International Journal of
eISSN: 2576-4454
Submit manuscript...
Research Article Volume 9 Issue 2
A review on developing machine learning techniques for estimation of reference evapotranspiration using meteorological data
Bhavya TR, Ananta Vashisth, P Krishnan, Monika Kundu, Shivprasad
Received: July 29, 2025 | Published: August 25, 2025
Citation: Bhavya TR, Vashisth A, Krishnan P, et al. A review on developing machine learning techniques for estimation of reference evapotranspiration using meteorological data. Int J Hydro. 2025;9(2):85-93. DOI: 10.15406/ijh.2025.09.00407
Abstract
Evapotranspiration (ET0) is a combined loss of water from surface of the earth and vegetative surface to atmosphere under the influence of weather parameters. The weather parameters that are influencing ET0 are air temperature, solar radiation, relative humidity and wind speed. Evapotranspiration (ET0) is the major loss of water received by the crop through rainfall and irrigation and also it represents major portion of total water budget of crop. The estimation of accurate value of ET0 helps in irrigation scheduling at different critical stages of crop for maximum crop production and also for hydrological and water resource planning based on available weather data. This review discusses various methods for estimating ET, including traditional empirical approaches and machine learning techniques. Traditional methods, which are computationally simple and data demanding, may fail to accurately estimate ET. Machine learning methods, SVM, ANN and RF increase the precision of ET estimation by extracting features from all weather particularly in datarich regions. However, model complexity can lead to difficulties in interpretation the performance of these methods may be limited in data-scarce areas. Collectively, future research should aim to improve data quality, optimize model generalizability, and explore methods that integrate physical processes with data-driven models. When selecting an ET estimation method, considerations should include data availability, model adaptability, estimation accuracy requirements, and technical operational complexity to meet the needs of specific research areas and applications.
Keywords: evapotranspiration, penman-monteith (FAO-56), machine learning techniques
©2025 Bhavya, et al. This is an open access article distributed under the terms of the, which permits unrestricted use, distribution, and build upon your work non-commercially.