Damor et al. (2025) Data-Driven Modeling of FAO-56 Penman–Monteith Reference Evapotranspiration Using Limited Meteorological Parameters through Artificial Neural Networks

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

Identification

• Journal: International Journal of Environment and Climate Change
• Year: 2025
• Date: 2025-11-19
• Authors: P. A. Damor, G. V. Prajapati, P. A. Pandya, H. D. Rank, Hiren Parmar, T. D. Mehta, Divya Patel, Devrajsinh I. Thakor
• DOI: 10.9734/ijecc/2025/v15i115131

Research Groups

Not explicitly stated, but likely an agricultural research institution or university in Junagadh, Gujarat, India.

Short Summary

This study developed and evaluated Artificial Neural Network (ANN) models to accurately estimate daily reference evapotranspiration (ET0) using limited meteorological inputs, demonstrating that 3-4 input variables provide optimal accuracy and efficiency for data-scarce regions.

Objective

• To develop and evaluate Artificial Neural Network (ANN) models for simulating daily FAO-56 Penman-Monteith ET0 using limited meteorological inputs at Junagadh station, Gujarat, India.

Study Configuration

• Spatial Scale: Point scale (Junagadh station, Gujarat, India).
• Temporal Scale: Daily.

Methodology and Data

• Models used: Artificial Neural Network (ANN) models, FAO-56 Penman-Monteith (as a reference), Gamma Test (for input selection).
• Data sources: Local meteorological station observations (maximum temperature, wind speed, solar radiation, relative humidity).

Main Results

• The Gamma Test identified maximum temperature (Tmax), wind speed (WS), solar radiation (SR), and relative humidity (RHmean) as the most influential predictors for ET0.
• The ANN model with three inputs (Tmax, WS, SR) achieved a Root Mean Square Error (RMSE) of 0.4722 mm/day, a coefficient of determination (R²) of 0.9463, a Nash-Sutcliffe Efficiency (NSE) of 0.9029, and a Mean Absolute Percentage Error (MAPE) of 9.70%.
• The ANN model with four inputs (Tmax, RHmean, WS, SR) yielded an RMSE of 0.4504 mm/day, an R² of 0.9652, an NSE of 0.9116, and a MAPE of 8.56%.
• Models with more than four inputs showed only marginal improvement, confirming that three or four parameter combinations offer optimal accuracy with computational efficiency.
• ANN models reliably replicate the nonlinear dynamics of ET0.

Contributions

• Demonstrates the reliability and efficiency of ANN models as a viable alternative to the FAO-56 Penman-Monteith method for ET0 estimation in data-scarce regions.
• Identifies optimal minimal input combinations (3-4 variables) for accurate daily ET0 prediction using ANNs.
• Provides a robust decision-support tool for irrigation scheduling and water resource management, particularly in arid and semi-arid regions with limited climatic data.

Funding

Not explicitly stated in the provided text.

Citation

@article{Damor2025DataDriven,
  author = {Damor, P. A. and Prajapati, G. V. and Pandya, P. A. and Rank, H. D. and Parmar, Hiren and Mehta, T. D. and Patel, Divya and Thakor, Devrajsinh I.},
  title = {Data-Driven Modeling of FAO-56 Penman–Monteith Reference Evapotranspiration Using Limited Meteorological Parameters through Artificial Neural Networks},
  journal = {International Journal of Environment and Climate Change},
  year = {2025},
  doi = {10.9734/ijecc/2025/v15i115131},
  url = {https://doi.org/10.9734/ijecc/2025/v15i115131}
}