Earth System Model Tuning Without Hyperparameters

Abstract This article introduces a new algorithm, KalmRidge, and demonstrates its ability to tune an Earth system model using idealized experiments. Unlike similar algorithms, KalmRidge eliminates the need for offline hyperparameter selection, thereby substantially reducing computational expense. Th...

Full description

Saved in:
Bibliographic Details
Main Authors: Nikki Lydeen, Timothy DelSole, Benjamin Cash
Format: Article
Language:English
Published: American Geophysical Union (AGU) 2025-07-01
Series:Journal of Advances in Modeling Earth Systems
Subjects:
Online Access:https://doi.org/10.1029/2024MS004607
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849729093142052864
author Nikki Lydeen
Timothy DelSole
Benjamin Cash
author_facet Nikki Lydeen
Timothy DelSole
Benjamin Cash
author_sort Nikki Lydeen
collection DOAJ
description Abstract This article introduces a new algorithm, KalmRidge, and demonstrates its ability to tune an Earth system model using idealized experiments. Unlike similar algorithms, KalmRidge eliminates the need for offline hyperparameter selection, thereby substantially reducing computational expense. This is done by rewriting the update equations for the ensemble Kalman filter as an equivalent ridge regression problem, then applying standard cross‐validation techniques to adaptively choose the regularization parameter. We propose that this algorithm, with time‐mean spherical harmonic projections as tuning targets, provides a promising, tractable approach for parameter estimation.
format Article
id doaj-art-594555f7f238428aa52e5ecfb5ab1fb9
institution DOAJ
issn 1942-2466
language English
publishDate 2025-07-01
publisher American Geophysical Union (AGU)
record_format Article
series Journal of Advances in Modeling Earth Systems
spelling doaj-art-594555f7f238428aa52e5ecfb5ab1fb92025-08-20T03:09:19ZengAmerican Geophysical Union (AGU)Journal of Advances in Modeling Earth Systems1942-24662025-07-01177n/an/a10.1029/2024MS004607Earth System Model Tuning Without HyperparametersNikki Lydeen0Timothy DelSole1Benjamin Cash2George Mason University Fairfax VA USAGeorge Mason University Fairfax VA USAGeorge Mason University Fairfax VA USAAbstract This article introduces a new algorithm, KalmRidge, and demonstrates its ability to tune an Earth system model using idealized experiments. Unlike similar algorithms, KalmRidge eliminates the need for offline hyperparameter selection, thereby substantially reducing computational expense. This is done by rewriting the update equations for the ensemble Kalman filter as an equivalent ridge regression problem, then applying standard cross‐validation techniques to adaptively choose the regularization parameter. We propose that this algorithm, with time‐mean spherical harmonic projections as tuning targets, provides a promising, tractable approach for parameter estimation.https://doi.org/10.1029/2024MS004607parameter estimationKalman filterhyperparameters
spellingShingle Nikki Lydeen
Timothy DelSole
Benjamin Cash
Earth System Model Tuning Without Hyperparameters
Journal of Advances in Modeling Earth Systems
parameter estimation
Kalman filter
hyperparameters
title Earth System Model Tuning Without Hyperparameters
title_full Earth System Model Tuning Without Hyperparameters
title_fullStr Earth System Model Tuning Without Hyperparameters
title_full_unstemmed Earth System Model Tuning Without Hyperparameters
title_short Earth System Model Tuning Without Hyperparameters
title_sort earth system model tuning without hyperparameters
topic parameter estimation
Kalman filter
hyperparameters
url https://doi.org/10.1029/2024MS004607
work_keys_str_mv AT nikkilydeen earthsystemmodeltuningwithouthyperparameters
AT timothydelsole earthsystemmodeltuningwithouthyperparameters
AT benjamincash earthsystemmodeltuningwithouthyperparameters