Ganjei et al. (2025) Evaluating climate change impacts on reference evapotranspiration using CMIP6 projections and machine learning in the Aras River Basin

Identification

• Journal: Modeling Earth Systems and Environment
• Year: 2025
• Date: 2025-10-18
• Authors: Simin Ganjei, Jalal Shiri, Sepideh Karimi
• DOI: 10.1007/s40808-025-02651-1

Research Groups

• Department of Water Engineering, Faculty of Agriculture, University of Tabriz, Tabriz, Iran
• Water Engineering and Science Research Institute (WESRI), University of Tabriz, Tabriz, Iran

Short Summary

This study evaluated future spatio-temporal trends of reference evapotranspiration (ET₀) in the Aras River Basin, Iran, using CMIP6 projections and machine learning, finding a consistent upward trend in ET₀, especially under the high-emission SSP5-8.5 scenario.

Objective

• To evaluate the future spatio-temporal trends of reference evapotranspiration (ET₀) in the Aras River Basin, Iran, under CMIP6 climate change scenarios.

Study Configuration

• Spatial Scale: Aras River Basin, Iran.
• Temporal Scale: Historical (1985–2022) and Future projections for three periods: 2021–2040, 2041–2060, and 2061–2080.

Methodology and Data

• Models used:
  • Statistical downscaling: LARS-WG 8
  • Global Climate Model (GCM): MPI-ESM1-2-LR (from CMIP6)
  • Machine Learning models for ET₀ prediction: Multiple Linear Regression (MLR), Random Forest (RF), Artificial Neural Network (ANN). (ANN was selected for projections due to superior performance).
• Data sources:
  • Daily meteorological data (precipitation, maximum temperature, minimum temperature) from 1985 to 2022.
  • CMIP6 climate change scenarios: Shared Socioeconomic Pathways (SSP1-2.6 and SSP5-8.5).

Main Results

• The Artificial Neural Network (ANN) model outperformed Multiple Linear Regression (MLR) and Random Forest (RF) in estimating ET₀.
• Future projections indicate a consistent upward trend in ET₀ across all stations in the Aras River Basin.
• The increase in ET₀ is substantially greater under the high-emission SSP5-8.5 scenario compared to SSP1-2.6.
• The most pronounced increase in ET₀ is projected at the Jolfa station, which currently exhibits the highest ET₀.
• Annual ET₀ at Jolfa station is projected to increase by up to 358.83 mm for the 2061–2080 period under SSP5-8.5.
• Seasonal analysis shows the most substantial increases in ET₀ are projected to occur in summer, intensifying water demand during the peak irrigation season.
• Spatial assessments reveal that low-elevation, warmer stations are likely to experience the most pronounced changes in ET₀.

Contributions

• Provides a comprehensive evaluation of future spatio-temporal ET₀ trends in the Aras River Basin using CMIP6 projections and advanced machine learning.
• Identifies the superior performance of ANN for ET₀ estimation in the region, offering a robust modeling approach.
• Quantifies projected increases in ET₀ under different SSPs, highlighting critical periods (summer) and locations (low-elevation, warmer stations) for water resource management.
• Offers valuable insights for hydrological modeling and water resource management strategies in the Aras River Basin under climate change.

Funding

• No external funding reported.

Citation

@article{Ganjei2025Evaluating,
  author = {Ganjei, Simin and Shiri, Jalal and Karimi, Sepideh},
  title = {Evaluating climate change impacts on reference evapotranspiration using CMIP6 projections and machine learning in the Aras River Basin},
  journal = {Modeling Earth Systems and Environment},
  year = {2025},
  doi = {10.1007/s40808-025-02651-1},
  url = {https://doi.org/10.1007/s40808-025-02651-1}
}