Using Simple, Explainable Neural Networks to Predict the Madden‐Julian Oscillation

Martin, Zane K.; Barnes, Elizabeth A.; Maloney, Eric

doi:10.1029/2021ms002774

i

Using Simple, Explainable Neural Networks to Predict the Madden‐Julian Oscillation

Supporting Files

2022
By Martin, Zane K. ; Barnes, Elizabeth A. ; Maloney, Eric

Details

Journal Title:

Journal of Advances in Modeling Earth Systems
Personal Author:

Martin, Zane K. ; Barnes, Elizabeth A. ; Maloney, Eric
NOAA Program & Office:

OAR (Oceanic and Atmospheric Research)
Description:

Few studies have utilized machine learning techniques to predict or understand the Madden‐Julian oscillation (MJO), a key source of subseasonal variability and predictability. Here, we present a simple framework for real‐time MJO prediction using shallow artificial neural networks (ANNs). We construct two ANN architectures, one deterministic and one probabilistic, that predict a real‐time MJO index using maps of tropical variables. These ANNs make skillful MJO predictions out to ∼18 days in October‐March and ∼11 days in April‐September, outperforming conventional linear models and efficiently capturing aspects of MJO predictability found in more complex, dynamical models. The flexibility and explainability of simple ANN frameworks are highlighted through varying model input and applying ANN explainability techniques that reveal sources and regions important for ANN prediction skill. The accessibility, performance, and efficiency of this simple machine learning framework is more broadly applicable to predict and understand other Earth system phenomena.
Keywords:

Environmental Chemistry General Earth And Planetary Sciences Global And Planetary Change General Earth And Planetary Sciences Global And Planetary Change General Earth And Planetary Sciences Global And Planetary Change
Source:

Journal of Advances in Modeling Earth Systems, 14(5)
DOI:

https://doi.org/10.1029/2021ms002774
ISSN:

1942-2466 ; 1942-2466
Format:

PDF
Publisher:

American Geophysical Union (AGU)
Document Type:

Journal Article
Funding:

Grant no. NA18OAR4310296 ; Grant no. NA18OAR4310299 ; Grant no. NA19OAR4590151
License:

https://creativecommons.org/licenses/by/4.0/
Rights Information:

CC BY
Compliance:

Library
Main Document Checksum:

urn:sha256:d44ecfc1ad03c0712ab2828902b22631ebe4935100ac0d0994b20ae40be2a398
Download URL:

https://repository.library.noaa.gov/view/noaa/58250/noaa_58250_DS1.pdf
File Type:

[PDF - 2.41 MB ]

Machine readable version (XML)

Download

ON THIS PAGE

Details Supporting Files

The NOAA IR serves as an archival repository of NOAA-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by NOAA or funded partners. As a repository, the NOAA IR retains documents in their original published format to ensure public access to scientific information.