCID: https://orcid.org/0009-0008-7373-8029)
Institutions
- Universidad Complutense de Madrid (ES)
- Consejo Superior de Investigaciones Científicas (ES)
- European Centre for Medium-Range Weather Forecasts (GB)
- Instituto de Geociencias (ES)
- University of Reading (GB)
Publication Details
- Journal
- Scientific Reports
- Published
- 2026-11-14
- DOI
- https://doi.org/10.1038/s41598-026-78245-3
- Primary Topic
- Climate variability and models
- Type
- article
- Field-Weighted Citation Impact
- 0.00