Tunca et al. (2026) Optimizing Reference Evapotranspiration Estimation in Data-Scarce Regions Using ERA5 Reanalysis and Machine Learning

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

Identification

• Journal: Agronomy
• Year: 2026
• Date: 2026-01-21
• Authors: Emre Tunca, Václav Novák, Petr Šařec, Eyüp Selim KÖKSAL
• DOI: 10.3390/agronomy16020253

Research Groups

Not explicitly mentioned in the provided text.

Short Summary

This study optimizes reference evapotranspiration (ETo) estimation in data-scarce regions by integrating ERA5-Land reanalysis data with machine learning models. It found that Extreme Gradient Boosting (XGBoost), when applied to bias-corrected ERA5-Land data, provides superior and computationally efficient ETo estimates.

Objective

• To optimize the estimation of reference evapotranspiration (ETo) in data-scarce regions by integrating ERA5-Land reanalysis data with machine learning (ML) models.

Study Configuration

• Spatial Scale: 33 stations across Turkey’s diverse climate zones.
• Temporal Scale: Daily meteorological data from 1981 to 2010.

Methodology and Data

• Models used: Random Forest (RF), Extreme Gradient Boosting (XGBoost), Extreme Learning Machine (ELM).
• Data sources: ERA5-Land reanalysis data, daily ground-based meteorological observations from 33 stations in Turkey.

Main Results

• ERA5-Land reanalysis data provides highly accurate solar radiation (Rs) and temperature (T) data.
• ERA5-Land variables such as wind speed (U2) and relative humidity (RH) exhibit systematic biases compared to ground observations.
• Among the tested machine learning models, XGBoost demonstrated superior performance (R² = 0.95, RMSE = 0.43 mm day⁻¹, and MAE = 0.30 mm day⁻¹) and computational efficiency for ETo estimation.

Contributions

• Provides a robust, regionally calibrated framework that corrects reanalysis biases using machine learning.
• Offers a reliable alternative for ETo estimation in areas where local measurements are insufficient for sustainable water management.

Funding

Not explicitly mentioned in the provided text.

Citation

@article{Tunca2026Optimizing,
  author = {Tunca, Emre and Novák, Václav and Šařec, Petr and KÖKSAL, Eyüp Selim},
  title = {Optimizing Reference Evapotranspiration Estimation in Data-Scarce Regions Using ERA5 Reanalysis and Machine Learning},
  journal = {Agronomy},
  year = {2026},
  doi = {10.3390/agronomy16020253},
  url = {https://doi.org/10.3390/agronomy16020253}
}