Ersoy et al. (2025) Exploring the Potential of Multi-Hydrological Model Weighting Schemes to Reduce Uncertainty in Runoff Projections

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Water
• Year: 2025
• Date: 2025-10-10
• Authors: Zeynep Beril Ersoy, Okan Fıstıkoğlu, Umut Okkan
• DOI: 10.3390/w17202919

Research Groups

Not explicitly stated, but likely involves hydrological research groups in Türkiye focusing on climate change impacts and water resources management.

Short Summary

This study evaluates hydrological model weighting strategies to reduce runoff projection uncertainty under future climate scenarios, introducing the Uncertainty Optimizing Multi-Model Ensemble (UO-MME) framework which dynamically balances calibration performance and projection uncertainty, achieving an average 30% reduction in uncertainty compared to standard methods.

Objective

• To evaluate the capacity of strategies that weight monthly scale hydrological models (HMs) to narrow runoff projection uncertainty under future climate scenarios.
• To introduce and test the Uncertainty Optimizing Multi-Model Ensemble (UO-MME) framework, which dynamically considers trade-offs between calibration performance and projection uncertainty.

Study Configuration

• Spatial Scale: Beydag and Tahtali watersheds in Türkiye.
• Temporal Scale: Monthly scale for hydrological model simulations; future climate scenarios for projections.

Methodology and Data

• Models used: Seven unspecified Hydrological Models (HMs), five unspecified General Circulation Models (GCMs).
• Data sources: 140 ensemble runoff projections generated from a modeling chain comprising five GCMs, two emission scenarios, two downscaling methods, and seven HMs. Historical simulation skill data was used for standard weighting approaches.

Main Results

• Standard weighting techniques (Bayesian model averaging, ordered weighted averaging, Granger–Ramanathan averaging) resulted in either marginal reductions or noticeable increases in runoff projection uncertainty.
• The Uncertainty Optimizing Multi-Model Ensemble (UO-MME) framework achieved average reductions in runoff projection uncertainty of approximately 30% across the two watersheds.
• UO-MME maintained high simulation accuracy, with Nash–Sutcliffe efficiency values exceeding 0.75.
• UO-MME helps temper the inflation of noisy GCM signals in runoff responses, providing more balanced hydrological projections.

Contributions

• Introduction of the Uncertainty Optimizing Multi-Model Ensemble (UO-MME) framework, a novel dynamic weighting scheme for hydrological models that considers trade-offs between calibration performance and projection uncertainty.
• Demonstration of UO-MME's superior performance in reducing runoff projection uncertainty (approximately 30% reduction) compared to static, historical-skill-based weighting methods.
• Provides a method to balance the influences of climate signals from GCMs in hydrological projections, leading to more robust outputs for water resources planning.

Funding

Not explicitly stated in the provided text.

Citation

@article{Ersoy2025Exploring,
  author = {Ersoy, Zeynep Beril and Fıstıkoğlu, Okan and Okkan, Umut},
  title = {Exploring the Potential of Multi-Hydrological Model Weighting Schemes to Reduce Uncertainty in Runoff Projections},
  journal = {Water},
  year = {2025},
  doi = {10.3390/w17202919},
  url = {https://doi.org/10.3390/w17202919}
}