Wagener et al. (2026) Metrics that matter: objective functions and their impact on signature representation in conceptual hydrological models

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-08-04
• Authors: Peter Wagener, Wouter Johannes Maria Knoben, Niels Schütze, Diana Spieler
• DOI: 10.5194/hess-30-4867-2026

Research Groups

• Institute of Hydrology and Meteorology, TUD Dresden University of Technology, Dresden, Germany
• Schulich School of Engineering, Department of Civil Engineering, University of Calgary, Calgary, Canada

Short Summary

This study systematically investigates the impact of objective function choice on the representation of various streamflow characteristics across 47 conceptual models and 10 hydro-climatically diverse catchments. The results show that objective function selection is often more influential than model structure choice in determining signature representation.

Objective

• Investigate the influence of objective function choice on the representation of hydrological signatures in conceptual hydrological models

Study Configuration

• Spatial Scale: Catchment scale (10 selected catchments from the CARAVAN dataset)
• Temporal Scale: Daily time step, with a 10-year calibration period and a 10-year evaluation period

Methodology and Data

• Models used: 47 conceptual hydrological models from the MARRMoT toolbox
• Data sources: Streamflow data from the CARAVAN dataset, meteorological forcing data from ERA5 (with an alternative Penman-Monteith-based PET estimation)

Main Results

• The results show that objective function selection is often more influential than model structure choice in determining signature representation.
• Some signatures, particularly those related to flow variability, are relatively insensitive to objective function choice.
• Other signatures exhibit large performance shifts across different objective functions.

Contributions

• This study provides generalisable guidance on which objective functions best reproduce which aspects of streamflow behaviour.
• The results help to better select objective functions most appropriate for a specific modelling purpose and provide insights into which metric combinations can cover a broader range of flow characteristics.

Funding

• Not specified in the provided text.

Citation

@article{Wagener2026Metrics,
  author = {Wagener, Peter and Knoben, Wouter Johannes Maria and Schütze, Niels and Spieler, Diana},
  title = {Metrics that matter: objective functions and their impact on signature representation in conceptual hydrological models},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-4867-2026},
  url = {https://doi.org/10.5194/hess-30-4867-2026}
}