Applicability of Hargreaves-Samani and Valiantzas’ Models for Solar Radiation Estimation in Coimbatore using Temperature and Humidity Inputs

Precise estimation of reference evapotranspiration (ET_₀) is crucial for effective irrigation planning and water resource management, particularly in semi-arid regions such as Coimbatore, Tamil Nadu. The FAO56 Penman–Monteith (FAO56-PM) method is widely regarded as the standard approach for ET_₀ estimation. Still, its application is limited by the requirement of multiple meteorological inputs, including solar radiation (R_s) and wind speed, which are often unavailable or unreliable in many regions. This study evaluates two simplified models for estimating R_s, the Hargreaves–Samani (HS) model and the modified version by Valiantzas (2017), which incorporates relative humidity (RH) along with temperature, and examines their impact on FAO56-PM ET_₀ estimation using 21 years (2004–2024) of NASA POWER meteorological data for Coimbatore. Results showed a strong correlation between measured R_s and estimated R_s values, with the HS model achieving an R² of 0.7866 and the Valiantzas’ model an R₂ of 0.7034. When used to compute ET_₀, both models showed excellent agreement with FAO56-PM ET_₀ derived from measured R_s, with R₂ values of 0.9477 (HS) and 0.9281 (Valiantzas), and low standard error estimates (SEE < 0.22 mm day^-1). Seasonal analysis revealed that the models accurately preserved the expected ET_₀ patterns, with peak values in summer and minima in winter, aligning with trends in solar radiation. The findings confirm that both HS and Valiantzas’ models provide reliable alternatives for R_s estimation and ET_₀ computation in data-scarce environments. The HS model, due to its simplicity and robustness under clearsky conditions, is particularly well-suited for semi-arid climates. In contrast, the Valiantzas’ model offers added flexibility in regions with variable humidity. These results highlight the potential of simplified radiation models to enhance irrigation planning when comprehensive weather data are unavailable.