Barreiro-Fonta et al. (2025) Assessing Climate Change and Reservoir Impacts on Upper Miño River Flow (NW Iberian Peninsula) Using Neural Networks

Identification

• Journal: Water
• Year: 2025
• Date: 2025-12-12
• Authors: Helena Barreiro-Fonta, Diego Fernández-Nóvoa
• DOI: 10.3390/w17243514

Research Groups

• Environmental Physics Laboratory (EPhysLab), Centro de Investigación Mariña (CIM), Universidade de Vigo, Spain.

Short Summary

This study utilizes artificial neural networks to project the impacts of climate change on the Upper Miño River, finding that while high-emission scenarios intensify hydrological extremes, reservoir operations can significantly mitigate these effects by redistributing seasonal water availability.

Objective

• To evaluate projected changes in streamflow and reservoir regulation in the Upper Miño basin by the end of the 21st century under different Shared Socioeconomic Pathways (SSP2-4.5 and SSP5-8.5).

Study Configuration

• Spatial Scale: Upper Miño River basin (catchment area: 4344 $km^2$) and the Belesar reservoir (storage capacity: 654 $hm^3$), Northwest Iberian Peninsula.
• Temporal Scale: Historical baseline (1985–2014) compared against late-century projections (2070–2099).

Methodology and Data

• Models used: Nonlinear Autoregressive networks with Exogenous inputs (NARX) neural networks for both hydrological cycle modeling and reservoir operation simulation; WRF-ARW v4.3.3 regional climate model for downscaling.
• Data sources: Iberia01 gridded dataset (0.1° resolution) for historical precipitation and temperature; Confederación Hidrográfica del Miño-Sil (CHMS) for daily reservoir inflow, outflow, and storage; CMIP6 Global Climate Models (MPI-ESM1-HR and an 18-model ensemble).

Main Results

• SSP5-8.5 Scenario: Projected mean streamflow decline of >15%. Low-flow conditions (10th percentile) are expected to decrease by ~11%, while high-flow conditions (99.997th percentile/one-year return period) are projected to increase by ~5%.
• SSP2-4.5 Scenario: Projected mean streamflow decline of <15%. Low-flow conditions decrease by ~5%, while high-flow conditions show a significant projected decrease of ~30% relative to the historical period.
• Reservoir Impact (SSP5-8.5): The Belesar reservoir moderates extremes by redistributing water from winter to summer; downstream summer flows are projected to be more than double the natural upstream flows, and the one-year return period high-flow value is reduced by ~15%.
• Ecological Flow: Under SSP5-8.5, the number of days failing to meet ecological flow requirements is projected to rise sharply in spring, from <5 days to ~25 days per season.

Contributions

• Provides a robust, data-driven framework using neural networks to model complex, non-linear reservoir management decisions without requiring explicit operational rules.
• Quantifies the specific "buffer effect" of large-scale infrastructure in mitigating climate-induced hydrological intensification in the Iberian Peninsula.
• Demonstrates the transferability of ANN-based hydrological modeling using standard atmospheric variables (temperature and precipitation) for regional climate impact assessments.

Funding

• Xunta de Galicia (Project ED431C 2021/44 - Grupos de Referencia Competitiva).
• Interreg POCTEP program (Project RISCPLUS, 0031RISCPLUS6_E).
• Xunta de Galicia postdoctoral grant (ED481D-2024-004).

Citation

@article{BarreiroFonta2025Assessing,
  author = {Barreiro-Fonta, Helena and Fernández-Nóvoa, Diego},
  title = {Assessing Climate Change and Reservoir Impacts on Upper Miño River Flow (NW Iberian Peninsula) Using Neural Networks},
  journal = {Water},
  year = {2025},
  doi = {10.3390/w17243514},
  url = {https://doi.org/10.3390/w17243514}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)