Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models

Abstract The Madden–Julian Oscillation (MJO) is an influential climate phenomenon that plays a vital role in modulating global weather patterns. In spite of the improvement in MJO predictions made by machine learning algorithms, such as neural networks, most of them cannot provide the uncertainty le...

Full description

Saved in:
Bibliographic Details
Main Authors: Haoyuan Chen, Emil Constantinescu, Vishwas Rao, Cristiana Stan
Format: Article
Language:English
Published: American Geophysical Union (AGU) 2025-05-01
Series:Journal of Advances in Modeling Earth Systems
Subjects:
Online Access:https://doi.org/10.1029/2023MS004188
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327236763615232
author Haoyuan Chen
Emil Constantinescu
Vishwas Rao
Cristiana Stan
author_facet Haoyuan Chen
Emil Constantinescu
Vishwas Rao
Cristiana Stan
author_sort Haoyuan Chen
collection DOAJ
description Abstract The Madden–Julian Oscillation (MJO) is an influential climate phenomenon that plays a vital role in modulating global weather patterns. In spite of the improvement in MJO predictions made by machine learning algorithms, such as neural networks, most of them cannot provide the uncertainty levels in the MJO forecasts directly. To address this problem, we develop a nonparametric strategy based on Gaussian process (GP) models. We calibrate GPs using empirical correlations and we propose a posteriori covariance correction. Numerical experiments demonstrate that our model has better prediction skills than the artificial neural network models for the first five lead days. Additionally, our posteriori covariance correction extends the probabilistic coverage by more than 3 weeks.
format Article
id doaj-art-44df9ddbd407418da50ed7226e5c7b57
institution Kabale University
issn 1942-2466
language English
publishDate 2025-05-01
publisher American Geophysical Union (AGU)
record_format Article
series Journal of Advances in Modeling Earth Systems
spelling doaj-art-44df9ddbd407418da50ed7226e5c7b572025-08-20T03:47:57ZengAmerican Geophysical Union (AGU)Journal of Advances in Modeling Earth Systems1942-24662025-05-01175n/an/a10.1029/2023MS004188Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process ModelsHaoyuan Chen0Emil Constantinescu1Vishwas Rao2Cristiana Stan3Department of Industrial and Systems Engineering Texas A&M University College Station TX USAMathematics and Computer Science Division, Argonne National Laboratory Lemont IL USAMathematics and Computer Science Division, Argonne National Laboratory Lemont IL USADepartment of Atmospheric Oceanic and Earth Sciences George Mason University Fairfax VA USAAbstract The Madden–Julian Oscillation (MJO) is an influential climate phenomenon that plays a vital role in modulating global weather patterns. In spite of the improvement in MJO predictions made by machine learning algorithms, such as neural networks, most of them cannot provide the uncertainty levels in the MJO forecasts directly. To address this problem, we develop a nonparametric strategy based on Gaussian process (GP) models. We calibrate GPs using empirical correlations and we propose a posteriori covariance correction. Numerical experiments demonstrate that our model has better prediction skills than the artificial neural network models for the first five lead days. Additionally, our posteriori covariance correction extends the probabilistic coverage by more than 3 weeks.https://doi.org/10.1029/2023MS004188Madden–Julian oscillationGaussian process modelsprobabilistic forecastinguncertainty quantificationsubseasonal‐to‐seasonal prediction
spellingShingle Haoyuan Chen
Emil Constantinescu
Vishwas Rao
Cristiana Stan
Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
Journal of Advances in Modeling Earth Systems
Madden–Julian oscillation
Gaussian process models
probabilistic forecasting
uncertainty quantification
subseasonal‐to‐seasonal prediction
title Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
title_full Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
title_fullStr Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
title_full_unstemmed Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
title_short Improving the Predictability of the Madden‐Julian Oscillation at Subseasonal Scales With Gaussian Process Models
title_sort improving the predictability of the madden julian oscillation at subseasonal scales with gaussian process models
topic Madden–Julian oscillation
Gaussian process models
probabilistic forecasting
uncertainty quantification
subseasonal‐to‐seasonal prediction
url https://doi.org/10.1029/2023MS004188
work_keys_str_mv AT haoyuanchen improvingthepredictabilityofthemaddenjulianoscillationatsubseasonalscaleswithgaussianprocessmodels
AT emilconstantinescu improvingthepredictabilityofthemaddenjulianoscillationatsubseasonalscaleswithgaussianprocessmodels
AT vishwasrao improvingthepredictabilityofthemaddenjulianoscillationatsubseasonalscaleswithgaussianprocessmodels
AT cristianastan improvingthepredictabilityofthemaddenjulianoscillationatsubseasonalscaleswithgaussianprocessmodels