Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data

Krzysztof Chalupka Caltech, Tobias Bischoff Caltech, Frederick Eberhardt, Pietro Perona Caltech
Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, PMLR R14:38-47, 2016.

Abstract

We show that the climate phenomena of El Nino and La Nina arise naturally as states of macro variables when our recent causal feature learning framework (Chalupka2015a, Chalupka2015b) is applied to micro-level measures of zonal wind (ZW) and sea surface temperatures (SST) taken over the equatorial band of the Pacific Ocean. The method identifies these unusual climate states on the basis of the relation between ZW and SST patterns without any input about past occurrences of El Nino or La Nina. The simpler alternatives of (i) clustering the SST fields while disregarding their relationship with ZW patterns, or (ii) clustering the joint ZW-SST patterns, do not discover El Nino. We discuss the degree to which our method supports a causal interpretation and use a low-dimensional toy example to explain its success over other clustering approaches. Finally, we propose a new robust and scalable alternative to our original algorithm Chalupka2015b, which circumvents the need for high-dimensional density learning.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR14-caltech16a, title = {Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data}, author = {Caltech, Krzysztof Chalupka and Caltech, Tobias Bischoff and Eberhardt, Frederick and Caltech, Pietro Perona}, booktitle = {Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence}, pages = {38--47}, year = {2016}, editor = {Ihler, Alexander and Janzing, Dominik}, volume = {R14}, series = {Proceedings of Machine Learning Research}, month = {25--29 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r14/main/assets/caltech16a/caltech16a.pdf}, url = {https://proceedings.mlr.press/r14/caltech16a.html}, abstract = {We show that the climate phenomena of El Nino and La Nina arise naturally as states of macro variables when our recent causal feature learning framework (Chalupka2015a, Chalupka2015b) is applied to micro-level measures of zonal wind (ZW) and sea surface temperatures (SST) taken over the equatorial band of the Pacific Ocean. The method identifies these unusual climate states on the basis of the relation between ZW and SST patterns without any input about past occurrences of El Nino or La Nina. The simpler alternatives of (i) clustering the SST fields while disregarding their relationship with ZW patterns, or (ii) clustering the joint ZW-SST patterns, do not discover El Nino. We discuss the degree to which our method supports a causal interpretation and use a low-dimensional toy example to explain its success over other clustering approaches. Finally, we propose a new robust and scalable alternative to our original algorithm Chalupka2015b, which circumvents the need for high-dimensional density learning.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data %A Krzysztof Chalupka Caltech %A Tobias Bischoff Caltech %A Frederick Eberhardt %A Pietro Perona Caltech %B Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2016 %E Alexander Ihler %E Dominik Janzing %F pmlr-vR14-caltech16a %I PMLR %P 38--47 %U https://proceedings.mlr.press/r14/caltech16a.html %V R14 %X We show that the climate phenomena of El Nino and La Nina arise naturally as states of macro variables when our recent causal feature learning framework (Chalupka2015a, Chalupka2015b) is applied to micro-level measures of zonal wind (ZW) and sea surface temperatures (SST) taken over the equatorial band of the Pacific Ocean. The method identifies these unusual climate states on the basis of the relation between ZW and SST patterns without any input about past occurrences of El Nino or La Nina. The simpler alternatives of (i) clustering the SST fields while disregarding their relationship with ZW patterns, or (ii) clustering the joint ZW-SST patterns, do not discover El Nino. We discuss the degree to which our method supports a causal interpretation and use a low-dimensional toy example to explain its success over other clustering approaches. Finally, we propose a new robust and scalable alternative to our original algorithm Chalupka2015b, which circumvents the need for high-dimensional density learning. %Z Reissued by PMLR on 04 October 2026.
APA
Caltech, K.C., Caltech, T.B., Eberhardt, F. & Caltech, P.P.. (2016). Unsupervised Discovery of El Nino Using Causal Feature Learning on Microlevel Climate Data. Proceedings of the 32nd Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R14:38-47 Available from https://proceedings.mlr.press/r14/caltech16a.html. Reissued by PMLR on 04 October 2026.

Related Material