Dissertation · 2019 · Open access
Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting
Utrecht University Repository (Utrecht University) · Published 1 January 2019
In brief
“Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting” is a 2019 dissertation by Paul Petersik, published in Utrecht University Repository (Utrecht University). OpenAlex counts 0 citations as of 20 November 2026. It is open access: the full text is free to read at Utrecht University Repository (Utrecht University).
- Citations
- 0
- Authors
- 1
Abstract
This study introduces two novel statistical models for the prediction of the El Niño-Southern Oscillation\n(ENSO). The first model is termed Deep Ensemble (DE). This neural network method was recently\ndeveloped by Lakshminarayanan et al. (2017). A DE is applied for the first time in the prediction of the\nENSO. Instead of only predicting one value for Oceanic Niño Index (ONI), as it is done by other statistical\nENSO models, the DE predicts the mean and the standard deviation of a Gaussian distribution. In this\nway, the forecast comes with an estimation of the predictive uncertainty, which is a novel feature in the\nprediction of the ENSO by statistical models. The predictor variables for the DE are chosen such that\nthey represent different aspects of the ENSO dynamics. The memory component of the subsurface is\nincluded by the use of the warm water volume (WWV) of the equatorial Pacific. As a driver of more\nstochastic effects, the area averaged zonal wind stress anomaly in the west Pacific (WP) is added to the\npredictor variables. To include information about the interdecadal changes in the background state of\nthe Pacific, the amplitude of the leading empirical orthogonal function (EOF) of the 5-year running-mean\nsea surface temperature anomaly (SSTA) field is used. Other predictor variables which are used are the\ndipole mode index (DMI) of the Indian Ocean Dipole (IOD), two variables from the theory of evolving\ncomplex networks as well as the ONI itself. The trained DEs show similar prediction skills as other\nstatistical ENSO models do that are currently used in operational ENSO forecasting. As expected, the\nDEs assign low predictive uncertainties to forecasts with a small lead time. In contrast, for very long lead\ntimes, the DEs predict the climatological distribution of the ONI. Unfortunately, decadal variations in\nthe estimated predictive uncertainty are not clearly visible within the forecasts. This can be attributed\nto the low amount of available training data.\nThe second model is a so-called Encoder-Decoder (ED) model which is inspired by the architecture of\nAutoencoders (AEs). The ED is used to predict the entire SSTA field in the Pacific ocean between\n30◦S-30◦ N and 120◦ E to 80◦ W. Therefore, the prediction provides information on the spatial pattern of\nthe anomalies. The model shows generally weaker skills than the DE regarding the ONI. Hence, it has a\nweaker prediction skill in comparison to other statistical models. However, it is still remarkable that the\nED can make skillful predictions given the little amount of data which was available for the training in\nrespect to the complexity of the model. This shows that the bottleneck architecture of the ED effectively\nprevented overfitting.\nBoth models are analyzed onto…
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Petersik, P. (2019). Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting. Utrecht University Repository (Utrecht University). http://dspace.library.uu.nl:8080/handle/1874/383388
MLA 9
Petersik, Paul. “Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting.” Utrecht University Repository (Utrecht University), 2019. http://dspace.library.uu.nl:8080/handle/1874/383388.
Chicago
Petersik, Paul. 2019. “Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting.” Utrecht University Repository (Utrecht University). http://dspace.library.uu.nl:8080/handle/1874/383388.
Harvard
Petersik, P. (2019) ‘Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting’, Utrecht University Repository (Utrecht University). Available at: http://dspace.library.uu.nl:8080/handle/1874/383388.
IEEE
P. Petersik, “Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting,” Utrecht University Repository (Utrecht University), 2019.
BibTeX
@phdthesis{petersik2019deep,
title = {{Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting}},
author = {Petersik, Paul},
year = {2019},
url = {http://dspace.library.uu.nl:8080/handle/1874/383388}
}
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 forecasting
- ensemble deep learning
- encoder-decoder neural network
- Oceanic Niño Index
- forecast uncertainty
- Indian Ocean Dipole
- sea surface temperature anomalies
- decadal variability
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?
Petersik, P. (2019). Deep Ensemble and Encoder-Decoder neural networks for ENSO forecasting. Utrecht University Repository (Utrecht University). http://dspace.library.uu.nl:8080/handle/1874/383388
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 Utrecht University Repository (Utrecht University): sign in and open the full text from this page.
Who wrote this paper?
Paul Petersik.
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.