Toward routing river water in land surface models with recurrent neural networks

Lima, Mauricio; Deck, Katherine; Dunbar, Oliver R. A.; Schneider, Tapio

doi:https://doi.org/10.5194/hess-29-3145-2025

Articles | Volume 29, issue 14

https://doi.org/10.5194/hess-29-3145-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/hess-29-3145-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 29, issue 14

Research article

|

24 Jul 2025

Research article |

| 24 Jul 2025

Toward routing river water in land surface models with recurrent neural networks

Mauricio Lima, Katherine Deck, Oliver R. A. Dunbar, and Tapio Schneider

Data sets

ERA5-Land hourly data from 1950 to present J. Muñoz Sabater https://doi.org/10.24381/cds.e2161bac

The GRDC GRDC Data Download https://portal.grdc.bafg.de/

River discharge and related historical data from the Global Flood Awareness System Joint Research Center, Copernicus Emergency Management Service https://doi.org/10.24381/cds.a4fdd6b9

Global river hydrography and network routing: baseline data and new approaches to study the world's large river systems (https://www.hydrosheds.org/hydroatlas#download) B. Lehner and G. Grill https://doi.org/10.1002/hyp.9740

Global hydro-environmental sub-basin and river reach characteristics at high spatial resolution (https://www.hydrosheds.org/products/hydrobasins#downloads) S. Linke et al. https://doi.org/10.1038/s41597-019-0300-6

Model code and software

limamau/Rivers: Release v0.1.1 (v0.1.1) Mauricio Lima https://doi.org/10.5281/zenodo.13752864

Short summary

Machine learning is playing an increasingly important role in hydrological modeling. In this paper, we introduce an adaptation of existing machine learning models for simulating streamflow in river basins, redesigning them with the goal of integrating them in climate models. We demonstrate the effectiveness of our adapted model by showing that it outperforms a physics-based river model. These results motivate further studies of the use of machine-learning-based river models inside climate models.