Brigode et al. (2025) Using century-long reanalysis and a rainfall-runoff model to explore multi-decadal variability in catchment hydrology at the European scale

Identification

• Journal: Hydrology and earth system sciences
• Year: 2025
• Date: 2025-10-22
• Authors: Pierre Brigode, Ludovic Oudin
• DOI: 10.5194/hess-29-5535-2025

Research Groups

• Univ Rennes, CNRS, Géosciences Rennes – UMR 6118, Rennes 35000, France
• Sorbonne Université, Université PSL, EPHE, CNRS, Milieux Environnementaux, Transferts et Interactions dans les hydrosystèmes et les Sols, METIS, Case 105, 4 place Jussieu, 75005 Paris, France

Short Summary

This study evaluates the capacity of century-long global reanalyses (NOAA 20CR, ERA-20C) to simulate multi-decadal catchment hydrology across over 2000 European catchments using a rainfall-runoff model, finding reasonable performance, especially for mean flows, and revealing significant alternating wet and dry periods.

Objective

• To evaluate the ability of a modeling methodology, using global reanalyses as inputs for a rainfall-runoff model, to identify trends and/or periodicities of catchment hydrology at the European scale, despite coarse spatial resolution and modeling uncertainty.
• To assess how well global reanalyses perform in providing climate forcings for hydrological models to reproduce observed streamflow at daily and decadal timescales.
• To determine if simulation performance depends on spatial scale (catchment size) and hydrological processes (catchment regimes).
• To verify if the simulated low-frequency variability aligns with observations and other simulation results.

Study Configuration

• Spatial Scale: European scale, covering 2128 catchments with areas larger than 100 km². Climate forcings used have resolutions ranging from 0.1° (approximately 91 km²) to 1.4° (approximately 17000 km²).
• Temporal Scale: Century-long simulations, with reanalyses spanning from 1836 (NOAA 20CR) and 1900 (ERA-20C). Calibration period: 1996–2010. Evaluation periods: 1982–1995 and 1903–1995.

Methodology and Data

• Models used: GR4J (conceptual rainfall–runoff model) and CemaNeige (snow accumulation and melting module).
• Data sources:
  • Climate Forcings: MSWEP (V2) daily precipitation (1979–2019, 0.1°), ERA5 reanalysis daily air temperature (1980–2019, 0.25°) for reference calibration. ERA-20C reanalysis (1900–2010, 1.4°) and NOAA 20CR (v3) reanalysis (1836–2015, 1.0°) for long-term simulations.
  • Catchment Data: Daily streamflow series from 2128 European catchments, compiled from CAMELS-CH, CAMELS-FR, CAMELS-GB, NVE, SMHI, SAIH-RODEA, LamaH-CE, and GRDC datasets.
  • Ancillary Data: GRanD dataset (for dam storage capacity), EU-DEM (v1.1) for catchment delineations and hypsometric data.

Main Results

• Both NOAA 20CR and ERA-20C reanalyses, when coupled with the GR4J model, can reasonably reproduce daily and multi-decadal streamflow variability across Europe, particularly for mean flows.
• Simulation performance generally improves with increasing catchment size, but challenges persist for Mediterranean and snow-dominated regions.
• Consistency between the climate forcings used for model calibration and simulation is crucial; calibrating with the same reanalysis (e.g., NOAA 20CR) used for simulation yields better performance than calibrating with higher-resolution reference datasets.
• The study reveals significant multi-decadal variations in streamflow, characterized by alternating wet and dry periods across Europe, rather than consistent linear trends.
• Performance varies regionally, with the best results observed in catchments in western France, Scandinavia, and the British Isles, and the lowest in eastern Europe, the Mediterranean, and the Iberian Peninsula.
• NOAA 20CR generally shows slightly better performance than ERA-20C, with notable differences in some regions (e.g., temporal correlation for mean flows in eastern Europe is approximately 0.8 for NOAA 20CR versus 0.4 for ERA-20C).

Contributions

• Provides the first large-scale (European) assessment of the suitability of century-long global reanalyses (NOAA 20CR, ERA-20C) as inputs for conceptual rainfall-runoff models to explore multi-decadal hydrological variability.
• Highlights the critical importance of ensuring consistency between calibration and simulation climate forcings, suggesting it can be more impactful than the spatial resolution of the calibration data for long-term hydrological modeling.
• Offers a valuable framework for understanding long-term hydrological trends and anticipating future changes in water resources and hydrological extremes under climate variability.
• Emphasizes the prevalence of multi-decadal wet and dry cycles over linear trends in European streamflow, underscoring the utility of long-term reanalyses for such analyses.

Funding

• This work was supported by the French National program EC2CO (Ecosphère Continentale et Côtière).

Citation

@article{Brigode2025Using,
  author = {Brigode, Pierre and Oudin, Ludovic},
  title = {Using century-long reanalysis and a rainfall-runoff model to explore multi-decadal variability in catchment hydrology at the European scale},
  journal = {Hydrology and earth system sciences},
  year = {2025},
  doi = {10.5194/hess-29-5535-2025},
  url = {https://doi.org/10.5194/hess-29-5535-2025}
}