Multi‐Model Ensemble and Reservoir Computing for River Discharge Prediction in Ungauged Basins
Journal of Geophysical Research: Machine Learning and Computation · 2026-08 · 3 (4)
DOI: 10.1029/2025jh001196 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Geophysical Research: Machine Learning and Computation, online erschienen am 05.08.2026. Verfasst von Mizuki Funato und 1 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 73 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Japan Society for the Promotion of Science.
Abstract (Originalsprache)
Despite the necessity for accurate flood prediction, many regions lack sufficient river discharge observations. Although numerous models for daily river discharge prediction exist, achieving high accuracy, interpretability, and efficiency under data‐scarce conditions remains a major challenge. We address this with a novel method, HYdrological Prediction with multi‐model Ensemble and Reservoir computing (HYPER). Our approach applies Bayesian Model Averaging (BMA) to 47 “uncalibrated” lumped conceptual hydrological models. A reservoir computing (RC) model, a type of machine learning model, is then trained via linear regression to correct BMA output errors, a non‐iterative process ensuring computational efficiency. For ungauged basins, we infer the required BMA and RC weights by mapping them to catchment attributes from gauged basins, creating a generalizable framework. Evaluated on 87 Japanese basins, in a data‐rich scenario, HYPER (median Kling‐Gupta Efficiency, KGE, of 0.65) performed comparably to a benchmark Long Short‐Term Memory (Kling‐Gupta Efficiency (KGE) 0.58) but required only 3% of its total computational time. In a data‐scarce scenario (where only ∼20% of basins are gauged), HYPER maintained robust performance (KGE 0.60) by leveraging the physical structure of the ensemble. In contrast, the LSTM's performance degraded substantially (KGE ‐0.07) due to data insufficiency. These results demonstrate that calibrating individual conceptual hydrological models can be circumvented when using a sufficiently large ensemble assembled and combined with machine‐learning‐based bias correction. HYPER provides a robust, efficient, and generalizable solution for discharge prediction, particularly in ungauged basins. By eliminating basin‐specific calibration, HYPER offers a scalable, interpretable framework for accurate hydrological prediction in diverse data‐scarce regions.
Förderung
- Japan Society for the Promotion of Science
Zitieren
APA
Funato, M., & Sawada, Y. (2026). Multi‐Model Ensemble and Reservoir Computing for River Discharge Prediction in Ungauged Basins. Journal of Geophysical Research: Machine Learning and Computation, 3(4). https://doi.org/10.1029/2025jh001196
Vancouver
Funato M, Sawada Y. Multi‐Model Ensemble and Reservoir Computing for River Discharge Prediction in Ungauged Basins. Journal of Geophysical Research: Machine Learning and Computation. 2026;3(4). doi:10.1029/2025jh001196
BibTeX
@article{Funato2026,
author = {Funato, Mizuki and Sawada, Yohei},
title = {Multi‐Model Ensemble and Reservoir Computing for River Discharge Prediction in Ungauged Basins},
journal = {Journal of Geophysical Research: Machine Learning and Computation},
year = {2026},
volume = {3},
number = {4},
doi = {10.1029/2025jh001196}
}
EUVIMED