Preprint · 2026 · Open access
Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting
Preprints.org · Published 27 August 2026 · DOI 10.20944/preprints202608.1982.v1
In brief
“Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting” is a 2026 preprint by Saghar Ganji and 3 co-authors, published in Preprints.org. OpenAlex counts 0 citations as of 20 November 2026. It is open access under a CC BY licence: the full text is free to read at Preprints.org.
- Citations
- 0
- Authors
- 4
Abstract
Reliable prediction of the El Niño–Southern Oscillation (ENSO) at multi-year lead times remains challenging because forecast skill generally declines beyond 12–18 months. This study presents a streamlined Informer-based framework designed to predict the Niño 3.4 index up to 36 months in advance using one-dimensional climate time series. The proposed model was developed following an analysis of the earlier Multimodal ENSO Forecast framework, which showed that most of the long-lead predictive skill originated from the time-series branch, whereas the spatial branch added substantial computational cost with limited benefit. Historical simulations from CMIP5 and selected CMIP6 models were used for training, followed by calibration with GODAS observations from 1980 to 2000 and independent validation over 2001–2020. The model achieved anomaly correlation coefficient values of 0.92, 0.75, 0.56, 0.46, 0.48, and 0.41 at lead times of 1, 6, 12, 18, 24, and 30 months, respectively. At directly comparable lead times, its performance was consistently higher than that of the CNN benchmark, with the largest difference observed at 18 months. The model also maintained useful predictive skill at extended lead times and showed reduced sensitivity to the spring predictability barrier. In addition, the simplified architecture required substantially less computational effort than the original multimodal framework and supported inference on a standard CPU. These results demonstrate that an efficiently optimized time-series architecture, combined with climate-model-based data augmentation, can provide competitive and computationally practical ENSO forecasts at multi-year horizons.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Ganji, S., Naisipour, M., Saeedpanah, I., & Adib, A. (2026). Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting. Preprints.org. https://doi.org/10.20944/preprints202608.1982.v1
MLA 9
Ganji, Saghar, et al. “Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting.” Preprints.org, 2026. https://doi.org/10.20944/preprints202608.1982.v1.
Chicago
Ganji, Saghar, Mohammad Naisipour, Iraj Saeedpanah, and Arash Adib. 2026. “Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting.” Preprints.org. https://doi.org/10.20944/preprints202608.1982.v1.
Harvard
Ganji, S. et al. (2026) ‘Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting’, Preprints.org. doi:10.20944/preprints202608.1982.v1.
IEEE
S. Ganji, M. Naisipour, I. Saeedpanah, and A. Adib, “Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting,” Preprints.org, 2026, doi: 10.20944/preprints202608.1982.v1.
BibTeX
@misc{ganji2026enhanced,
title = {{Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting}},
author = {Ganji, Saghar and Naisipour, Mohammad and Saeedpanah, Iraj and Adib, Arash},
howpublished = {Preprints.org},
year = {2026},
doi = {10.20944/preprints202608.1982.v1},
url = {https://doi.org/10.20944/preprints202608.1982.v1}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Environmental Science › Global and Planetary Change › Climate variability and models
- ENSO
- ENSO forecasting
- informants
- Niño 3.4 index
- spring predictability barrier
- computational efficiency
Work with this paper
Most-cited papers in Environmental Science
- Biodiversity hotspots for conservation prioritiesNorman Myers, Russell A. Mittermeier, Cristina Goettsch Mittermeier and 2 more · Nature · 2000 · 32,720 citations
- The Tragedy of the CommonsGarrett Hardin · Science · 1968 · 26,194 citations
- River flow forecasting through conceptual models part I — A discussion of principlesJ.E. Nash, John V. Sutcliffe · Journal of Hydrology · 1970 · 24,625 citations
- The value of the world's ecosystem services and natural capitalRobert Costanza, Ralph C. d’Arge, R.S. de Groot and 10 more · Nature · 1997 · 23,508 citations
- Climate change 2007: the physical science basisSusan Solomon, Dahe Qin, Martin Manning and 5 more · 2007 · 23,383 citations
Questions about this paper
How do I cite this paper in APA style?
Ganji, S., Naisipour, M., Saeedpanah, I., & Adib, A. (2026). Enhanced Informer-Based Deep Learning for Three-Year ENSO Forecasting. Preprints.org. https://doi.org/10.20944/preprints202608.1982.v1
How many times has this paper been cited?
0 times, according to OpenAlex on 20 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access at Preprints.org: sign in and open the full text from this page.
Who wrote this paper?
Saghar Ganji, Mohammad Naisipour, Iraj Saeedpanah, Arash Adib.
What is the DOI of this paper?
10.20944/preprints202608.1982.v1, permanently at https://doi.org/10.20944/preprints202608.1982.v1. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 20 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.