Preprints
https://doi.org/10.5194/egusphere-2026-5583
https://doi.org/10.5194/egusphere-2026-5583
18 Nov 2026
 | 18 Nov 2026
Status: this preprint is open for discussion and under review for Hydrology and Earth System Sciences (HESS).

Learning from dense meteorological networks: a regional LSTM model for streamflow simulation

Inmaculada C. González-Planet, Daniel Klotz, and Carmelo Juez

Abstract. Long Short-Term Memory (LSTM) networks have become the current state-of-the-art for streamflow modelling due to their ability to capture non-linear relationships and long-term temporal dependencies. In regional LSTM applications, meteorological forcings are typically aggregated at the catchment scale, which removes spatial variability and may limit the representation of key hydrological processes, particularly in large and heterogeneous basins. This study departs from this conventional practice by proposing a regional LSTM model that directly exploits the full spatial structure of raw meteorological observations without any prior spatial aggregation. The model integrates dynamic and static information from more than 700 meteorological stations across the northern Ebro River Basin (Spain) to simulate streamflow at 31 catchments. Missing data are handled using a masked mean approach, enabling direct use of observations. Compared to a benchmark regional LSTM based on catchment-averaged forcings, the proposed approach improves overall performance, and yields substantial reductions in error for medium- and high-flow conditions. These results indicate that preserving spatial meteorological variability enhances the representation of hydrological processes associated with runoff generation and peak-flow response. To further investigate the learned relationships, explainable artificial intelligence (XAI) techniques are applied. The analyses reveal physically consistent spatial and temporal attribution patterns, showing that the model dynamically adjusts the relevance of meteorological stations depending on hydrological conditions. This study demonstrates that spatial aggregation of meteorological inputs can limit their ability to capture hydrological dynamics in complex basins. The proposed framework provides a flexible and data-driven approach for exploiting dense observational networks in regional streamflow modelling.

Competing interests: At least one of the (co-)authors is a member of the editorial board of Hydrology and Earth System Sciences.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Inmaculada C. González-Planet, Daniel Klotz, and Carmelo Juez

Status: open (until 31 Dec 2026)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
Inmaculada C. González-Planet, Daniel Klotz, and Carmelo Juez
Inmaculada C. González-Planet, Daniel Klotz, and Carmelo Juez

Viewed

Total article views: 86 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
69 14 3 86 7 6
  • HTML: 69
  • PDF: 14
  • XML: 3
  • Total: 86
  • BibTeX: 7
  • EndNote: 6
Views and downloads (calculated since 18 Nov 2026)
Cumulative views and downloads (calculated since 18 Nov 2026)

Viewed (geographical distribution)

Total article views: 86 (including HTML, PDF, and XML) Thereof 86 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 19 Nov 2026
Download
Short summary
Accurate streamflow simulations are essential for water management. This study applies an artificial intelligence model and shows that using information from hundreds of weather stations, rather than averaging observations over entire catchments, improves predictions across a large river basin in Spain. The model automatically identifies the most relevant weather information for each location and reveals physically meaningful relationships between weather conditions and river flow.
Share