Ara et al. (2026) Assessing Climate Change Impacts on Future Precipitation Using Random Forest Statistical Downscaling of CMIP6 HadGEM3 Projections in the Büyük Menderes Basin

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

Identification

• Journal: Water
• Year: 2026
• Date: 2026-01-21
• Authors: Ismail Ara, Mutlu Yaşar, Gürhan Gürarslan
• DOI: 10.3390/w18020277

Research Groups

Not specified in the provided text.

Short Summary

This study developed and evaluated a hybrid machine-learning statistical downscaling framework to generate monthly precipitation projections for the 21st century in the Büyük Menderes Basin, western Türkiye, revealing a significant basin-wide drying trend under both moderate and high-emission climate change scenarios.

Objective

• To develop and evaluate a hybrid, machine-learning-based statistical downscaling framework to generate monthly precipitation projections for the 21st century in the Büyük Menderes Basin, western Türkiye.

Study Configuration

• Spatial Scale: Regional (Büyük Menderes Basin, western Türkiye) with station-scale (23 rainfall stations) projections.
• Temporal Scale: Monthly projections for the 21st century (2025–2099), analyzed in near (2025–2050), mid (2051–2075), and late (2076–2099) periods.

Methodology and Data

• Models used: Random Forest (RF) models, HadGEM3-GC31-LL global climate model (CMIP6), Quantile Delta Mapping (QDM) for bias correction, and a three-stage inverse distance weighting (IDW), Delta, and Variance rescaling approach for harmonization.
• Data sources: Monthly observations from 23 rainfall observation stations, ERA5 reanalysis predictors, and coarse-resolution general circulation model (GCM) fields.

Main Results

• Random Forest models demonstrated strong predictive skill, with test Nash–Sutcliffe Efficiency (NSE) values ranging from 0.45 to 0.81, RSR values from 0.43 to 0.74, and PBIAS values from −21.99% to +5.29%.
• Future projections indicate a basin-wide drying trend under both SSP2-4.5 and SSP5-8.5 scenarios.
• Under SSP2-4.5, mean annual precipitation is projected to decrease by approximately 12.2 mm (2025–2050), 19.6 mm (2051–2075), and 33.7 mm (2076–2099) relative to the baseline.
• Under SSP5-8.5, projected decreases are approximately 25.2 mm (2025–2050), 53.2 mm (2051–2075), and 86.9 mm (2076–2099).
• Late-century reductions in precipitation are projected to reach approximately 15–22% in several sub-basins.

Contributions

• Development and evaluation of a robust RF-based hybrid statistical downscaling framework combined with trend-preserving bias correction for generating station-scale precipitation projections.
• Provision of essential, high-resolution precipitation projections for water resources planning in semi-arid Mediterranean basins, highlighting a substantial decline in future water availability.

Funding

Not specified in the provided text.

Citation

@article{Ara2026Assessing,
  author = {Ara, Ismail and Yaşar, Mutlu and Gürarslan, Gürhan},
  title = {Assessing Climate Change Impacts on Future Precipitation Using Random Forest Statistical Downscaling of CMIP6 HadGEM3 Projections in the Büyük Menderes Basin},
  journal = {Water},
  year = {2026},
  doi = {10.3390/w18020277},
  url = {https://doi.org/10.3390/w18020277}
}