Articles | Volume 26, issue 19
https://doi.org/10.5194/hess-26-5085-2022
https://doi.org/10.5194/hess-26-5085-2022
Research article
 | Highlight paper
 | 
11 Oct 2022
Research article | Highlight paper |  | 11 Oct 2022

Improving hydrologic models for predictions and process understanding using neural ODEs

Marvin Höge, Andreas Scheidegger, Marco Baity-Jesi, Carlo Albert, and Fabrizio Fenicia

Viewed

Total article views: 6,039 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
4,373 1,595 71 6,039 75 64
  • HTML: 4,373
  • PDF: 1,595
  • XML: 71
  • Total: 6,039
  • BibTeX: 75
  • EndNote: 64
Views and downloads (calculated since 10 Mar 2022)
Cumulative views and downloads (calculated since 10 Mar 2022)

Viewed (geographical distribution)

Total article views: 6,039 (including HTML, PDF, and XML) Thereof 5,697 with geography defined and 342 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 17 Jul 2024
Download
Executive editor
This is a paper demonstrating the added value of hybrid modeling approaches for modeling hydrological features (flow and internal states). The paper is well written and will be sent out for a detailed scientific review. The combination of ANN with ODE-based hydrological models is a good way forward to construct hybrid modeling approaches. It is a significant step forward compared to the current state of the art.
Short summary
Neural ODEs fuse physics-based models with deep learning: neural networks substitute terms in differential equations that represent the mechanistic structure of the system. The approach combines the flexibility of machine learning with physical constraints for inter- and extrapolation. We demonstrate that neural ODE models achieve state-of-the-art predictive performance while keeping full interpretability of model states and processes in hydrologic modelling over multiple catchments.