Evaluation of artificial intelligence models for daily prediction of reference evapotranspiration using temperature, rainfall and relative humidity in a warm sub-humid environment

Accurate estimation of reference evapotranspiration is essential for agricultural management and water resources engineering applications. In the present study, ability precision three artificial intelligence (AI) models (i.e., Support Vector Machines (SVMs), Adaptive Neuro-Fuzzy Inference System (ANFIS) Categorical Boosting (CatBoost)) were assessed estimating daily (ET0) using limited weather data from five locations in a warm sub-humid climate Mexico. The Penman–Monteith FAO-56 equation was used as target ET0 values. Three different input combinations investigated, namely: temperature-based (minimum maximum air temperature), rainfall-based temperature, temperature rainfall), relative humidity-based humidity). Extraterrestrial radiation values all combinations. AI compared with conventional Hargreaves–Samani (HS) model commonly to estimate when only records are available. goodness fit terms coefficient determination (R2), Nash–Sutcliffe efficiency (NSE), root mean square error (RMSE) absolute (MAE). results showed that among evaluated, SVM outperformed ANFIS CatBoost modeling ET0. Further, influence humidity rainfall on performance investigated. analysis indicated significantly improved accuracy models. Finally, better response over HS method. can be an adequate alternative modeling.