Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events

<p>Data-driven techniques have shown the potential to outperform process-based models in rainfall–runoff simulation. Recently, hybrid models, which combine data-driven methods with process-based approaches, have been proposed to leverage the strengths of both methodologies, aiming to enhance s...

Full description

Saved in:
Bibliographic Details
Main Authors: E. Acuña Espinoza, R. Loritz, F. Kratzert, D. Klotz, M. Gauch, M. Álvarez Chaves, U. Ehret
Format: Article
Language:English
Published: Copernicus Publications 2025-03-01
Series:Hydrology and Earth System Sciences
Online Access:https://hess.copernicus.org/articles/29/1277/2025/hess-29-1277-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850250929100554240
author E. Acuña Espinoza
R. Loritz
F. Kratzert
D. Klotz
D. Klotz
M. Gauch
M. Álvarez Chaves
U. Ehret
author_facet E. Acuña Espinoza
R. Loritz
F. Kratzert
D. Klotz
D. Klotz
M. Gauch
M. Álvarez Chaves
U. Ehret
author_sort E. Acuña Espinoza
collection DOAJ
description <p>Data-driven techniques have shown the potential to outperform process-based models in rainfall–runoff simulation. Recently, hybrid models, which combine data-driven methods with process-based approaches, have been proposed to leverage the strengths of both methodologies, aiming to enhance simulation accuracy while maintaining a certain interpretability. Expanding the set of test cases to evaluate hybrid models under different conditions, we test their generalization capabilities for extreme hydrological events, comparing their performance against long short-term memory (LSTM) networks and process-based models. Our results indicate that hybrid models show performance similar to that of the LSTM network for most cases. However, hybrid models reported slightly lower errors in the most extreme cases and were able to produce higher peak discharges.</p>
format Article
id doaj-art-098b9a0b11ee4635bec033f80d007324
institution OA Journals
issn 1027-5606
1607-7938
language English
publishDate 2025-03-01
publisher Copernicus Publications
record_format Article
series Hydrology and Earth System Sciences
spelling doaj-art-098b9a0b11ee4635bec033f80d0073242025-08-20T01:58:03ZengCopernicus PublicationsHydrology and Earth System Sciences1027-56061607-79382025-03-01291277129410.5194/hess-29-1277-2025Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme eventsE. Acuña Espinoza0R. Loritz1F. Kratzert2D. Klotz3D. Klotz4M. Gauch5M. Álvarez Chaves6U. Ehret7Institute of Water and Environment, Karlsruhe Institute of Technology (KIT), Karlsruhe, GermanyInstitute of Water and Environment, Karlsruhe Institute of Technology (KIT), Karlsruhe, GermanyGoogle Research, Vienna, AustriaGoogle Research, Vienna, AustriaHelmholtz Centre for Environmental Research (UFZ), Leipzig, GermanyGoogle Research, Zurich, SwitzerlandStuttgart Center for Simulation Science, Statistical Model-Data Integration, University of Stuttgart, Stuttgart, GermanyInstitute of Water and Environment, Karlsruhe Institute of Technology (KIT), Karlsruhe, Germany<p>Data-driven techniques have shown the potential to outperform process-based models in rainfall–runoff simulation. Recently, hybrid models, which combine data-driven methods with process-based approaches, have been proposed to leverage the strengths of both methodologies, aiming to enhance simulation accuracy while maintaining a certain interpretability. Expanding the set of test cases to evaluate hybrid models under different conditions, we test their generalization capabilities for extreme hydrological events, comparing their performance against long short-term memory (LSTM) networks and process-based models. Our results indicate that hybrid models show performance similar to that of the LSTM network for most cases. However, hybrid models reported slightly lower errors in the most extreme cases and were able to produce higher peak discharges.</p>https://hess.copernicus.org/articles/29/1277/2025/hess-29-1277-2025.pdf
spellingShingle E. Acuña Espinoza
R. Loritz
F. Kratzert
D. Klotz
D. Klotz
M. Gauch
M. Álvarez Chaves
U. Ehret
Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
Hydrology and Earth System Sciences
title Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
title_full Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
title_fullStr Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
title_full_unstemmed Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
title_short Analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
title_sort analyzing the generalization capabilities of a hybrid hydrological model for extrapolation to extreme events
url https://hess.copernicus.org/articles/29/1277/2025/hess-29-1277-2025.pdf
work_keys_str_mv AT eacunaespinoza analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT rloritz analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT fkratzert analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT dklotz analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT dklotz analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT mgauch analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT malvarezchaves analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents
AT uehret analyzingthegeneralizationcapabilitiesofahybridhydrologicalmodelforextrapolationtoextremeevents